You are currently not logged into ManagerLeague
If you wish to log in, click here.
If you wish to sign up and join us, click here.
Wonderstar Analytics
The Transfer Market (06/06/2015 09:51)

 Hello everyone!

 

I've not had time to write blog posts for a good while, but I thought I'd quickly share some numbers on what the transfer market is up to these days - as many of you have noticed, player value is no longer an accurate gauge for how much a player will sell for, especially for higher quality, older players. Here's a few plots illustrating what the situation is:

 

These show how much players are being sold for across different ages (x-axis) and quality ranges (line darkness). The outer 4 plots split these into the 4 different positions, and the large centre plot shows this averaged across all four positions.

Most lower-Q players sell for around value, and younger players of all Q rarely go for >150% of value (though bear in mind these prices don't include credits that may have sweetened the deal). But it's really clear that players in their early 30s - up to mid 30s for keepers - are considered far more valuable by managers than their ML price suggests. These players can sell for 2-3x the list price (even 4x for elite goalkeepers), especially those in the top class Q bracket of 94+.

One of the reasons is likely the extended retirement age for outfield players, making those in their 30s more valuable than before. Another factor is that there are fewer top players in this age bracket, since this generation arrived with ball control set to 50, and without the benefit of the new elite youth academy boost so they're on average 3-4Q behind the latest generation. The rarity of experienced, high Q performers pushes the price up a bit.

But this only goes so far as an explanation. 

Why is there such a huge price inflation for players in general, and high Q players in particular?

 


 

Well, the main reason is that the game allows transfer fees to be determined by the market (with a few restrictions), but fixes player wages very low. So richer teams can stockpile very high quality players without incurring much of a wage bill. In other words, with a large cash reserve, you end up spending very little of that maintaining your high Q squad, and have lots then to spend on players to add to it. This is what's driving the prices up, and causing the richer/rebuild teams to hugely outspend the second-tier ones. As a result, you end up with all the expensive players congregating in a few very high Q squads, and it's impossible to compete without investing several billion dollars yourself.

This is different to real life. For most football clubs, player wages are by far their biggest expenditure, not transfer fees, and are the main limiting factor preventing a rich club buying up every great player. As an example, Manchester City paid an average of £12.34m in transfer fees for each first team player over the past few seasons. Meanwhile, the average annual wage they're paying those players is £5.33m. So in other words, a player's transfer fee in real life is, on average, equivalent to something like 2-2.5 years of wages for a large, rich club.

So what is this ratio in Manager League? Well, using ml-tools to get Q, age, and trasfer fee details for the last 2 seasons - and knowing the (fixed) salary costs for a player of a given Q in ML - we can calculate the same ratio. It's 32.8 - that is, a player's transfer cost in ML is, on average, worth more than 30 seasons of their wages.  

 

In Manager League, wages are 14 times cheaper than in real life ... or transfer fees are 14 times more expensive.

 

The effect of this of course is that the rich teams don't spend very much of their budget on wages. This means that they can afford to stockpile a lot of very high Q players - and spend all their savings bringing those players in, inflating the prices.

The knock-on effect is that to reach the upper echelons of the game, managers need to stockpile a similarly huge pile of cash, to get access to those elite players. Generally, this means spending many seasons developing and selling players, until they have enough money to buy their own competitive team. And it also means very few managers can be competitive at a time, since a single team can easily afford the wages of many elite players (leaving fewer available for other teams).

 


 

How do you fix this situation? Well, a complex approach would be to model player contracts in the game. If instead of treating players as commodities, they had a choice over where to move, you would allow the market to determine wages in the same way as it determines transfer fees and the two should rebalance.

A simpler route would just be to increase wages. There are two (related) reasons why this should help:

a) It should soak up some of the excess cash that's driving transfer price inflation; redirecting it out of the player pool. Teams with many high Q players would have less spending power, but lower level teams should have more, thanks to decreasing player prices.

b) It should prevent one team from being able to afford to stockpile 15-20 superstars, since the wages would be too much. This makes more of the elite players available to other sides, meaning fewer Q96-98 teams, and more competitive Q92-95 ones. Teams spending big would still have the edge, naturally, but it might no longer be necessary to spend 10 seasons rebuilding just to compete in the top division for a few seasons.

 

 

Here's a rough suggestion. In yellow, current ML salary costs (for a season, assuming 20 players) for different qualities. Even the very top teams currently spend only $100-150m a season on wages, i.e. around half the transfer cost of a top player. Under the suggested wage structure this would increase up to four-fold for a Q96 side, but very little for a Q80 team, making it more of a challenge to retain large stockpiles of top players all at the same time. This should lead to lower transfer prices, and more balanced sides with a mix of superstars and lower Q squad players - rather than 15 identikit cardboard cutout superstars all at the same Q.

 

Anyway, just something to ponder while you're being outbid for that Q93 36-year-old midfielder...

 

 

 

Read the full entry
Wonderstar Awards S116: League Of The Season (20/08/2014 06:21)

 

Here are the S116 league rankings - Seasons S112, S113, S114 and S115 are also available. Leagues whose sides performed well in the Champions League and Super Cup in S116, and/or had a lot of competition for their title or international spots, are ranked highly, regardless of how strong their average team Q makes them in theory (a Q-based measure of strength is provided for comparison).

 

For a rolling, weighted 5-season average of these scores check out the ManagerLeague League of Leagues! This is the first season we have had a full 5 seasons worth of data to include.


 

Season 116 Results

A quick recap of the methodology: Leagues receive ratings for Strength (based on player Q and teamstats), International Performance (based on their results, progress, goal difference and opposition strength in Champions League and Super Cup), and Competitiveness (based on how even the league is - leagues dominated by one or two teams will be lower ranked than one in which every team is very strong). Each rating ranges from 0 to 100, and a final league rating is made up of 2 x International Performance, plus 1 x Competitiveness, for a total range of 0-300. Strength is included in the table for comparison, and does not contribute to the overall rating - though of course stronger leagues would be expected to be both more competitive and perform better internationally. So which leagues were the most impressive in Season 116?

 

 

The Romanian league is strongest in S116, a spectacularly different story compared to last season's disappointing results. It's an unpredictable league, both in terms of the teams who win the title, and the performance of their sides internationally. But in S116 they lived up to their high strength, with a deep roster of strong teams in the league and all 5 international sides either qualifying from their groups or narrowly missing out. Norway also lived up to their strength - while the intense competitiveness of the top division has subsided a little, their sides put in very strong performances in the Champions League, where Blest reached the quarter-finals, and the Super Cup, with Krokeide only losing out in the final to High Society, who helped the USA to yet another 90+ rating. Only the lack of depth past the top five is holding back the US league at this point, their skill in the Champions League and Super Cup is well established now.

All three of those leagues were eclipsed in the Internation arena however by a resurgent Belgium. Darkness Endures may not have endured over the last ten seasons or so, but their return to the upper echelons of the league coincides with stiff competition from Bruges Blues and DansendHert. All three sides, along with alidas, reached the knockout stages in S116 and have propelledBelgium to 3rd place overall this season.

Other big success stories in S116 include Ireland, Australia, Denmark and Portugal with each posting big rises, while Netherlands, Indonesia and Hungary are headed in the opposite direction. Hungary's drop in ranking is particularly notable given that they finished 2nd as recently as S113. Spain's drop is even more precipitous, as they slipped 18 places to 32nd.

 

 

- Belizio

 

 

 

 

Click Here to return to the Wonderstar Analytics home page

 

Read the full entry
Wonderstar Awards S116: Manager Of The Season (13/08/2014 02:02)
As another new season begins, teams across ManagerLeague nervously cradle their trophies from the previous campaign, wondering whether their acievements will be enough to earn them the prize they really crave...



The S115 Wonderstar Manager Of The Season

 

For details about how these are calculated, and to properly prepare your enraged messages to me demanding to know why you were overlooked, see Season 112's winners. Note that for the first season, End of Season Cup performances give you a (small) boost to your rating, if you reach the QF or further in any of the top 3 cups (Q94+). And also check out the ManagerLeague Manager League, updated every season (in theory), which lists the top 100 managers in the game, according to total Manager Ratings since S112. So without further ado, the S116 Wonderstar Manager of the Season is...

 


 

It's the third season in a row that a Romanian manager has taken the award - testament to a run of unpredictable champions in a very strong league. In this case, recent CL champions Steaua Bucuresti were hardly underdogs, but dominated both the Romanian League and their SuperCup group to an impressive degree. Congratulations! Here are the rest of the top ten for S116:
 

 


Congratulations to everybody listed. Several big names occupied the top spots in season 116, with Champions League runners-up Beachlife, Norwegian league winners Goals Galore and veterans Blest all represented. Meanwhile, FC Brescia scored a notable treble with wins in the Italian league, League Cup, and Q98 EoS cup, as well as a last-16 appearance in the Champions League. 
 



 

Speaking of International tournaments, High Society brushed off the disappointment of last season's SuperCup final defeat and roared to the title, and in a nice piece of symmetry they followed up last season's International Manager of the Season award with a runner-up spot this time around. Beachlife take the overall title and their conquerors VN WinMySelf rank 3rd. Meanwhile Australia posted two semi-finalists (Champion FC finishing 6th in the table with a score of 94) and Bruges Blues made the last four of the Champions League - just one season after reaching the same stage of the Super Cup. That performance was the pick of an excellent season for Belgian sides in the international tournaments, with all four making the knockout stages. 
 

And finally, the top two managers in every league. Congratulations to everyone listed!

 

 

 


  

 

  

Click Here to return to the Wonderstar Analytics home page 
Read the full entry
Wonderstar Awards S115: League Of The Season (15/07/2014 04:13)


It's time for the fourth of our seasonal rankings of the 33 leagues - Seasons S112, S113 and S114 are found behind those links. Leagues whose sides performed well in the Champions League and Super Cup in S115, and/or had a lot of competition for their title or international spots, are ranked highly, regardless of how strong their average team Q makes them in theory (a Q-based measure of strength is provided for comparison).

 

For a rolling, weighted 5-season average of these scores check out the ManagerLeague League of Leagues!

 

A quick recap of the methodology: Leagues receive ratings for Strength (based on player Q and teamstats), International Performance (based on their results, progress, goal difference and opposition strength in Champions League and Super Cup), and Competitiveness (based on how even the league is - leagues dominated by one or two teams will be lower ranked than one in which every team is very strong). Each rating ranges from 0 to 100, and a final league rating is made up of 2 x International Performance, plus 1 x Competitiveness. Strength is included in the table for comparison, and does not contribute to the overall rating - though of course stronger leagues would be expected to be more competitive and perform better internationally. So which leagues were the most impressive in Season 115?

After a quietly strong under-the-radar performance in Season 114, the USA emphatically broke their glass ceiling, providing 3 of the season's 4 finalists. After two borderline perfect international seasons - 25 group stage wins last time round were supplemented by 23 more this season, as well as 9 wins and one defeat in the knockouts - the US league is one of only two to have increased its score each of the last three seasons. And with the best international score two seasons running, the only thing holding the league back now is its relative lack of strength in depth: The top teams have excelled, but there's a question mark over whether the sides behind them are ready to make the step up as and when those top teams fade away or go through rebuilds. For now though, this is certainly a period of American dominance, and ominously, that competitiveness is improving year-on-year too.  

England are the other league to have posted three consecutive rating increases. While the US dominated the finals, English teams actually did similarly to the USA's previous season, racking up slightly more points as they produced 4 group stage qualifiers (and one third place). In fact, only eventual champions Vagabonds managed to knock out any English teams in the Champions League. On the back of both this and an envious strength-in-depth, England top the league table for the second season running. Norway crept higher with a similarly strong international performance, and Greece continued to overperform their modest strength levels thanks to good performances by Death House FC and Aristotelis Forest.

Belgium shot up the rankings, the International league are starting to look more like a 5-qualifier league, and Serbia bounced back following a disappointing S114. In an opposite reversal-of-fortune, Portugal and Germany fell back after their strong S114 performances. Most dramatically perhaps, Romania collapsed from 2nd to 16th. The table gives each league in order (including the change in ranking since S113), with columns showing the number of international places assigned to each league, their strength, international performance and competitiveness. Ranks are in brackets:

 

 - Belizio

 

 

 

 

Click Here to return to the Wonderstar Analytics home page

 

Read the full entry
Wonderstar Awards S114: League Of The Season (15/07/2014 03:24)

 

Following on from S112 and S113, here is a ranking of the 33 leagues currently represented in the game by their S114 international performances and strength-in-depth. Leagues whose sides performed well in the Champions League and Super Cup, and/or had a lot of competition for their title or international spots, are ranked highly, regardless of how strong their average team Q makes them in theory (a Q-based measure of strength is provided for comparison). 

 

For a rolling, weighted 5-season average of these scores check out the ManagerLeague League of Leagues!

 

A quick recap of the methodology: Leagues receive ratings for Strength (based on player Q and teamstats), International Performance, based on their results, progress, goal difference and opposition strength in Champions League and Super Cup, and Competitiveness, based on how even the league is (leagues dominated by one or two teams will be lower ranked than one in which every team is very strong). Each rating ranges from 0 to 100, and a final league rating is made up of 2 x International Performance, plus 1 x Competitiveness. Strength is included in the table for comparison, and does not contribute to the overall rating - though of course stronger leagues would be expected to be more competitive and perform better internationally. So which leagues were the most impressive in Season 114?

In season 114, several of the traditionally strong leagues justified their 5 Champions League spots and rose to the top. Chief amongst them are the English and Romanian leagues, with nearly identical scores. England and Romania were the two most competitive leagues, knocking Norway down to 3rd. Meanwhile, the USA performed best overall in the Champions League and SuperCup. While this might seem surprising at first - no US team reached the semi-finals in either contest - in fact every single US side were dangerous, with an average score of 16pts and +12 goal difference in the group stages, and three quarter-finalists. In total during the group stages American teams won 25 matches between them, only drawing 5 and losing 5. Greece also posted very strong international scores again, with an unexpected SuperCup semi-finalist in Aristotelis Forest and a Champions League quarter-finalist in Panathinaikos FC Mouries.  

Indonesia, Germany, Spain, Portugal, Argentina and Canada made big jumps upwards, while Denmark, Bulgaria and Serbia showed the biggest declines. The Baltic League scored an impressive 114pts, mainly thanks to FC NVilnia's shock SuperCup runThe table gives each league in order (including the change in ranking since S113), with columns showing the number of international places assigned to each league, their strength, international performance and competitiveness. Ranks are in brackets:

 

 

 

- Belizio

 

 

 

 

Click Here to return to the Wonderstar Analytics home page

 

Read the full entry
Wonderstar Awards S115: Manager Of The Season (11/07/2014 19:04)

It's time to get all dressed up for yet another awards ceremony! That's right, it's...




The S115 Wonderstar Manager Of The Season

 

For details about how these are calculated, and to properly prepare your enraged messages to me demanding to know why you were overlooked, see Season 112's winners. And also check out the ManagerLeague Manager League, updated every season (in theory), which lists the top 100 managers in the game, according to total Manager Ratings since S112. So without further ado, the S115 Wonderstar Manager of the Season is...

 


 

The second successive Romanian Manager of the Season thoroughly earned the award, taking the 12th strongest squad - narrowly avoiding relegation with a 12th placed finish the previous season - to an astonishing league title! It's -iBonciu-'s first national level trophy, and will be hard to top for sheer achievement. Congratulations! Elsewhere in Romania, Titanul took the 13th strongest squad in Romania and dragged them into the Champions League with a third placed finish, almost as impressive an achievement (and not the first time Titanul has posted an excellent Manager Rating). Here are the rest of the top ten for S115:

 



 

Congratulations to everybody listed. If S114 was a season for the underdog, S115 included several bigger sides managing to combine strong domestic and international campaigns. We also have two former Managers Of The Season back in the top ten, with Whiting and 1milanfan both posting exceptional ratings of 95+. Elsewhere, Crvena Zvezda made their first top ten finish - but their fourth successive Manager Rating of 80+, an exceptional achievement.



 

Internationally in S115 the US league dominated, and this is reflected in the International Manager Of The Season rankings. While Vagabonds carried off the ultimate prize of the Champions League trophy, our top three managers are the other three finalists, relative minnows by comparison (but who isn't next to Vagabonds?). See Above (a joke that does not work quite so well with my horizontal format) took the top spot after reaching a first ever international final, while their conquerors Loz Cruzados, and Brad G Strikers (winner of no fewer than three runners-up medals in S115), also scored exceptionally highly. The top five were rounded out by two widely acclaimed veteran managers making strong runs into the knockout stages. 
 

And finally, the top two managers in every league. Congratulations to everyone listed!

 

 

 


  

 

  

Click Here to return to the Wonderstar Analytics home page 
Read the full entry
Wonderstar Awards S114: Manager Of The Season (11/07/2014 05:45)

 A mere season-an-a-bit late, your patience has been rewarded Ladies & Gentlemen! It's finally time for...


 

The S114 Wonderstar Manager Of The Season

 

For details about how these are calculated, and to properly prepare your enraged messages to me demanding to know why you were overlooked, see Season 112's winners. And also check out the ManagerLeague Manager League, updated every season (in theory), which lists the top 100 managers in the game, according to total Manager Ratings since S112. So without further ado, the S114 Wonderstar Manager of the Season is...

 


 

Congratulations! Despite an ageing squad that had lost both value and quality every season since S109, Dora-Flavius summoned a herculean effort from his players and equalled his best ever finish, third in the Romanian league. For this achievement Dora-Flavius is crowned Manager of the Season and is eligible for a free team analysis and the respect of his peers! The rest of the top ten were as follows:

 



 

Congratulations to everybody listed. S114 was a season for the underdog - Munchen Carpet were the only side in the top ten who played internationally, and were also the only league champions (in fact they secured the double in the International league). The remaining teams with high manager ratings ranged from small clubs heroically avoiding automatic relegation in difficult leagues - lorde_team, Valminor and Berntree Unite - to unlikely SuperCup and Champions League qualifiers, such as Alex FC and Steaua 1971 Dragasani. 



 

Internationally, it was another minnow who took top honours: FC NVilnia of the Baltic League caused a major shock by winning their Super-Cup group, before their unlikely run was halted by eventual finalists FC RealManager. Calgary Kings flew the flag for Canada after River City's strong performance in S113, providing their Champions League opponents with a much tougher challenge than expected, while Giarra Djinns, Tôi Yêu SFVP and Aristotelis Forest all made impressive runs into the latter stages of the Champions League and Super Cup. 

 

And finally, the top two managers in every league. Congratulations to everyone listed!

 

 

 


  

 

  

Click Here to return to the Wonderstar Analytics home page 

 

Read the full entry
Gaining Part 6: The Gaining Formula (29/05/2014 07:17)

"There are so many factors involved in this "gaining-bit", each of them alone has almost no impact."

- Spinner


A few months ago when I started this blog, I spent the first few posts discussing gains. After five posts of varying detail I moved onto some other things, some of them controversial. And I've spent much of my time focusing on match strategy, which I feel is the most interesting part of the game. But the most common thing I still get asked when people pm me out of the blue? "How do I get my players to gain!"

Some context now, which you can skip over if you're just looking for the raw numbers (next section). But I've been thinking around the whole gaining strategy idea for a good while now, and I think I've reached quite a different conclusion to the majority of managers on here, so if you'll indulge me: I believe a lot of the 'rules' for gaining that you read on the forums and help blogs are either outright false, or greatly exaggerated. Chief among these is the idea that you need to have 8 very old players on the pitch alongside 3 youths for them to gain properly. Generally, the advice is to have an average lineup age of over 30.

I don't do this. In my friendlies I have a range of ages - rookies, young players, first teamers, a couple of veterans, and an average lineup age around 24 - for the simple reason that I'm building a whole squad, not a handful of players for sale, and I don't want my 23 and 27 year olds to start declining because I'm dedicating all my friendlies to a mob of 38 year old goalkeepers (and also because that method of gaining feels too unrealistic to me, and makes the game less enjoyable). I went into this strategy with open eyes, figuring that I probably wouldn't gain quite so well on my younger players but hey-ho. Instead, I've gained 3300 attributes over 6-and-a-half seasons, with 131 additional friendlies that I could have taken along the way if I were a little more organised. The short story is, ignoring the prevailing wisdom here doesn't seem to have done me too much damage.

The longer story is that it's not simply a case of "lots of attributes, but spread too thinly to be worthwhile" either. I compared myself this season to another team of similar quality to me, regularly in the top-10 gainers for the season, who were very obviously adopting the lots-of-very-old-goalkeepers strategy. They've been raising players over the past few seasons, and in that time they paid $675m for ten players, and sold them (or have them currently valued at) $980m. The rest of their squad are worth next to nothing (and have to be replenished at a cost of $25m or so every season due to retirements). The team's youth players were sacked to make way for the bought players, some of whom went for big money, while others fizzled and made a loss. So that's an average profit of just over $30m per player, after an average training period of just over a season each - total $27m/player/season. Training 6 or 7 players at a time that equates to a benchmark of around $150-200m per season.

Just my top 7 purchases over the past few seasons cost me $167m and are now nudging $600m. At an average of around 2 seasons development each, that works out at $26m profit per player, per season: Comparable, in other words, to the rate a relatively successful youth farm team is getting. But there's a big difference: I've not just got those 7 players. I've also been training a whole bunch of other players too, in place of the large numbers of (otherwise useless) veterans. So my 8 youths since I began the game weren't sacked, and after an average 4 seasons development each are now worth another $620m. Plus there's a further $500m or so still locked up in the rest of my squad. That extra billion-and-a-bit is extra value that the youth farm has sacrificed in the pursuit of an extra attribute or two from their chosen ones. There's a very strong argument I think that the strategy is not as effective as people assume, once you really look at the numbers. Despite all those gains, the 6-youth strategy earns:

 

- Roughly an extra $1million a year on each of those youths (and maybe some credits in a gaining contest?).

 

And it sacrifices:

 

- Roughly $1billion on the rest of their squad over 6 seasons.

- Any chance of being competitive in the league, or in player cups, for the whole period they're building up cash.

- The flexibility to minimise tax by buying/selling just the right value of players per season.

- ALL of their teamstats. Buying twelve 40-year-olds every season will do that.

 

Pick your poison I guess. For certain the 6-youth strategy will work for some people, in some circumstances, but don't just pursue it blindly is all I'm saying. The sacrifices you're making may not be worth it after all.


 

The Model

I know, I know! So far so anecdotal. I might have been lucky. The team I looked at might have been unlucky. Having said that, I don't think they consider themselves to be, and they're regularly in the top rankings for gains. I'm not naming them because I don't want to give the impression I'm spotlighting a 'failed' team; on the contrary they seem to have carried out the strategy very successfully. Which is entirely my point. 

So is this example just an outlier, or is there a general lesson we can draw? And if there is, why might the strategy not be as effective as people think? Time to look at the data. Previously I took a look at how player gains per match might be affected by surrounding your players with veterans. The conclusion, from examining 21 high-gaining teams with different strategies? It basically doesn't help at all, and the key thing is simply to play young players in as many games as possible. But that's based on just 21 teams. What if we expanded the analysis to 1500 teams? Perhaps then we could get a much clearer idea of what the most important factors are.

For this, I've got hold of every player's attribute gains this season - up to round 21 - for teams in either Div1, or Div2 of the top 16 leagues. After removing Bot teams, that means we're looking at over 36,000 players, and well over a quarter of a million attribute gains. Along with their gains, I've also gathered the publicly available info about these players' seasons: Their number of appearances (in league, CL, SC and friendly games), their starting attributes for the season, the team they're on etc. etc. The approach is going to be similar to our Player Attributes analysis from a while back, in that we have a dependent variable (Gains) and we're going to use multiple linear regression to establish what factors predict how many gains a player earns, and their relative importance. For those of you more statistically inclined, note that we'll be using standardized variables throughout, as the scales are very different for all the variables I'm going to use.

So, first off, what goes into the model? We're going to look at:

 

Dependent Variable (what we're trying to predict):

Number of Gains. Fairly self-explanatory, just the number of attributes earned by that player, by round 21.

 

Independent Variables (what we think might affect gains):

1. The age of the player. We'd expect younger players to gain better than older ones, so this should come out of the analysis as important.

2. The average age of the player's teammates next to him on the field. In other words, if he's a midfielder, the average age of the other midfielders on the pitch. If playing directly alongside a veteran helps specifically, we should see this improving gains.

3. The average age of the player's teammates in different positions. So for a midfielder, this looks at the average lineup age of defenders, attackers etc. If having a high average lineup age is important, this number should be high.

4. The player's Quality. Presumably, higher quality players will gain more slowly. If so, this should come out as important. We'll call it simply Quality.

5. Friendly Games. The number of friendlies played. Note that since Player Cups appearances are not recorded in player details, this variable probably accounts for them too (players who get loads of friendlies tend to also be played in custom cups a lot).

6. Competitive Games. These are league, champions league and super cup appearances. Similar to above, cup matches are not recorded in player details, so probably this variable also captures cup games (first team players are likely used in competitive games).

 

Dummy Variables (things I included to mop up various effects and improve the accuracy of the analysis, but aren't themselves interesting):

1. Division. This was included to account for the fact that division 2 players get more department cup and league cup games before round 21, neither of which are included above.

2. A constant, which accounts for the regular training and camps available to every player even if they've played no matches. In practice, because we're standardizing variables no explicit constant is required in the model - but it's effectively there.

 

Discarded Variables (things I looked at and found to be unrelated to gains, or that we just can't include):

1. Potential. I'm not scouting 36,000 players for you, I just don't have that kind of cash I'm afraid! Fortunately, potential should average out since it's unlikely to be strongly correlated with any of the variables we are including. But just remember that's another factor.

2. Teammate quality. I checked this actually, both the average quality of players in the same position and different positions. Neither affected gains at all so I won't be revisiting them.

3. Opposition quality. There's no way of telling how good opponents were in friendlies, but in any case, we've looked at this before and found little effect. And also, the average teammate quality is a pretty good indication of opponent quality too - but that had no effect, as mentioned above.

4. Cup and Player cup games. As noted above, these are captured instead by the friendly and competitive game counts, and also to a certain extent by division. Just remember when you see "Friendly Games" it probably represents all non-competitive matches, not just friendlies, and "Competitive Games" will probably also include cups.

 

I ran the regression analysis separately for each position. For goalkeepers, average age in the same position isn't included, since you don't have multiple goalkeepers on the pitch at the same time. Apart from that one variable, the others - Age, Average Lineup Age, Competitive Games, Friendly Games and Player Quality - were very similar across all four positions. That's a good sign, as it suggests the amount of noise is pretty low, and we're getting quite an accurate estimate for each variable. So I just average them together to produce a pie chart based on how important each factor was in determining gains. Here you go...


 

The Data

 

 

The Gaining Pie. 

 

OK, let's step through each slice. By far the most important factor is the number of matches, which account for well over half of the variation in attribute gains across players. Because there are far more of them a season, friendlies are the most important, but competitive matches help a huge deal too (and remember, the effect of cups and players cups is incorporated into those slices too). If you want to see gains, get your players on the pitch at every opportunity. This means there's a HUGE cost associated with playing 8 old players in each game - you're effectively wasting around 30% of your gaining potential on those players.

The next biggest factor is player age. Pretty unsurprising, but younger players gain better than older ones. The interesting thing perhaps is that this is only half as important as the number of games played - so veterans playing lots of friendlies can end up earning a lot of attributes, while a 23-year-old left warming the bench might earn fewer. It also means that it's still worth trying to improve your mid-aged players, e.g. 25-30, if they're important for your first team. Play them in plenty of games and they should still rack up some attributes. Let's pause for a moment. Age and number of games account for almost 90% of gains in this analysis. Just remember that if you take nothing else from today: A young player, given plenty of matches in a season, will gain a lot of attributes regardless of the team around them.

The next few factors might help a bit at the margins. Player quality reduces gaining, and the importance is still fairly high, around 9%. Pretty unsurprising, but if you're taking part in a gaining contest on the forums, more important than your gaining strategy is the player you begin with. You want to look for as low quality a player as possible - and preferably one with low alternative stats too, like low keeping for an outfielder. Obviously it's unlikely the faster gaining will help them overcome the poor starting point, but if your aim is to win bag loads of credits on the forums, that's perhaps not so important. In a follow-up post I'll be going into much more detail about player Q and its effect on gaining.

Finally, age. The factor hailed as by far the most important thing to improve your gains - and in whose name billions of dollars worth of useful gains are sacrificed - comes in at a mealy 3.3% importance. And of that 3.3%, by far the most important aspect is the age of players in the same position. In other words: Make sure your young strikers are playing alongside a veteran attacker, but don't worry so much about the overall lineup age. In actual fact, this 0.6% may well actually be 0%, since players (especially in friendlies) are often played out of position - so the 0.6% could come from a midfielder benefitting from defender age because he's playing in defence or an attacker benefitting from a keeper playing alongside him in attack. This all matches the conclusions from the previous lineup age analysis too. Basically, it seems to make no difference. And of course, as always with age, remember the useful variable could be experience instead - so be wary of using old, inexperienced players to get the same effect.

So what about potential, where would that fit in? It will remain an open question how useful stars are - it's impossible to get a big enough sample size with known star ratings. So here's an idea: If you're a Div 1 team or a Div 2 team in one of the 16 biggest leagues (ranked by Q) during S114 you could send me your player's star ratings along with their id, and if we get enough we could run an estimate. Anyone who sends me this data before the end of the season will get the result of the analysis sent to them once I've run it. Any interesting results from the analysis will not be posted publicly to the blog until at least one season later - so S116 at the earliest. My best guess is that potential would be somewhere around 5% - less than player Q (even low star players gain fast) but maybe a touch higher than teammate age. Would be interesting to finally put a firm number on it though!


 

The Advice

Long article, I know, so let's boil all this down into a few pieces of advice - my own alternative to the advice lists that are already out there.

 

1. This is the most important thing: Decide what your aims are. Are you trying to build a team? Maximise sales and build up cash? Win a gaining contest? The answer here will affect your whole approach so keep it in mind throughout the rest of the advice - and for any other advice you may read elsewhere.

2. The most important thing is matches. Now, obviously, the best thing to do is play 200 friendlies a season. But if you're not buying credits that's hard. Don't worry though: You'll get 100 open friendlies, and then just play as many sets as you have credits available (up to 50/season). Your 15 each season will probably be spent on training camps and stadium upgrades, but you may have some left over, especially if you win player awards in your department, get promoted, win a cup (including player cups), etc. etc. If you're trying to win a gaining contest there's no alternative to playing 200 friendlies. If you're just trying to get some decent gains and build a team up, don't worry about it so much. You can compensate by playing extra...

3. ...Cup matches. Both player cups and real cups are a great source of extra games. Just player cups account for 17+ games a season - if you can be competitive and push that close to 50, that extra 30 games or so is worth close to 50-60 extra friendlies. So I would never recommend blowing your teamstats by buying too many players at once - it just ruins your chances of competing and getting those extra games. The exception is if you're very credit-rich and can afford to host your own, uncompetitive cups. In that case you can get 48 games a season, though it will cost you 112 credits to do so. But otherwise, you'll want to focus your efforts on getting as far as possible in any cup contest you play in (prioritise it over the league!). Get your teamstats trained up, especially penalties. Often it will be worth paying a couple of credits to face teams of the same Q - you should sometimes win back your entry fee by winning the cup. And league/department cup wins are important too, as each one nets you an extra competitive game.

4. Keep a large, balanced squad. NOT A DOZEN GOALKEEPERS! If you want those extra cup games, you'll need to play people in position. And you'll need close to 30 players, so fitness doesn't kill you in later rounds. Finally, what I mean by 'balanced' here is a lot of players around the same Q, e.g. six strikers between Q80-Q85 for example, not two strikers of Q90 and 4 of Q72. This means if you're in a Q-limited cup, you have more capable players you can rotate in to help you win. No point beating your opponent in fitness if you're fielding Q72 in a Q90 cup. And remember - you don't have to sacrifice gains on your chosen few players here. So long as there's a veteran alongside them, the rest of the team can be adapted to maximise your chance of winning extra matches. And be sure to set up your events so that if the result is not in doubt after 60mins, you take the opportunity to bring off your best players and sub on your weakest, or the ones you want to gain.

5. Run a training camp every weekend, preferably extreme (but high is fine if you can't afford the extra $10m a year extreme costs). Stamina is great to focus on, but if you're selling, it's all about primary stats. Everyone suggests the opposite, but the Q of your player - and how much he gained the previous season - are much bigger selling points than the low level of a primary stat. A Q78, Sh79 striker who gained +6Q last season will fetch a higher price than a Q77, Sh77 striker gaining +5Q. The listed value ($89m v $63m) is a big anchor, and having two points lower shooting will not be valued at $25m, especially when that Q77 player gained less the previous season. Otherwise, get popular stats (e.g. perception) from 59/69/79 to 60/70/80. Your guys will turn up in searches more often that way (as well as just looking better when people scan them). 

6. Make sure there is 1 veteran in each outfield position (def, mid, att) at all times in all games. Then fill out the rest of your side with whoever you want to gain, whether that be first teamers, youths, or whatever. The 6 players you most want to gain should be on the wings, given things to do, and subbed on or off at half time, just like most people recommend. But you can still be gaining well on another 4 or 5 players, too.

7. If your aim is to maximise sales, think about how long you want to develop each player. A 17-year-old might not develop as quickly as you hoped in season 1, but have a great season at age 19 (in general players seem to gain faster or just as fast at 18-20 than at 17). If you want to maximise sales, you need to hang on to players when they're still developing, and sell them when their development is peaking. To have this flexibility, and still be able to sell 5 players a season to maximise your profit (4 purchases plus youth player) you can't only be developing 6 players. Developing 10 gives you the option of holding on to each one for an average of two seasons (which will vary depending on when you judge each player is going to maximise your income) and still leaves 4 slots in your lineup for veterans, one in each position. Finally, this also reduces risk. If a player is fizzling you can cut him loose for 70% (or even sack him) without torpedoing such a high proportion of the season's input. If I was looking to just raise money, this is the approach I'd take.

8. Train wherever possible. This is another reason to keep a big squad - you can't increase your friendlies, but you can decrease the number of times your players rest. If you're really trying to maximise gains, you should be taking individual training any time you're within Stamina/10 + 2 of 100% fitness, otherwise you're just throwing away fitness.

9. If (and ONLY if) your single aim is to raise one player above all others and win a gaining contest on the forums - it's going to all come down to stats and potential. Obviously you've found a 5* player, but is he low enough quality? Get the lowest total attributes you can (including keeping!), with the exception of stamina (and make sure that's 70/80, not 75/85). Train the lowest stat at camps, even if that's keeping. Only play friendlies when that player is available and put him alongside at least one veteran. Give him every task you can - all the free kicks, penalties and corners in the world. Play offside trap and long ball if his perception is low, play continental if he can gain passing instead. Forget goalkeepers - build a squad that can get you through every possible cup round, even while carrying this wretched creature. And good luck!

 

- Belizio

 

 

  

Click Here to return to the Wonderstar Analytics home page

 

 

Read the full entry
The ManagerLeague League of Leagues! (16/05/2014 22:50)

 

Which leagues are the strongest? I've developed a rating that relies on results - and not theoretical quality - to determine which leagues are the strongest in a given season. Here I will keep and update a 5-season, weighted rolling sum of these ratings, which should reflect the true strength of each league in international competition. This serves at least three main purposes:

1. It's interesting, and provides some national-level bragging rights for those who've contributed to a successful league!

2. It lets you see which leagues are improving over time, and which are stagnating or going backwards. This is of course only one possible measure - active players and team Q are also relevant here - but especially for small leagues trying to strengthen and develop (and remain represented in the game) this could be a useful way to track progress.

3. It provides an at-a-glance measure of which leagues are being given more international places than they justify with their performances, and which deserve to have their places increased. Although the developers do not currently appear to use international performances as a major basis for allocating these spots, they do frequently tweak and improve the game, and also make adjustments to CL/SC allocations over time. Having a consistent measure available for international performance may encourage this to be used more as a basis in the future, which I think would be fairer, and discourage 'tanking' in the Super Cup or Champions League. At the very least, it provides some hard evidence to point to if you decide to request extra spaces for your league in the future (Poland! Greece! Australia! Ireland! Singapore!). Currently 7 leagues receive five spots, 17 receive four, 7 receive three places and 2 receive just two spots.

Each season's overall score is given, and the breakdowns can be seen in the page for that season. The overall score is a weighted sum: 5 x the most recent season, plus 4 x the next most recent, and so on down to 1 x the 5th most recent season. This is the first table featuring a full 5 seasons.

Seasons included in the current averages: S112, S113, S114, S115S116. 

 

  

- Belizio

 

 

 

  

Click Here to return to the Wonderstar Analytics home page 

 

 

Read the full entry
Wonderstar Awards S113: League Of The Season (16/05/2014 22:10)

 

A couple of weeks ago I suggested a way of measuring league strength without using team Q, and posted an accompanying table, ranking the 33 leagues currently represented in the game by their S112 international performances and strength-in-depth. Leagues whose sides performed well in the Champions League and Super Cup, and/or had a lot of competition for their title or international spots, would be ranked high, regardless of how strong their average team Q made them in theory. After another season of international competition - congratulations Hotspurs FC and Panathinaikos FC Mouries - the updated table is featured below. I'll also from now on keep a track of scores over the previous 5 seasons. This way we can see which leagues outperform their inherent strength, which are improving over time, and which deserve more or fewer Champions League and Super Cup places.

 

For a rolling, weighted 5-season average of these scores check out the ManagerLeague League of Leagues!

 

A quick recap of the methodology: Leagues receive ratings for Strength (based on player Q and teamstats), International Performance, based on their results, progress, goal difference and opposition strength in Champions League and Super Cup, and Competitiveness, based on how even the league is (leagues dominated by one or two teams will be lower ranked than one in which every team is very strong). Each rating ranges from 0 to 100, and a final league rating is made up of two parts International Performance, and one part Competitiveness. Strength is included in the table for comparison, and does not contribute to the overall rating - though of course stronger leagues would be expected to be more competitive and perform better internationally. So which leagues were the most impressive in Season 113?

Well, it was a great year for Scandinavia, as Sweden improved their ranking to move into the top ten, Norway ranked first overall, and Denmark made the biggest improvement of any league, and also outperformed their strength more than anyone else. Hungary, Greece and Poland also showed very strong performances, while Brazil, Baltic LEague and France all overperformed their Strength considerably. Portugal underperformed, while Argentina and the Arab League slipped back as they failed to replicate strong performances from S112. In terms of international performance, Norway and USA ranked strongly after dominated the Champions League group stages, but both Champions League and Super Cup featured a Hungarian side in the semi-finals, and a Polish team in the final, pushing those two leagues to the top of the pile.

The table gives each league in order (including the change in ranking since S112), with columns showing the number of international places assigned to each league, their strength, international performance and competitiveness. Ranks are in brackets:

 

 

 

 

- Belizio

 

 

 

  

Click Here to return to the Wonderstar Analytics home page 

 

 

Read the full entry
Wonderstar Awards S113: Manager Of The Season (13/05/2014 07:50)

 

Dust off the tuxedo! Draft those acceptance speeches! Plan out your strategy for taking maximum advantage of the free bar! It's time for...

 

The S113 Wonderstar Manager Of The Season

 

For details about how these are calculated, and to properly prepare your enraged messages to me demanding to know why you were overlooked, see Season 112's winners. And also check out our new feature: The ManagerLeague Manager League! This will be updated each season, and lists the top 100 managersin the game, according to total Manager Ratings. So without further ado, the Manager of the Season is...

 

Whiting of Raccoon City Team, Italy

 

Congratulations! Whiting guided Raccoon City Team to 3rd place in Italy, added the Italian League cup, and stayed competitive in the Champions League - all with a small squad, the tenth highest teamstat total in the league, and an average lineup quality of around 91 by the end of each game. A fine example this season of excellent squad management, with 42 games squeezed from just 11 players over Q80 - proving young players can be brought through even as trophies are added. Whiting receives a complementary analysis from Wonderstar analytics, but as always the real award is the respect you guys will inevitably shower him with! And here is the rest of Season 113's top ten, the very best of the best last season:

 

 

Interestingly, the top three this season were all in the top 12 last season too (out of over 500 eligible teams!), and two sides from last season's top ten - Northcountry Timberwolves and Imparaveis - finished 14th and 15th this time around, scoring 93 each. These managers really are showing consistently exceptional results for their team strength, and all rank very high on the all-time ManagerLeague Manager League. Cork Hibernian leap from 12th in their breakthrough title-winning season to second this time around, after competing hard domestically, reaching the cup final, and very nearly managing a surprise qualification from their Champions League group. Harcipocok FC overperformed a moderate strength in both the Hungarian league and the Super Cup, while AKNEO followed last season's Champions League triumph with a run to the Super Cup final and a domestic league and cup double. Tunetul and Zmajevi Beograd overperformed in the Romanian and Serbian top divisions respectively to reach 7th, while Steaua Bucuresti reached second place and won the league cup (champions Dinamo Red also finished 12th in this season's Manager Table). Angels Share had an impressive season in Norway, reaching 4th, while Toreknall took the title. Finally, Avantu Tulle finished 2nd in France, despite a drastically lower Q than the teams around them. 

 

 

Internationally, AKNEO took the overall International Manager Of The Year Award for their excellent run to the Super Cup final - the side that beat them, Panathinaikos FC Mouries of Greece, are third on this table having faced a very slightly easier set of opponents. Between them, Hotspurs FC get recognition for fulfilling their dark horse prophecy and winning the Champions League, but the remaining two places in the top five belong to smaller clubs who proved stronger than expected in the group stages: River City FC of Canada, and Cork Hibernian of Ireland.

 

And finally, the top two managers in every league. Congratulations to everyone listed, and good luck in Season 114!

 

.

 

 

 

  

 

  

Click Here to return to the Wonderstar Analytics home page 

 


Read the full entry
Wonderstar Awards S113: Player of the Season (12/05/2014 00:06)

 

Welcome to Season 114! I hope you all had a successful or enjoyable season - or if not, that your team's incompetence and failure was at least somewhat entertaining, and you enjoyed sacking the worthless, money-grabbing layabouts. And of course a new season means time for a new set of Wonderstar Awards! This season we are expanding the awards. Look out for team/manager awards, and league awards, in separate posts soon.

 

But here we're introducing a new category: Players of the Season! Now I know what you're thinking - there are already awards, or at least rankings, for the top players in each league. So hopefully these will look a little different, and are calculated in slightly more sophisticated ways than just average performance. The awards are for division 1 players in any league, and include:

 

Goalkeeper Of The Season

Defender Of The Season

Defensive Midfielder Of The Season

Attacking Midfielder Of The Season

Attacker Of The Season

The Sad Keanu Award For The Biggest Superstar To Spend The Whole Season Sitting On The Bench On His Own Eating Sandwiches


For each of these awards I'll briefly explain the methodology, announce the winners for both Seasons 112 and 113, and then let the managers responsible for these players bask in reflected glory. All but the Sad Keanu Award also have youth equivalents, for player aged 23 or younger.


 

Goalkeeper Of The Season

There are four components I'll be using to calculate Goalkeeper of the Season: Average performance, Strength of opposition, Matches played and (for Season 113 onwards) Save percentage. The rationale for this? A goalkeeper has the most specialised job on the field, and can't really be held responsible for the number of chances opponents are given. Fortunately performance in goalkeepers tends to reflect the amount they had to do, so we'll just remove half of the effect of opponent quality/teamstats (to take into account that it's harder to save against elite strikers). The approach is as follows: Correct performance and - if relevant - save percentage for opposition strength; determine how far above average these metrics were, given the number of matches played; and combine the two. By these measures the top goalkeepers were:

 

Season 112 Goalkeeper Of The Year: Steinar Gjelstad (Ф UNFORGIVEN Ф)

Season 112 Young Goalkeeper Of The Year: Nace Narobe (FC Ruwenda)

Season 113 Goalkeeper Of The Year: Norbert Németh (Levski Bulgaria)

Season 113 Young Goalkeeper Of The Year: Lei Vranckenm (Bombero AC)

 

Steinar Gjelstad put in an average performance of 67 against the best Bulgaria could offer in Season 112 - alas it was not enough to save Unforgiven from relegation, but he's still going at the ripe old age of 38 and may yet see one last season in the top flight. Season 112 was not a great one for young goalkeepers, but Nace Narobe emerged from FC Ruwenda's youth academy in Season 107, made the position his own at the tender age of 19, and by Season 112 was posting average performances of almost 60 in the German top division. Bulgaria seems to be the place to discover goalkeeping talent, since Season 112's 3rd highest rated keeper, Norbert Németh, went two places better last season, posting an exceptional average rating of 72 and averaging more than 5 successful saves per game. He can add Wonderstar Goalkeeper Of The Year to his glittering haul of awards. Unfortunately, the shame of relegation was too much for poor Norbert, who has since stolen the identity of Northern Irish legend Pat Jennings and scarpered to the greener grass of Munchen Carpet. A snip at $35m. For those of you in need of a top keeper and kicking yourselves at missing the opportunity, I'll just note that Vito Palumbo of Eresson - second best keeper of last season, and still in his twenties - is currently on the market at a generous 30% discount. Top young keeper Lei Vranckenm has just broken into the first team at AC Bombero in time for their first ever Super Cup campaign, and may be harder to prise away from his club.


 

Defender Of The Season

I'm not interested in your fancy-dan, marauding, Roberto-Carlos-lite full-back who spends his whole time in the opposition half attempting stepovers and taking free kicks. No - that's a midfielder. A real defender spends his time in his own box making blocks, clearances, headers, getting blood and sweat and dirt all over him and surreptitiously raking his studs down the opposing striker's hamstring. And he has excellent hair. With all that in mind, Defenders are calculated in a similar way to goalkeepers, except that as well as removing some of the effect of opponent strength, we'll remove ALL of the effect of assists and goals for defenders. We're not impressed. This leaves us with the bit of performance that comes from actually defending - and by that metric the Defenders of the Season are:

 

Season 112 Defender Of The Year: Kompany (TMS FC)

Season 112 Young Defender Of The Year: Vegard Killi (Dragunia Soccer)

Season 113 Defender Of The Year: Asis Nasseri (Melissokomoi)

Season 113 Young Defender Of The Year: Smidts (DansendHert)

 

When Team BWO went into hibernation in Season 111, Kompany moved to Singapore and hasn't looked back, posting ratings of between 71 and 77 in each of the last three seasons, helping TMS FC to second place in the league last year. Norwegian Vegard Killi is currently flying under the radar at Brazilian second-division club Dragunia, but had an impressive season in the top flight in S112, despite a relatively low Q rating of just 86. Last season Asis Nasseri was one of the few bright lights for Melissokomoi as they struggled in the Greek league, and despite having his best years behind him at 35, Asis is looking forward to a new challenge in Bulgaria at newly promoted AFC Dzhebel UnitedA Champions League Semi-Finalist - twice - before he was 23, last season Smidts was the most impressive pillar in one of ManagerLeague's most fearsome defensive units (only Seventh Heaven allowed fewer chances than DansendHert in any top division last season). Expect him to shut down the world's best strikers for another decade.


 

Defensive Midfielder Of The Season

The top performers in any league are all the same type of player: Midfielders entrusted with the task of taking free-kicks and corners, who consequently rack up the assists and goals against weaker opponents. In fact, glance at a random team card and there will virtually always be one standout midfielder. Are they really the most important player on the team? Well sure, they might post sky-high performances, but if other players could do the same with set-piece duty, then they're not necessarily so special. And if you read a match report from time to time, you'll know there's another, less replaceable role - the all-action defensive midfielder, breaking up play, winning 50-50 balls and keeping your attacks flowing with accurate passes. Here we acknowledge those players by stripping away the effects of assists and goals from performance, to find which players have truly been running the show in a steady, unshowy way:

 

Season 112 Defensive Midfielder Of The Year: Marius Glad (SHARP United)

Season 112 Young Defensive Midfielder Of The Year: Aleksa Ašanin (Grien Bey Pakkers)

Season 113 Defensive Midfielder Of The Year: Mattéo Nguyen (Toreknall)

Season 113 Young Defensive Midfielder Of The Year: Adrien Leroy (millou_psg)

 

Marius Glad has been a model of consistency in the Swedish top division, and in Season 112 he posted a rating of 83 from just 4 goals and 4 assists. Meanwhile, Aleksa Ašanin was developing into the prototypical midfield destroyer for Grien Bey Pakkers, with 2 assists and 13 foul points in a suspension-shortened 23-game season. An average performance rating of 81 made him worth it. Toreknall emerged champions last season of what is probably the strongest and most competitive league in the world, and did so thanks to an exceptional defensive performance (only Goals Galore gave up fewer shots to their opponents in Norway). While Mattéo Nguyen spent the previous season racking up assists from set pieces, in S113 he played just as important a role protecting the defence and setting up chances for his teammates - posting a performance of 91 despite just ten assists and goals combined. And those of you looking for an elite central midfielder might take a look at Adrien Leroy of Millou_psg, who posted a performance rating above 90 from only 11 assists.


 

Attacking Midfielder Of The Season

Time for the showboaters! This list is probably going to look quite similar to most people's "greatest midfielders in the world" guesses. The only real difference is that opposition strength is taken into account, so flat-track bullies in weaker leagues are slightly downgraded compared to those who've proven themselves against the elite. But assists and goals are given full credit, so if you racked up 75 assists from free-kicks then fair play to you!

 

Season 112 Attacking Midfielder Of The Year: Marius Log (Alidas)

Season 112 Young Attacking Midfielder Of The Year: Vadis Odjidja (Bruges Blues)

Season 113 Attacking Midfielder Of The Year: Tomasz Hołota (WKS SLASK)

Season 113 Young Attacking Midfielder Of The Year: Adrien Leroy (millou_psg)

 

Marius Log is posting his best performances at the tail end of a long career, and in Season 112 starred for Belgian team Alidas in both the league and (although it isn't counted here) the Super Cup. Interesting to note that he managed this with perception in the mid-70s, too. The best young prospect (at least back in the mists of time that were Season 112) was also plying his trade in the Belgian top division, as Vadis Odjidja rated 91 for Bruges Blues. Tomasz Hołota ran riot in the Polish league last season, with a performance rating of 97 and direct involvement in 48 league goals, more than half WKS Slask's league total. And Adrien Leroy? His performances for a generally weaker team were not just enough to win defensive midfielder of the year last season, but also attacking midfielder. A genuine all-round overperformer who has the potential to blossom if the team around hims improves, or if a bigger fish snaps him up.


 

Attacker Of The Season

In a similar approach to goalkeepers, for this award we're going to combine performance with shooting accuracy (at least for Season 113, where we have the data). A quick note of caution - shooting accuracy is calculated per team (though generally the better performer on a team will have contributed the best shooting percentage anyway), and of course that places more of the focus on finishing, and less on carving out the chances themselves. But of course creating chances seems to have a very strong effect on performance anyway, so this combined measure might get a little closer to what most managers seek in a top striker: Impeccable goalscoring chops.

 

Season 112 Attacker Of The Year: Claudio Gonzalez (Los Cruzados)

Season 112 Young Attacker Of The Year: Czernia (DansendHert)

Season 113 Attacker Of The Year: Marco Arnautovic (FC GPRO)

Season 113 Young Attacker Of The Year: David Johnson (Classic Ipswich)

 

How do you become perennial champions of a league? Having a striker like Claudio Gonzalez helps a bit. The veteran posted a career-high season in S112 with 37 goals and an average performance of 86. Meanwhile, a 16-goal striker might seem like an odd tip, but when you take into account the team he was on, Kurniawan Firdaus actually put in a terrific shift for a relegated team with horrible teamstats. No surprise then to see multiple platinum boots in his trophy cabinet from less challenging seasons. His side now languish in the Indonesian 3rd division, playing (and losing) at low fitness, and Kurniawan is still in his prime... here is a striker with high potential elsewhere if his employers can be persuaded to cash in. As for younger prospects, DansendHert are putting together a scarily good future spine, and Czernia's 27-goal season (plus 4 Champions League goals) wins him Young Attacker of S112. Last season, Marco Arnautovic fired FC GPRO to the German title, hitting the target with around 5 in every 6 shots. And after a transitional year following his move from Multiple Scoregasms, David Johnson hit his stride for Classic Ipswich, scoring 34 goals in the league and Super Cup and posting an average rating of 78.


 

Sad Keanu Award For The Biggest Superstar To Spend The Whole Season Sitting On The Bench On His Own Eating Sandwiches

This award goes to the highest Q outfield player to have played no league or international matches all season. Try tempting their owner with a cheeky bid!

 

 

Shareef al-Hamadhani circa S112. (Q95, Emperor of Thorns. Took a season break in S112 but was back last season.)

 

Thierry Henry circa S113. (Q96, Sauda. Wondering when he is finally going to be trusted to break into the first team.)

 


 

Kudos to everyone who has helped train these players, and especially their teams who are playing a system which gets the best out of them! Looks like just in real life, Belgium is a real hotbed of talent right now.

 
 

- Belizio

 

 

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
The ManagerLeague Manager League! (11/05/2014 08:08)

"Please don't call me arrogant, but I'm European champion, and I think I'm a special one."

- José Mourinho

 

This is a list of the top 100 managers in the game, as judged by my fair and impartial formulae. The score for each manager is the sum of their Manager Ratings, beginning in Season 112. The longer a manager stays at the top, and the more they outperform their team strength, the more points they will rack up and the higher they will appear on this table.

Any manager listed here is a top manager, and reaching the upper echelons is considered an outstanding achievement.


The table, updated to include the results from Seasons S112S113, S114S115 and S116:

 

 

A simple text list of each team in the top 400 is provided below, together with their rank. This is so you can search for a particular team, such as your own (using ctrl+F or whatever works on your system/browser), and if they are ranked 1-100, find them more easily in the table above. Note the team names are listed, not the manager names:

1: Crvena Zvezda 2: Raccoon City Team 3: Hells Angels 4: JustinLaFieber 5: Cork Hibernian 6: MilanAC 7: Bavaria Bulldogs 8: DansendHert 9: IMPARAVEIS FC 10: Parma AC 11: Beachlife 12: Club Sportivo Sergipe 13: FC Judoshiai 14: prince 15: Blest 16: Calgary Kings 17: Seventh Heaven 18: Los Cruzados 19: The Red Warriors 20: Vallaboys 21: Brad G Strikers 22: Lauburu 23: FC Vegan 24: More Beer FC 25: sihanoukville united 26: Bruges Blues 27: Tottenham Oz Spurs 28: TMS FC 29: JK United 30: mitalik 31: Versailles IF 32: Giarra Djinns 33: AKNEO 34: WKS SLASK 35: - STEAUA Bucuresti - 36: FC Ruwenda 37: FC Brescia 38: Indonesian Legends 39: Sovrino SC 40: FC NVilnia 41: Viva el Talian 42: Northcountry Timberwolves 43: Las Leonas 44: The Invincibles 45: SHARP United 46: pge fc 47: Avantu Tulle 48: High Society 49: rene1293 50: Munchen Carpet 51: FC Bortebom 52: Nagyrede FC 53: Majesty in Heaven 54: Old Folks FC 55: CORK CITY 56: Kamu FC 57: Atletico Madrid CF 58: Real Carlikos FC 59: Invisible Sun 60: Fishing club 61: Champion FC 62: Tôi Yêu SFVP 63: L0V3 UNITED 64: PoolEver 65: Taffelstickers 66: Grien Bey Pakkers 67: Krazy Kickers 68: Panathinaikos FC Mouries 69: FC Bananas 70: Dazzas Reds 71: FC Independiente 72: FC bal op het dak 73: Klaebs Kings 74: Red and Black Attack 75: P1cass0 76: Vagabonds FC 77: Manfredonia 78: Dinamo Red 79: millou_psg 80: CzarneKoszule 81: Arabian Samurai 82: Timberwolves FC 83: ERZURUM SK 84: FC Nielsen 85: FC GPRO 86: FC Dondukovo 87: VN WinMySelf 88: Yukas 89: Partick Thistle Nil 90: Hunters FC 91: Lubricentro Gareca 92: FC MUCU SEINI 93: Quebracho FC 94: Goals Galore 95: Melres D C 96: FC Bioul 97: Original 98: SPORTACADEMY 99: Craiova - Maxima 100: ALL SPURS 101: Brave Orc 102: 11 BALAS 103: Riverhead Blue Waves 104: Malabar United 105: R E S T R A I N 106: -ToRoS KaPLaNLaRI- 107: Band of Brothers FC 108: Halifax City 109: Aguia FC 110: La Revancha Futbol Club 111: FC Manish Daskalovo 112: Connemara United 113: MyDogGabe 114: Laudrup Finten 115: Ghuss 116: Loewit United 117: FC Dzhebel 118: DEATH HOUSE FC 119: Sons of Anarchy 120: Alex FC 121: Fc Craggy Island 122: Losers FC 123: Toripolliisit 124: GDB Racing 125: ICPU AllStars 126: FC Knudde 127: Adamska DaviDs Team 128: Slycujos Gladiators 129: Tadinates 130: CSKA LULIN 131: APACHE 132: BYM FC 133: Cunvamiu 134: Royal Arsenal FC 135: Kotabaru United 136: red rebels fc 137: FC Jaubinho 138: Ricfig FC 139: PSCM raiders 140: RKS Motor 141: viitorul dragasani 142: Dúnedain Rangers 143: The Good Rebellions 144: io9456 FC 145: GKS 1977 Belchatow 146: Amethyst 147: Lajhar tevek FC 148: Dilligaf United 149: HFC Poustanidea 150: Comtek 151: Lipin Lad 152: DIV DEDOV 153: Sibodo FC 154: Master MZ 155: Mew2010 156: Real Antoñico CF 157: Audere Est Facere 158: Lyngby BK 159: Digital Monster 160: FC Verrebroek 161: FC Bengal 162: cantera 163: AC-SMØLF 164: Gosma Verde FC 165: Rapid Fotbal Club 166: INDEPENDIENTE SANTA FE 167: DINAMO ZARAGOZA 168: Hotspurs FC 169: Melbourne Bushrangers 170: ferrari007 171: KA Barcelona 172: Honda 173: Tanus 174: Sonnenhotels Eger 175: Baronie 176: Galatasaray AS 177: AC GUMUSGOZE 178: RB Salzburg 179: Bakkens Drenge 180: Nestorians FC 181: tai_arsenal 182: Old Zeller 183: Skituljci 184: HFC Aristotelis 185: Woolwich Football Club 186: Spartans ™ 187: EgaBurgas 188: Stay With Kop 189: Man City BG 190: Anarchists 191: Super Troopers 192: scratch 193: Caipiras 194: Phoenix Baia Mare 195: Newcastle Singapore 196: Bombero AC 197: Wersten Utd 198: Bosphorus 199: NongHoo 200: Southampton 201: Wurzel United 202: Franco Canadiense 203: perssowens everton 204: HJK Helsinki 205: ewing united 206: DIRTY JOBS 207: Mew 2512 208: Vilnius Iron Wolves 209: Classic Ipswich 210: Northern lights 211: Bedda Team 212: PS-NanGo 213: P.S.N. FC 214: A Portuguesa de Desportos 215: FC Zguravci 216: BANDIRMASPOR 217: Republica MOLDOVA 218: AvengersFC 219: fc mrima 220: Ipswich Town FC 221: Shadow Killer FC 222: Legion_A_kickers 223: IRISH ROVERS 224: alidas 225: Deportivo Pereira 226: camel fc 227: CAFigu 228: IF Limhamm Bunkeflo 229: Harcipocok FC 230: Operative_Greenkeepers FC 231: Huskies 232: Oak Rangers 233: Partisan Marseille 234: Krokeide BK 235: C R Vasco da Gama 236: Squadra United 237: Tunetul 238: Persegres Gresik 239: Scarborough FC 240: Hentridge FC 241: London Plague 242: Fk Dubocica 1923 243: lorde_team 244: FC HVKTMM 245: THB SOTTRUM 246: Berntree Unite 247: Atletico Trini 248: ITB FC 249: Shakhtar FC 250: Angels Share 251: Lions of Darkness 252: Tottenham Welsh 87 253: SPFC pachecoorp SPFC 254: Zmajevi Beograd 255: TRIFYLLI 256: PINGULINS 257: Ourimbah Tigers 258: LUPI DEL MOLISE 259: LUSITANIA FC 260: lero7 261: Stiinta-Dragasani-RO 262: La Prestigiosa 263: Juventus F C 264: Claptown Eagles 265: istanBuLLsFB 266: League of Nations 267: Rising Star FC 268: Voros es Fekete 269: PTERODACTYL 270: FC Real Manager 271: Three Star 272: Giannena Athletic Clup 273: Plovdiv 274: Rebaixado 275: fcs 86 276: Werder Bremen Walle 277: maxima 278: STEAUA1971 DRAGASANI 279: River City FC 280: Wendake fc 281: Holland 1 - A-tack 282: Fino Gardino 283: TNSBO 284: AFC Dzhebel United 285: Rossmore United 286: ASLANSARAY 287: Team Natsuko 288: Csengele GG FC 289: Moving IL 290: BEKASI 1988 291: Nottingham Forest FC 292: Watever 293: OLIMPO 294: FC Performanta AI 295: Zwillinge Krieger V.VI 296: Cinkak FC 297: Salgótarjáni BTC 298: ns cempaka 299: Champ20ns 300: Melissokomoi 301: Zenit Football Club 302: Proton R3 FC 303: Lazio_VN 304: RoadRunners 305: Arslanlar FC 306: FC Rockon 307: Hökarängens BK 308: Galatasaray_Amsterdam 309: St Domingo FC 310: Raus finest 311: BLACK PEARL 312: White Tigers 313: Darkness Endures 314: Fatal Justice 315: Kypseli Aliens 316: CHARLYALIKANTE 317: kavalacio fc 318: Habaneros 319: Bangalore Hitmen 320: Lakehead FC 321: RG Football Club 322: Zjednoczeni Piotrków 323: tympass 324: Samin Indonesia 325: Exiled Eleven 326: New Era 327: Thieu Lam Tu 328: Stela Rosa Belgrade 329: Lolita 330: Aix FC 331: ORZEŁ OLKUSZ 332: FC Clay 333: LongѮHổѮPhụng 334: -THE WARRIORS- 335: +FC Bayern München+ 336: GRYFON 337: Nyomdelore FC 338: Csak-A-Diósgyőr 339: Ruskyniai TM 340: Comigteam 341: Green Brigade Fc 342: Pacifica 343: fc boca seniors ac jk pk 344: Mellunmäki FK 345: Chester City FC 346: Liverpruul 347: Cleveland City FC 348: Grobari 1970 349: Watain 350: GSE United 351: Steppenwolf 352: Sauda 353: Busby babes 354: Chocobo Knights 355: Gaz metan medias89 356: Slip Heights FC 357: Rudeboys FC 358: New Generation 359: Portsmouth 360: Bolton Wonders 361: LokomotivNN 362: Ghiri Mannari 363: CA Huracan 364: sansai villa 365: Granvegen Rovers 366: Emperor of Thorns 367: Toreknall 368: Caritas 369: Komren united 370: AC Jim Beam 371: Cá Gỗ 372: Thermal KiSSeS 373: Kitchener Saints 374: Grand Old Team 375: tm punishment 376: Juventus29 377: Violet Blood 378: The Lucky Turtles 379: Evil Team 380: Labu Bedepot FC 381: Joauma FC 382: UD Las Palmas 383: Kick Ass Ninja Penguins 384: The Dream 385: RationalXI 386: The Crows 387: ZOLA TEAM 388: Israel 389: Corfebol 390: Rain City 391: Kujawiak 392: FC ADELAIDE CROWS 393: Os Juniores 394: Tollcross Rovers 395: INDIAN COLTS 396: Valminor 397: ACMilan89 398: Sandmen 399: Manche United 400: Asocial Team

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Wonderstar Awards S112: League Of The Season (04/05/2014 09:08)

No man(ager) is an island, entire of itself, every man is a piece of the continent, a part of the main.

- John Donne

 

Well, the end of Season 113 is upon us, and in Belizio's disorganised world that means it's time to catch up on some of the awards from Season 112 :-)

In a previous post I introduced an award recognising the top managers in the game - with the emphasis on overperforming their basic team strength. The aim of this award was to recognise some of the more skilful players at the game, and highlight some impressive achievements that might otherwise have flown beneath the radar - taking a low strength team to third in the league, that kind of thing.

 

For a rolling, weighted 5-season average of these scores check out the ManagerLeague League of Leagues!

 

But as the poem referenced above reminds us, no manager is an island. So who has the biggest and best continent? The leagues in ManagerLeague vary a lot in depth (number of active teams) and quality. You can measure the strength of a department in a few ways - the default way to do so is by averaging the team quality of the top 12 teams; by this measure the strongest leagues would be the likes of Norway, England and Romania. But of course we don't need to rate teams on paper - as we've seen, a good manager can get a team performing well above its quality level - so better to rate them on results! The leagues in ManagerLeague don't face off directly (there's no equivalent to the World Cup) but they do square off indirectly through their top clubs in the Champions League and the SuperCup. So I've put together a table based on each league's average performance in these competitions in Season 112, similar to the performance measure used to calculate the Manager Ratings. The more games won by teams from a particular league, and the further they go in the contest, the higher the league's rating's going to be.

And because the best leagues should be exciting and competitive, we can also throw in a rating to measure the strength and depth of a league - essentially, how strong are the other teams compared to the top few? If one side runs away with the title and has it sewn up by round 25, that's less exciting than one where the top five positions are still all up for grabs at the end of the season. 

So, I've programmed the numbers machine to give each league two ratings: One for performance and one for depth. Each ranges from 0 to 100, and a final league rating is made up of two parts performance, and one part depth. So which leagues were the most impressive in Season 112? The table gives each league in order, with columns showing the number of international places assigned to each league, their strength (based on both player Q and team stats, not used in the final rating), their international performance and their competitiveness. Ranks are in brackets:

 

 

Vietnam are our top league for Season 112. Well done guys, for competing strongly both at home and in the international contests! The top four leagues - Vietnam, Poland, Australia and Hungary - were very close, and despite all getting only 4 international places they were way out in front on that score. Akneo took the Champions League title for Poland, PoolEver and Lajhar Tevek reached the semi-finals of each contest, VNWinMySelf were runners up in the Super Cup, and Melbourne Bushrangers and Klaebs Kings both had strong runs for Australia. And plenty of other sides from those countries finished in the top three in their groups. All four leagues should be hassling Spinner to change the distribution of entries!

Norway were our fifth ranked side, not so much due to their performance - which were middling, and much lower than you'd expect from the strongest league on paper - but because of a phenomenally tight top division. Just eleven points separated 1st from 7th, and Goals Galore took the title with just 58 points - that's 6 fewer than the previous season, when they finished third... This season Toreknall have secured the title before the final round of matches, but the race for the international spots is still very exciting.


- Belizio

 

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Champions League S113: Group Stage Roundup (15/04/2014 22:48)

Wonderstar Analytics Champion's League Group Stage Roundup

 

A shorter version of this was originally posted on the forums, but I thought I'd place a more permanent and detailed article here so you can admonish me when the teams I tip go out in the first round :-) Others generally provide good coverage of the Champions's League and SuperCup in the forums, so this is mainly an attempt to put some of the teams in context and give a rough guide to how strong they all are. It's quite a quick analysis which shouldn't be taken too seriously, but hopefully it may be a slightly better guide than just glancing at the Q value next to each team.

I've rated each team in 3 areas, putting each rating on a 1-100 scale. I've then combined the three scores to give an overall Rating, also on a 1-100 scale:

1. Team Quality at the start of the competition (a crude but up-to-date measure of strength).
2. Last season's strength (a more sophisticated measure taking into account teamstats and the attributes of the players who actually lined up in league matches, but which does not include squad changes since last season).
3. The Manager Rating 
from last year (a reflection of how much better the manager got his/her team playing than you'd expect just from the stats).

So on this hopefully-better-than-Q-alone-but-still-pretty-arbitrary-if-we're-being-honest scale, which teams look most likely to progress to the knockouts? Original pre-tournament thoughts in green, updates as the tournament progresses in Yellow.


 

GROUPS 1 AND 2

 

Group 1With the highest strength and team quality in the competition, an above average manager rating, and four trophies in the past two seasons, it's very hard to see Vagabonds as anything less than a serious contender for the trophy. A kind draw should make progression a relative breeze - Group 1 is perhaps the least competitive of all, with two clear favourites to progress. Assuming Vagabonds go through, Hotspurs FC (formerly Cursed FC, formerly Hotspurs FC) look best placed to join them. Beyond that, the group looks evenly balanced, though Toros Kaplanlari, Tanus and JustinLaFieber all posted elite Manager Ratings last season and may outperform their modest strength. Round 1 made it clear that Toros Kaplanlari do not intend to take the competition seriously this season, while Vagabonds and Hotspurs/Cursed FC both ran out easy winners. In Round 2 Werder Bremen Walle and River City established themselves as the most likely rivals for the top two, but Werder Bremen's defeat to JustinLaFieber in Round 3 leaves River City v Hotspurs FC in Round 5 looking potentially decisive, with both teams possibly entering that clash on 9pts. Vagabonds beat Hotspurs in Round 4, likely securing the group - Werder Bremen Walle and River City aren't far behind, but have tougher remaining schedules than Hotspurs FC do. Hotspurs FC all but secured second place in Round 5 with a confident victory over River City; a late comeback to beat Werder Bremen Walle 4-3 in Round 6 confirmed their qualification. QUALIFIED: VAGABONDS, HOTSPURS FC.


Group 2If Vagabonds and Cursed FC have a relatively straightforward path to the knockout stages, they should prepare for a stiff challenge there - Group 2 is possibly the toughest of all, with a host of elite managers and three very strong sides battling for just two places in the next round. Adding to the competitiveness is the incentive of avoiding Vagabonds by finishing top. While Blest and VN WinMySelf have the strongest squads on paper, Phoenix Baia Mare were very strong last season and have an excellent manager. Those expecting a three-horse race would do well to be wary of FC Dzhebel and WKS Slask, who will be strong opponents and could take advantage if the top three slip up. Sure enough, WKS Slask pulled off a mild upset by grinding out a victory over VN WinMySelf, while FC Dzhebel and Blest took advantage to move top. Phoenix Baia Mare put some pressure on themselves with a frustrating draw against The Red Warriors. The group opened up even further after Round 2, with all four matches decided by just a single goal. The three lowest quality teams were all complicating the picture with strong performances, with Red Warriors and JK United pushing Dzhebel and VN WinMySelf close, and FC Vegan upsetting Blest. Phoenix Baia Mare got a crucial win against WKS Slask, but in Round 3 the Poles bounced back with victory over FC Dzhebel. Blest beat VN WinMySelf in the crucial clash, which seemed to put the Vietnamese runners-up out of contention - but in Round 4 they bounced back with a thrilling victory over Phoenix Baia Mare. With Blest, Dzhebel and Slask all winning, the top five are separated by just 3 points. Blest and WKS Slask both won in Round 5 to move clear at the top; VN WinMySelf beat Dzhebel to get within 3 points of the leaders. With Blest defeating Slask in Round 6, they should be safely through thanks to an excellent goal difference - although it will take a point against FC Dzhebel to be sure. A 1-0 win over FC Vegan was enough to haul VN WinMyself into second place on goal difference ahead of WKS Slask - qualification coming down to an epic goal difference duel between those two sides in the final round of matches. In the end, WKS Slask dramatically turned around their two-goal deficit with a 7-0 victory over JK United, with VN WinMySelf's 4-1 win over Red Warriors proving not quite enough. Congratulations to the Poles for their qualification, and to Blest who secured top spot with a win over FC Dzhebel. And a tip of the hat to both JK United and The Red Warriors for their sporting behaviour: With nothing technically to play for themselves, both put out their strongest lineup in the final games, ensuring a fair contest for second place. :-) QUALIFIED: BLEST, WKS SLASK.

 

Quarter Final Picks:

After a dramatic qualification, WKS Slask face a tough challenge, to put it mildly. After 23 league and CL matches, Vagabonds have 22 victories and a frankly terrifying +99 goal difference. Some crumbs of comfort: +42 of that goal difference came courtesy of three matches against sides who weren't trying, and their only league defeat came at the Dog House just a couple of rounds ago. Vagabonds can be beaten, but it would be a major shock if WKS Slask were the team to manage it.

Blest vs Hotspurs FC is much harder to predict. While Blest qualified impressively from a tough group, Hotspurs FC - dark horses for the competition according to no less an authority than Hotspurs manager SpudGreg - duly beat the weaker teams in their group, and lost to the very strong one, making it hard to judge their strength against sides of the calibre of Blest. League form is patchy for both sides - Hotspurs FC lie 8th in their league as of R16 (but that underrates them somewhat, being eight points clear of 9th and just two behind 4th) while Blest have a very similar record, sitting 6th in Norway. But the Norwegians have shown an upturn in form recently, and have the slight edge on Hotspurs in terms of player quality and penalty training, so they go in as razor-thin favourites by my book.

 

 


 

GROUPS 3 AND 4

 

Group 3Group 3 looks very open, with a mixture of strong teams (Munchen Carpet, FC Real Manager, FC Mucu Seini) and good managers (DansendHert, Imparaveis, Melissokomoi). The tag of favourite then probably goes to the team with both - Raccoon City Team, who topped a group featuring Vagabonds last season. The simple rating combination above probably underestimates the chances of Munchen Carpet; if their young squad begins to play well together they have the potential to challenge for qualification. A draw between Raccoon City Team and DansendHert in Round 1 isn't fatal for either side, but it allowed FC Real Manager and Imparaveis to move top with victories. Munchen Carpet edged out Greek runners-up Melissokomoi, and followed that up in round 2 with a narrow victory over FC Mucu Seini. Imparaveis joined them at the top with a close victory against Raccoon City Team, while DansendHert also picked up three points but FC Real Manager were held to a draw. Round 3 was a better round for Raccoon City Team, as DansendHert were beaten and Munchen Carpet held to a draw. FC Real Manager beat Imparaveis to move top, before fighting off the challenge of Munchen Carpet in Round 4 - though Imparaveis stayed on their coat-tails after dismantling Melissokomoi. DansendHert won to stay in contention, but a second draw for Raccoon City Team probably leaves them with too much to do in the final three rounds. Rounds 5 and six saw the top two pull away and secure qualification - Imparaveis beat Munchen Carpet and Mucu Seini, while FC Real Manager finished top after victories over DansendHert and Raccoon City Team. QUALIFIED: FC REAL MANAGER, IMPARAVEIS.

Group 4A relatively tough lineup in Group 4, where Kavalacio seem like they're due a good run in this contest, and English champions Beachlife have their eyes on a second title in four seasons. Indonesian Legends have the raw team Q to challenge the top two, and further down MilanAC and Ewing United should be more dangerous opponents than their Q suggests. In the two most telling fixtures from Round 1, Kavalacio and Beachlife laid down markers with impressive wins over Indonesian Legends and MilanAC. In round 2, Kavalacio dismissed the challenge of Dorostol to move top, but Indonesian Legends roared back into contention with a 2-1 win over Beachlife. Round 3 saw Beachlife fall further behind with defeat against Kavalacio, and a surprise win for Toripolliisit against Indonesian Legends. The Finnish side's unexpected run finally came to halt against Beachlife, and defeat to Kavalacio in Round 5 saw them drop back to fourth. Dorostol and Indonesian Legends both won in Round 4, and the Indonesians won their crucial clash in Round 5. With Beachlife unexpectedly dropping points, the race for second opened up for Indonesian Legends. After a straightforward win over ewing united, Indonesian Legends' strong goal difference meant their last round victory over MilanAC was enough to secure qualification - two early goals settling the nerves before eventually running out 5-0 winners. Beachlife leapfrogged Toripolliisit for third, but the Finns deserve credit for an unexpectedly strong challenge. QUALIFIED: KAVALACIO FC, INDONESIAN LEGENDS.

 

Quarter Final Picks:

FC Real Manager take on Indonesian Legends in another tough match to predict. Though Indonesian Legends have the quality edge, the Romanians finished top and unbeaten in what was probably a marginally tougher group, and have the experience of a recent SuperCup victory - all of which would normally make them favourites. However, Andres Iniesta has been ruled out for the Romanians through injury, weakening their midfield, leaving them with only two high-quality substitutes, and probably forcing a switch away from the 4-4-2 formation that served them so well in the Group Stage. This gives Indonesian Legends a golden opportunity to progress, especially if the match goes to extra time.

The description for Group 4 begins with a tip for Kavalacio to go on a run in this contest, and the Greeks duly dominated their group. More surprising was Imparaveis guiding the 5th strongest side in their group to qualification with a game to spare, justifying their stellar manager rating. Although it's dangerous to bet against such an accomplished manager, a 100% group record, a quality advantage of +1.5 and a deep bench is too hard to ignore - Kavalacio are slight favourites here.

 


 

GROUPS 5 AND 6

 

Group 5Though these are two of the weakest groups, they may be among the most interesting this season, since their pairing gives these 16 teams the best chance of a Champions League Quarter Final they are likely to have for some time. In Group 5, the slightly stronger of the two, Sons of Anarchy and Craiova - Maxima field the strongest squads, but Klaebs Kings are a regular in the knockout stages and have the International Manager of the Year guiding them - they won their group with the 7th strongest squad last seaon, so now they have the third strongest the model makes them narrow favourites. Looking beyond the top three, Parma AC may be a decent outside bet. Round 1 saw all the top four win, but in Round 2 Parma AC put the cat among the pigeons with a victory over Klaebs Kings. Sons of Anarchy beat Parma AC in Round 3, while Klaebs Kings and Craiova - Maxima both won as expected. Craiova - Maxima followed that up with a win in Round 4, but all heck broke loose in the remaining games as Klaebs Kings, Sons of Anarchy and Parma AC all tasted unexpected defeat. The carnage obviosuly puts the Americans in control, but left Real Antonico and FC Knudde with manageable schedules, having both already played three of the top four sides. Defeat for both in Round 5 simplified the group however, and after an agonising defeat to Craiova - Maxima, Klaebs Kings challenge ended up on life support, before finally being snuffed out by Sons of Anarchy. In a sign of how tight this group is, Klaebs Kings dropped to 7th, but with a +2 goal difference. Parma AC picked up 9 points in a strong run at the end, but fell short by goal difference to Sons of Anarchy. QUALIFIED: CRAIOVA - MAXIMA, SONS OF ANARCHY.

Group 6Taffelstickers, Ricfig FC and PoolEver all reached the knockout stages last season, with PoolEver going the furthest, to the semi-finals. This year the Norwegians and Portuguese look to have the strongest squads, but they may also face competition from several other teams in the group. Cork Hibernian in particular may be worth keeping an eye on, posting an impressive Manager Rating last season on their way to ousting perennial Irish champions Connemara. Taffelstickers, PoolEver and Cork Hibernian won their opening matches, with SPFC Pachecoorp holding Ricfig to a draw. In the second round, Cork Hibernian and Ricfig were held by the two Brazilian sides, leaving Taffelstickers and PoolEver to move clear at the top of the group. In Round 3 Taffelstickers took control of the group with a 2-0 win over PoolEver; following that with a victory over unbeaten Ricfig should all but secure qualification for the Norwegians. Meanwhile Cork Hibernian kept pace with the leaders with a second victory; but they dropped points again in Round 4 leaving them a tough challenge in their final three matches. Taffelstickers beat Ricfig and PoolEver secured three points against CheatAh in Round 4, then Round 5 victory for the Norwegians against Cork Hibernian and an emphatic PoolEver win over Ricfig FC left the Irish underdogs as the only team who could stop the top two at that point. A win over PoolEver in Round 6 kept them in contention, but the Hungarians did just enough with a victory over Pachecoorp in the last round. Although RicFig were slightly disappointing, the preview above now looks prescient, with dark horses Cork Hibernian almost snatching qualification with a side ranked 62nd of 64 (!!!) in terms of Q. They can now turn their attention to an incredibly tight domestic title defence with their heads held high, while PoolEver and Taffelstickers return to the knockout stages for successive seasons. QUALIFIED: TAFFELSTICKERS, POOLEVER.

 

Quarter Final Picks:

Sitting fifth in an increasingly strong USA league, and with an ageing defence figuring out retirement options, Craiova Maxima could be forgiven for having a now-or-never attitutude to the Champions League, and a strong performance in the group stages suggests they're in with a decent chance. After all, it worked for Chelsea! First they'll take on PoolEver of Hungary, who reached the semi-finals last season. My bet is they'll come up two rounds short this time around, and Craiova Maxima will reach the quarter finals.

Taffelstickers v Sons of Anarchy is very finely poised - the Swedes have the edge in team Q, but Taffelstickers came through their group with more assurance, and look to have a slight edge on teamstats. Taffelstickers top the Norwegian league as of Round 16, which suggests they're getting the best out of a strong team - that makes their manager rating of 10 from last season look like it's underestimating them, so we'll give the edge to Taffelstickers here. But the upsets in this contest usually come at the expense of high-Q Norwegian teams, and group winners - and Taffelstickers are both.

 


 

GROUPS 7 AND 8

 

Group 7. In Groups 7 and 8 we firmly leave the land of opportunity behind and enter the groups of death. Ten of these 16 teams have ratings of 84 or above; twice as many as the other groups have, and (at least) 8 of them will be out by the time the Quarter Finals roll around. In Group 7 a host of strong teams are nearly impossible to pick between - Champion FC have the highest team quality, but the model likes LongѮHổѮPhụng and Los Cruzados based on past performance. KA Barcelona and Czarne Koszule are always in contention, one should never discount the English team - least of all one as strong and experienced as Giarra Djinns - and Bruges Blues overperformed their Q as well as anyone in S112. Bosphorus must be wondering what they did to deserve this draw. Yet in round 1 Bosphorus pulled a hard-fought and unlikely victory out the bag against LongѮHổѮPhụng - in a group this unforgiving that could be a fatal wound for the Vietnamese. Meanwhile Giarra Djinns and KA Barcelona scored important wins against CzarneKoszule and Bruges Blues, while Champion FC made a huge statement by thumping Los Cruzados 4-0. Round 2 emphasised the open nature of this group with draws between Los Cruzados and Giarra Djinns, and LongѮHổѮPhụng and KA Barcelona. CzarneKoszule moved back into contention with a win over Bosphorus, and Champion FC secured the points against Bruges to move top. A Round 3 victory over Giarra Djinns put Champion FC firmly in control, while Los Cruzados picked up three points to stay alive and KA Barcelona won a crucial matchup with CzarneKoszule. A 1-1 draw between Bruges Blues and LongѮHổѮPhụng probably spells the end for both teams. In Round 4 Champion FC extended their lead with a win over LongѮHổѮPhụng but Los Cruzados scored a crucial victory against KA Barcelona to leave the battle for second wide open. Los Cruzados, KA Barcelona, CzarneKoszule and Giarra Djinns were all within a point of eachother with three games to play. Another victory in Round 5 secured impressive qualification for Champion FC, but the melee behind them is still very open after a draw between Giarra Djinns and KA Barcelona. In Round 6 all of the chasing pack scored the victories they needed, so four sides remained in contention going into the final game. CzarneKoszule capped a thrilling comeback in the group after two early defeats to beat Los Cruzados and join their compatriots WKS Slask in reaching the knockout stages. The trophy might yet stay in Poland. QUALIFIED: CHAMPION FC, CZARNEKOSZULE.

Group 8Group 8 is if anything even tougher, with three title contenders in Goals Galore, Brad G Strikers and recent winners Dinamo Zaragoza, joined by last season's finalists Las Leonas, one of the game's top managers in Northcountry Timberwolves, and no obvious weak teams. The model gives the edge to the three strongest teams in terms of Q, though picking two from those three is a fool's errand. In the first round, Brad G Strikers won a terrific US-Canada matchup 5-3, while Dinamo Zaragoza and Goals Galore also posted impressive wins to make us a little more confident in the top-three picks. Round 2 shook things up a bit though, with Brad G Strikers and Goals Galore (living up to their names so far with 12 goals between them) drawing 2-2 and Crvena Zvezda roaring into contention with an impressive win over Dinamo Zaragoza. Las Leonas and Northcountry Timberwolves also stayed in the hunt with victories. In Round 3 Northcountry Timberwolves picked up a valuable point against Goals Galore, while Brad G Strikers moved top after an impressive demolition of last year's finalists. Crvena Zvezda couldn't maintain their momentum, slipping up against Manfredonia, while Dinamo Zaragoza moved up to 6 points. The Spanish champions went one further in Round 4 with a close victory against Northcountry Timberwolves to move second; their slip-up against Crvena Zvezda now seems very long ago. Speaking of the Serbians, their brief challenge has faded, losing 4-1 in Round 4 to Brad G Strikers. "With Goals Galore also winning, it still looks likely that Dinamo Zaragoza v Goals Galore in the final Round will be crucial, but it's getting harder to see a successful challenge from outside the current top 3." And with those words, Belizio proved himself quite the fool ahead of Round 5, where shock wins for Crvena Zvezda and Manfredonia over Goals Galore and Dinamo Zaragoza blew the race for second open enough that even 6th-placed Timberwolves had a potential path out of the group. After Brad G Strikers beat Dinamo Zaragoza in Round 6 to secure top spot, Goals Galore moved up to second place, and eventually qualified with a win over the Spanish side in the last game. This was one of the most exciting and open groups, and I'm a little disappointed to say goodbye to Crvena Zvezda, who capped off an absurdly up-and-down campaign by thumping Northcountry Timberwolves 4-1. QUALIFIED: BRAD G STRIKERS, GOALS GALORE.

 

Quarter Final Picks:

With all the other four Norwegian and American sides topping their groups, this was the only matchup between sides of the two strongest countries before the quarter finals - and Brad G Strikers came out top. CzarneKoszule qualified in thrilling fashion from a very difficult group, but a much more serene progression for Brad G Strikers makes me suspect we'll see a second round-of-16 victory for the USA over Poland.

Finally, Champion FC took their group in style, fulfilling a pre-tournament prediction from Flirtybee that the Australian sides might be dark horses. They'll face Goals Galore, who have been hotly tipped for a couple of seasons, and have already broken the Q98 barrier. Despite the spectacular player quality on show in this tie - the highest of all the Round-of-16 games - both teams have their weaknesses, losing once each and also racking up 3 draws between them. Another tough one to predict, and while it's hard to imagine all of the Norwegian and American sides getting though without at least one upset, it's also tough to tip defeat for the highest Q team in the contest. Goals Galore by a whisker.

 

 

 

- Belizio

 

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Wonderstar Awards S112: Manager of the Season (11/04/2014 17:52)

 "I wouldn't say I was the best manager in the business. But I was in the top one."

- Brian Clough


Welcome to Season 113 everybody!

It's been a hectic finish to Season 112 for me, but I'm now back in the country and starting to power through the analyses for last season - so if you've got one ordered, expect it before the first league game. Thanks everyone for being patient! Next month should be a lot easier. In the meantime, let me congratulate everyone who had a successful season, especially those who've signed up for Wonderstar Match Analytics. Of the 15 such teams who began in the top division of their league:


14 finished in the top 6
10 finished in the top 3
5 won their national league title
2 more won the league cup (and 4 others reached the final)
Almost all - 13 out of 15 - outperformed their Q and teamstats


Very, very impressive! Of course, the type of manager who buys credits and signs up for extra detail about their team is often already ahead of the pack - and certainly some of my clients are among the most respected managers in the game - but it is nice to see the attention to strategy paying off so impressively. Well done everyone, and also to those who gained promotion from further down the league ladder! The Wonderstars also outperformed their modest Q this season (11th highest in the department), finishing 3rd and taking home the Department cup. They even unexpectedly won their first Player Cup and promptly followed it up with two more in quick succession - life is fun down here in the lower divisions too!

This got me thinking - which managers have managed to wring the best out of their teams? So that's what the inaugural Wonderstar Manager of the Season aims to reward.


 

Wonderstar Manager Of The Season

To work out the Manager Of The Season, I calculate a Manager Rating. This reflects how much each team has over- or under-performed relative to their basic team properties: The quality of their players, and their total teamstats. Everything else - finding players who outperform their Q, designing tactics that give you an advantage, training the right balance of teamstats for your strategy etc. etc - these factors will all help improve the manager rating. I have posted the details further down, but roughly speaking here is how it works:

 

- Your Strength (based on your lineup quality and number of teamstat stars) is calculated and compared to the teams you played.

- Your Performance (based on goal difference, points, and bonuses for winning silverware or finishing in better positions) is also compared to your opponents.

- This is done for both domestic and international contests then combined, and a Manager Rating is calculated which reflects how high your Performance was relative to your Strength. Low Strength, High Performance gives you a good Rating!

- This means your player Q, injuries/suspensions, total teamstats and opponent strength are all taken into account.

 

So in total, there are 33 League Managers of the Season (one for each national league of course), one International Manager of the Season (best Manager Rating from their Champions League or Super Cup campaign) and one overall Wonderstar Manager Of The Season! Below I'll list the top-10 overall, the top 5 international, and the top 2 from each league, but if you'd like to know what your score is just message me. Remember that only teams finishing in the top 12 of the first division will have a score though. 


 

Results

Without any further ado, here are the top managers of Season 112. Congratulations to our overall champion, the Season 112 Wonderstar Manager Of The Season:

 

1MilanFan of MilanAC, Germany

 

1MilanFan finished ninth in Season 111, but despite a decrease in team quality - only the 10th best in the league on paper - in Season 112 they fought Kamu FC all the way to end, finishing second by just a single point. Twenty-three consecutive seasons in the top division is an impressive achievement too, in a game where aggressive rebuilding is often considered crucial to build an elite side. Don't underestimate them in the Champion's League this season! 1MilanFan will get a full complimentary team analysis for last season as a prize - though of course the prestige and respect of fellow managers is the real award :-) Here's the full top ten, representing the very elite of managerial talent last season:

 

 

 

Imparaveis capped their 12th consecutive top-flight season with their third league title, and performed stronger than expected in a tough SuperCup group. JustinLaFieber have made rapid progress since promotion three seasons ago, and took the domestic double in the Netherlands - their first ever national trophies - along with a creditable SuperCup performance. Raccoon City Team dominated the Italian league, also winning the double, and reached the quarter finals of the Super Cup - winning a group featuring Vagabonds, Blest and Zi Huha. Las Leonas dominated the Argentinian league and made an unexpected run to the Champion's League final. Fishing Club missed out on the Thai title by just a point, but had a much better goal difference and won the League Cup instead, while Northcountry Timberwolves held off a ferocious challenge from theoretically-stronger River City to win their 12th consecutive Canadian title by goal difference, adding the League Cup too - finally toppling FlirtyBee will take more than just higher quality and teamstats. Harcipocok FC took the 11th strongest team in Hungary to 3rd place, and FC Vegan won the Danish league with the 5th strongest side. Rounding out the top ten, Klaebs Kings' dedication to hidden stats earned their modestly-strong side victory in both the Australian league and a tough Champion's League group. And just missing out on the top ten - but worth mentioning - Cork Hibernian broke Connemara's stranglehold on the Irish title, and the two top Belgian teams also surprised a few people - Bruges Blues won four of their seven Champion's League games, and DansendHert proved last season's MLCL run was no one-off with another romp to the Semi-Finals.

And looking just at performances in the Champion's League and Super Cup, it was Klaebs Kings who outperformed their strength most impressively, beating 6 stronger sides to win their Champion's League group and ultimately reach the quarter finals. Congratulations! The rest of the top five were as follows:

 

 

Both Champion's League finalists AKNEO and Las Leonas are on there, reflecting the slightly surprising nature of that match-up in what was a very strong field this season. Ewing United reached the knockout stages of the Super Cup, while Zi Huha put up a stronger than expected fight in one of the most competitive Super Cup groups, demonstrating that their recent appearance in the CL final was built on more than just a strong squad. 

Finally, the Wonderstar Manager of The Season for every league (based on both domestic and international scores) is listed below:

 

 

  

Congratulations to everybody listed. There will be another set of awards at the end of this season, and also be a "running total" type contest, rewarding those managers who skillfully and consistently stay in the upper echelons. Good luck! Meanwhile I'm off to run some analyses, but look out for our other Wonderstar Award soon: League Of The Year.

 

- Belizio

 


 

Manager Rating Details

Here are the details (this is subect to revision year-on-year but gives you a good idea of how the scores are calculated):

 

1. The Manager Of The Season is open to Division 1 sides only. But if you're a top manager, you should find yourself in the elite soon enough! Furthermore, to be eligible your team must have finished in the top 12 of your National League, i.e. avoided outright relegation. That gives us 396 teams from 33 National Leagues.

2. The Manager Rating is calculated from the difference between Performance (how well your team did) and Strength (how good your team is on paper relative to your opponents). It is converted to a 100-point scale which is roughly linear (meaning there are about as many teams with score 50 as there are with score 70, or with score 10). A weaker team who perform well should end up with a high Manager Rating, a team performing about as expected should get around 50.

3. Manager Ratings are calculated separately for a) domestic performance, and b) Champions League (CL) or Super Cup (SC) performance, then combined. Sides not in international competition just use their domestic score, those in international competition get 80% of their score from their domestic performance and 20% from performance in the Champions League or Super Cup. In theory, the way it is set up, being in the CL or SC should be neither an advantage nor a disadvantage: If you perform well in both contests it will improve your score compared to someone who performed the same as you domestically, but didn't play internationally. On the other hand, if you underperform internationally it will drag your score back down. In practice, teams who outperform their Strength in the league also tend to outperform their Strength internationally, so 20 of the top 25 managers came from teams in CL or SC last season. But this does mean that if you want a high Manager Rating, you need to take the CL or SC seriously.

4. Your Domestic Strength score is calculated from two factors: Your average lineup quality and your total teamstats (in stars). Lineup quality is the actual quality you put out on the field in league matches, not your team quality - it is calculated from the number of appearances each player made, so if you suffered from injuries and suspensions all season, don't worry - that's taken into account. Your lineup quality is then compared to that of the other top-12 teams in your league - the bottom four teams are ignored as these can sometimes include bot teams, sides deliberately getting relegated, or just weak opponents. Next, teamstats are measured sometime during the middle of the season and the total number of stars added together - this is also compared to your top-12 domestic rivals and combined with quality to form your relative strength (i.e. how good your team is compared to your domestic rivals). A stat which is at 0.5 stars is counted as zero, since 0.5 stars can mean anything below 56, not just 51-55. The formula for combining teamstats and quality is determined by a regression analysis which will be refined year-on-year, but lineup Q is roughly 5-6 times more important than teamstats for predicting performance.

5. Your International Strength score is calculated in the same way, but this time comparing your lineup quality and teamstats to the other teams in your CL or SC group (but ignoring the side who came 8th since sometimes they have given up). Impotantly, your lineup quality is based on your league lineups - so if you play a B squad in the Champion's League or Super Cup your manager rating will be damaged! That's deliberate, part of being an elite manager is rotating your squad effectively to be competitive in every match.

6. Your Domestic Performance score is calculated by comparing your goal difference to the other top-12 teams in your league, comparing your points in the same way, combining those two factors equally, and then adding increasing bonuses for finishing in the top 5 or winning the League Cup. Winning the title and the league cup both get you the largest bonus (though winning the title also means you got more points, so in practice that's worth more to your performance score even though the bonus is the same). Because points and league position both improve your score, winning games is the most important factor - but goal difference does help too, so losing 3-2 instead of 3-0 will still be better for your performance rating.

7. Your International Performance score is calculated by comparing your goal-difference and points to the other teams in your group (ignoring the side who came 8th). Bonuses are then added for finishing position in the group, as well as progression through the competition - obviously winning the whole thing gets you the biggest bonus!

8. The Super Cup and Champions League are weighted equally, and domestically all leagues are also weighted equally. But because your opponents are likely to be stronger in the Champion's League, or in a tougher league, the same team would have a lower Strength score when playing in those tougher contests (because of harder opponents), so they'd get a higher Manager Rating for the same Performance. In other words, a Q92 team finishing 2nd in Norway would get a much higher Manager Rating than if they finished 2nd in Thailand.

9. Finally, Strength score are deducted from Performance scores to give Manager Ratings, the domestic and international components are combined, and everything is converted to the 100-point scale.

10. The exact formulae for calculating Manager Ratings may be tweaked season-on-season. There are also no guarantees of accuracy - in particular, large volume sales or purchases of players at a particular time have the potential to impact the strength ratings, and therefore the manager ratings (for example, a team sells all its good players just before I download all the player data, making their strength seem lower than it really was during the season). I'll carefully check the teams who finish very high, but with ~500 division 1 teams per season I can't check them all! The good thing is these effects are self-limiting, i.e. they can't last for multiple seasons so shouldn't have much of an effect on the overall Manager Table. And while I can't guarantee perfect accuracy or unchanging formulae, I do guarantee that no changes are made to benefit any particular team over another.

 

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Player Attributes: Caveats & Details (24/03/2014 08:46)

"Tackling? I don't know what to believe any more. The numbers drive me to madness."

- Belizio

 

This isn't a full blog post, rather a place to store a bunch of caveats, musings and technical details related to this already-way-too-long manuscript about player attributes. If you haven't read it yet, go there first! Fair warning though - I end up claiming that Shooting is only the fourth most important attribute for a striker. And I do so with a straight face.


 

Caveats and Details

I've mentioned some caveats as I've gone along, but I think I'm going to keep a list of them down here, along with some more technical details and asides that i think are important for fully understanding the patterns we've seen above. I'm happy to update this too as people suggest things! First though, a caveat to the caveats: I've considered and factored in all of the issues below, and I still think the analysis is useful (and gives us a lot of unique, objective data that complements the intuition and rumour that we currently rely on). Put it this way - if I thought any of the caveats made the analysis worthless, I wouldn't have posted it at all! Having said that, as always everybody is free to judge the evidence for themselves believe however much they like - after all, the more accurate you are, the greater your advantage as a manager.

 

:-)

 

1. The crowdsourced and regression formulae here are trying to predict one thing, and one thing only - match performance. In general, the performance rating is a pretty good measure of how useful a player was during a match, but it won't be perfect. Any improvements you can make to how performance is measured will change the formulae a bit.

UPDATE: It's become apparent from reading the forums that quite a few people see performance as not just imperfect, but virtually useless, and that it has little effect on match outcomes. I have my own reservations about performance, but it so happens that I have a dataset of 456 matches with both the performance ratings and scores for each team. The difference in performance correlates exceptionally strongly with the final score (R=0.88). Here it is visually:

Figure 1: Difference in goals scored versus difference in match performance for 456 games 

 

So although perfromance doesn't perfectly predict match outcome - luck plays a role - it's pretty damn close. The better-performing team in ManagerLeague win their game around 90% of the time, which is way more predictable than real life. It's also worth noting that performance is more closely tied to goals scored than it is to goals conceded. That's probably why performance is more predictable for attackers and midfielders than it is for defenders (see Figure 2 further down) - it's because defender performance is more dependent on how well their opponents played.

2. Value is not determined by a formula, it is determined by the market. So if a majority of managers judge players using their Q value, rather than these more sophisticated formulae, then your striker with excellent heading might not fetch as high a price as you'd hoped. On the other hand, you should be able to find value by buying players who are likely to outperform their Q.

3. The attributes used in each position will probably change with tactics, formations and position. For example, if you shoot from distance your midfielder shooting might become more important, and conversely your attacker heading might be required more, since you should earn more corners after saves. Alternatively, if you're shooting only when safe, then shooting and perception should be more critical for your strikers to avoid the offside trap and finish, while midfielder might end up using passing more. And of course a winger could use a different balance of attributes than a defensive midfielder.

4. Attributes are not the only factor to explain performance! Amongst other things, fitness, tactics, teamstats, opposition quality, lineup age, hidden attributes and many more factors are involved. Most of these should average out in the regression analysis, giving us pretty stable attribute estimates, but that does not mean they aren't going to have an effect in your matches.

5. Having said that, player attributes are very important for performance. Here's a way to think about it. If you take two random players in a given position, the one with the higher ability (according to the regression formula) has around a 70% chance of performing better over the course of a season.

6. The regression formulae help you predict performance more accurately than the standard Q formulae. This graph shows how much of a player's performance is predicted solely by his attributes (known as R-squared):

 

http://www.managerleague.com/images/uploaded/1061224/rsquared.png

Figure 2: Amount of performance explained by standard Q formulae (solid bars) and regression-derived formulae (lighter bars). Using the regression formulae gives a significantly better prediction of performance than Q alone. Certain positions are less predictable than others, for example defender performance is probably more heavily determined by opponent skill than attacker performance is. Even the simple Q formulae tell you a lot about performance - with these sample sizes, an R-squared of even 1% would be sufficient to demonstrate a real relationship; here R-squared reaches as high as 37%.The rest of the variance is split between everything else - formation, lineup age, minutes played, opposition strength, pressure, fitness, hidden attributes, positioning, playstyle, and - biggest of all - chance. With the exception of chance, it's highly unlikely any of those factors on their own gets very close to explaining the amount of performance that attributes do. Attributes aren't the whole story, but they're almost certainly the biggest part of it.

 

7. The regression analysis might be affected if an attribute is correlated with something else that affects performance. For example see the entries above on Goalkeepers and Midfielders for how Age, Passing, Perception, Heading and other stats could be influenced by positioning, roles at set pieces, or gaining rates throughout the season.

8. Some stats have a benefit or cost that isn't easily captured by performance. For example Stamina helps replenish fitness between games, while Age affects the value of the player, as well as the performance of the players around him.

9. Regression estimates will change with additional data; they shouldn't be considered perfectly accurate. All stats estimated at >4% (i.e. every positive value except stamina for midfielders or defenders) is considered to be reliably greater than zero by the regression. The average error of an attribute's importance is about +/- 4%. Likewise, that means even if an attribute is rated at 0% in the regression, it's still possible that it's actually 2% or something, because of that margin of error.

10. The "Tackling = Ball Control" theory is not perfect, and it could be totally wrong (though how you explain the regression then, I have no idea). For example, we might expect tackling to matter more for midfielders than it does. At some point I will run a more detailed analysis comparing tackling attributes to actual ball control outcomes in matches. That should give a more decisive answer. Much like Friedrich Nietzsche, it also raises the troubling question of whether a cruel and indifferent Spinner might have allowed us to be mislead until now.


- Belizio

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Contest Winners (21/03/2014 05:16)

"Perplexity is the beginning of knowledge."

 

- Khalil Gibran

 

Last week I set you guys a small challenge, while I finished up Season 111's Match Analytics: To guess how important each attribute was for performance in a match, all else being equal. Of two otherwise-identical midfielders, would the one with higher tackling or higher speed perform better over the course of a season? Does an extra perception point on your attacker help him more or less in a game than an extra heading point? And do older players really perform better than younger ones with similar stats? To figure this out myself I ran a large multiple linear regression, to see how each of the 8 visible attributes and age contributed to performance for 5000 players. And I offered some prizes for those of you whose predictions most closely matched the results of the analysis.

 I'll discuss the results in detail in a separate post, but in the meantime here are the contest winners. Congratulations!


 

Goalkeepers

The closest five guesses for goalkeeping stats were:

1. Rational

2. Elligen
3. Carrotcruncher
4. FlirtyBee
5. Belizio
 

Ratty correctly identified Keeping as being the only primary stat undervalued by the standard Q formula - it should actually be weighted closer to 45%. Passing and Stamina on the other hand had no identifiable effect on performance, while Age was significant at 20% - in other words if you have two similar goalkeepers, play the older one!


 

Defenders

The closest five guesses for defending stats were:

1. Elligen

2. Carrotcruncher
3. Wiggle
4. FlirtyBee
5. Mystz0r

 

Elligen was second closest for goalkeepers, but had the most accurate guess for defender performance. Defenders show a very balanced reliance on four main attributes - Tackling, Speed, Perception and Age, all at around 20% - with Passing and Heading making up most of the remainder.


 

Midfielders

The closest five guesses for midfield stats were:

1. Moshu

2. FlirtyBee
3. Elligen
4. Carrotcruncher
5. Getzome

 

Most people got pretty close with midfielders, who need to be flexible - every attribute except Keeping contributed in some way to performance. Moshu was nearest, mainly because he rated Passing quite highly. Despite the fact that lots of attributes help midfielder performance, Passing (32%) is still the most crucial one, closely followed by Speed (26%), both of which most people underrated. Perception is quite important but at 11% was actually overrated by most people, as were Shooting and Tackling. Age was worth about 10%.


 

Attackers

The closest five guesses for attacking stats were:

1. Mr Uttisrud

2. FlirtyBee
3. Truthseeker
4. Wiggle
5. Rational

This is the one that tripped most people up - myself included - and revealed the most interesting results. Mr Uttisrud got closest out of everybody, and wins the credits. I'll go into the details in the next blog post, but for now just note that almost everyone had Shooting as the most important attribute, except for Mystz0r who rated Speed higher. Guess what? Shooting's not even in the top three.

And nor is Speed.


 

Overall

Here's the overall euclidean distance between everyone's guesses and the results of the regression, rounded to 2 decimal places. Distance in brackets, smaller is better! FlirtyBee didn't get any of the individual prizes, but finished in the top four for each - and wins the prize for Closest Overall. Kudos also to Carrotcruncher who ranked 2nd, 3rd and 4th for Defenders, Goalkeepers and Midfielders. 

1. FlirtyBee (0.47)

2. Rational (0.49)
3. Carrotcruncher (0.49)
4. Elligen (0.55)
5. Wiggle (0.55)
6. Belizio (0.59)
7. Mystz0r (0.61)
8. Getzome (0.61)
9. Mr Uttisrud (0.61)
10. Moshu (0.68)
11. Spinner (0.69)
12. TruthSeeker (0.69)
 
 
Overall you can see that everyone's idea of what attributes are important was at least as accurate as the standard Q formulae (Spinner in the table above), and most were a big improvement. In other words, in broad terms the results of the analysis match up to what experienced players have intuited from playing the game. But there are some very interesting differences, which I'll go into in the next post... 
 
 

- Belizio

 

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Player Attributes (21/03/2014 04:49)

 "Do not be blinded by the Q."

- Spinner

 

Last week I set you guys a small challenge, while I finished up Season 111''s Match Analytics: To guess how important each attribute was for performance in a match, all else being equal. Of two otherwise-identical midfielders, would the one with higher tackling or higher speed perform better over the course of a season? Does an extra perception point on your attacker help him more or less in a game than an extra heading point? And do older players really perform better than younger ones with similar stats?

I think this is an interesting question to ask the community, because when you play this game you quickly realise that the simple formulae for quality in different positions is an imperfect measure of how good a player actually is. I'm talking of course about this one:

 

Figure 1: Player Quality Formulae for each position

 

To work out the quality of a player, you'd find his position, multiply each of his stats by the numbers given in the table, and add them together. But of course, this means that a midfielder for example could gain 10 heading and 20 perception and his quality value wouldn't change. Yet he'd certainly play better in a match - he'd be more likely to score at corners, play more intelligent passes, make successful interceptions and so on. So can we come up with a better set of formulae?


 

Crowdsourcing A New Formula

One approach is to use the "wisdom of the crowds" - how important do managers in the game feel each attribute is? Let's say you:

- Persuade roughly a dozen managers from 9 different leagues to estimate how important each attribute is, and
- Those managers have 353 full ManagerLeague seasons under their belt and a combined national-level trophy haul of 28 pots, and
- You mix everything all together

Then you might get something that looks a bit like this:

 

Figure 2: Attribute Importance for each position, Crowdsourced

 

So that's an alternative based on the combined wisdom and intuition of a bunch of experienced managers. I think it's valuable - so how does it differ from the standard Q formulae? Well, the first thing that pops out is how much more balanced these formulae are: Each position relies on a wider range of attributes, and is less dependent on the primary stat than the Q formulae imply. Perception is rated highly for every position, which is not that surprising given Spinner's advice that Speed and Perception are now more important in the latest version of the sim. Stamina is held to be moderately useful for every position. Older players are generally considered to play better than younger ones. You could do worse than use this as your guide to player ability, instead of Q. But could you do even better?


 

Measuring A New Formula

Our second approach then is to look at a large number of players in similar teams, and see how strongly associated each attribute is with their performance over the course of a season. There are some more details in the previous post, but roughly speaking we could:

- Find a large number of players - say 5000 - who all play in similar strength leagues, and
- Record their attributes, age, and average season performance, and
- Run a big multiple linear regression for each position to see how strongly each attribute predicts performance

Then, you'd get something like this instead:

 

Figure 3: Attribute Importance for each position, Calculated from Multiple Linear Regression

 

So that all looks OK, Goalkeepers pretty much what we expected, Defenders look reasonable, midfielders fine, attackers... hang on a minute.

Yes, I imagine one stat in particular stands out as being a bit... unexpected. I'll step through each position in turn, and address that in a bit more detail when it comes up. But apart from that, the strengths of different attributes actually make quite a bit of sense. For each position, the formula above is closer to the crowdsourced estimates in Figure 2 than it is to the simple Q table. In other words, the intuition players have about which attributes are over or underrated actually seems to be quite accurate, and is corroborated by the regression. What's more, more experienced managers guessed attribute strengths that were closer to the regression estimates in figure 3: Both the amount of experience a manager had (number of ML seasons played) and the number of national-level trophies they won correlated with their accuracy in the contest. Intuition and objective measurement lead you in roughly the same direction, and more experienced intuition matches up even more closely to what we measure.

So whether you look at pure intuition or a completely objective analysis, you get quite a similar story: More attributes are important than just the 20 in the Q formulae, Perception is indeed important for every player, primary stats are overrated, Age helps (for certain positions at least). Having said that, there are also some pretty big differences, so let's go through things position by position. As we go I'll discuss why I think certain attributes are important for each position, using my own understanding of the sim, as well as reasons why we might see differences across the three methods.

The way we're going to do this is look at all three methods for assessing attributes in each position: The original Q formula, the crowdsourced intuition, and the objective regression. I've also included a simple average of the three for each position. The reason I do this is because each method has different strengths and weaknesses, and in such cases you normally end up with more information when you combine data from different places. So for example, the Q formula is the only one that comes directly from Spinner. But, the sim has been tweaked since these formulae were designed, it's very simple, and it ignores a lot of attributes for each position. Meanwhile, the crowdsourced data reflects the average intuition of managers currently playing the game, so it's more up to date and also incorporates what people have learned from reading Spinner's announcements or the forums. On the other hand, it's based very closely on the Q formula, perhaps too closely - so if the Q formula isn't that accurate, the crowdsourced estimates might not be either. That goes for 'myths' and misunderstandings too - especially if they've been included in a help blog somewhere, people are likely to read them and accept them, without necessarily questioning how true they are. And intuition is pretty useful, but it's subject to all of the brain's usual weaknesses - including a tendency to overinterpret noisy data for example. The regression meanwhile 'looks at' a lot more data than the average human manager, and in a rigorous, carefully designed way. But it's limited by the data you put into the analysis, which in this case is all league games around Q80. If tactics and attributes act differently at other Q levels for example, that might affect the results. And of course they're not exact - roughly speaking, each estimate is probably within about 4% of the true underlying value.

A simple average of the three helps to smooth some of this out. Of course, you can average things in any way you wish - I'd be inclined to weigh the regression data more highly than the other two for example, since the Q and crowdsourced methods are so similar to eachother. But you might have a deep distrust of maths, and decide to completely ignore the regression results. That's all fine! Gathering the data and laying it out gives us all a chance to weigh things up ourselves and use our own judgment. So, let's take a look at each position.


 

Goalkeepers

Here's a summary of the goalkeeper formulae from our three sources: The original Q formula, crowdsourced intuition, and objective regression:

 

Figure 4: Important Attributes for Goalkeepers, estimated from three sources

 

First thing to note: They're all quite similar I'd say, and that's probably because the goalkeeper is the most specialised and simple position on the field. There's only a few things they're ever called on to do - saving shots, claiming crosses, stopping penalties, and passing the ball back out to start a new attack. I've not really seen goalkeepers committing fouls, running out of the box to make clearances, dribbling round an attacker, dealing with backpasses, coming up for corners in stoppage time or anything else like that - the sim keeps things pretty uncomplicated. Keeping is clearly the most important attribute, which makes sense given that the most important part of the job is stopping goalbound efforts. Perception is next, and it's quite possibly used when running out to claim a cross, saving a penalty, and possibly to avoid giving away corners by 'saving' a shot that is going wide. Most managers seem to think the Q formula slightly overrates perception here, and the regression agrees. Speed can probably be thought of as 'reflexes' for goalkeepers and may be important for making saves, as well as claiming crosses. The regression starts telling a different story when it comes to the last three attributes though. Passing doesn't seem to affect performance in any meaningful way - perhaps the success of long passes goalkeepers tend to make are determined more by the receiving player - and Stamina also has no noticeable effect on performance. It's perhaps a good time to remind ourselves that we're only dicussing effects on match performance - so even if stamina doesn't help in-game, it obviously adds some value to the player by letting him play more matches between rests.

Finally, the most interesting thing revealed by the regression is how important Age seems to be. Though several managers already buy into the importance of playing veterans, and we all know that the average lineup age affects your team's chances, the regression suggests that the age of a player is in and of itself a much bigger driver of individual performance than commonly thought. There are a couple of slightly complicated caveats to bear in mind here. Firstly, attributes were measured at the end of a season. An older player gains less over a season, so his end-of-season attribute values will be close to his average values for the season, whereas a younger player with the same end-of-season attributes may have actually had lower values (hence lower performance) earlier in the season, before catching up thanks to gains. This could cause age to seem about 2-3% more important than it actually is. Secondly, age is on a different scale to everything else: 17-40ish, rather than 30ish-99. This means that while gaining one year in age is about as important as gaining one point in perception, over a whole sample of players age will actually account for a little less of the variation in performance than you might expect, simply because ages aren't as different between players. For example, two players might have a speed difference of 6 on average, and an age difference of 3. Even though each age point is more important than each speed point, it's the difference in speed there that has the biggest overall effect (6x15% vs 3x20%). Finally, it may well be that it's experience which actually improves performance, and that age is just a good estimate of that, so be cautious about filling your team with veterans who've been sitting on the bench all their career. Overall though, age is still accounting for somewhere between 10% and 20% of performance, which is hard to ignore.


 

Defenders

Here's a summary of the defender formulae from our three sources: The original Q formula, crowdsourced intuition, and objective regression:

 

Figure 5: Important Attributes for Defenders, estimated from three sources

 

There's a little more difference between the three methods for defenders. First off, intuition agrees with the regression - Tackling is important, but overrated at 36%. Defenders tackle quite a lot during a game, but they're also required to make a lot of blocks, interceptions and clearances, which might rely a bit more on some of the other stats. Passing is also a little overrated, but still quite useful, while Speed and Perception - likely important for making interceptions, and, in the latter case, offsides - are if anything slightly underrated. On the other hand, Heading doesn't seem to be as important as we thought, perhaps because defensive headers don't need to be accurate, and aerial duels also rely on speed and perception. Stamina is also heavily overrated in terms of match performance, and the estimate from the regression is not significant (a rule of thumb for the regression values: Any values below 5% should be considered a little shaky, and anything over 10% rock solid). Notice how stamina and heading have similar patterns in Figure 3, i.e. nothing for GK, low for Def/Mid and high for attackers? It could be that stamina acts as strength in aerial duels, and is just a less important characteristic than heading for these situations. Shooting doesn't predict performance in defenders, though it's quite likely that the true value is a couple of percent or so - I have seen my defenders shoot, it's just very uncommon. Finally, Age is once again a surprisingly important factor. All the same caveats apply as discussed above for goalkeepers, but it does seem that a veteran defender should play better on average than a rookie with similar stats.


 

Midfielders

Here's a summary of the midfielder formulae from our three sources, the original Q formula, crowdsourced intuition, and objective regression:

 

Figure 6: Important Attributes for Midfielders, estimated from three sources

 

First and foremost, midfielders have to be flexible. Almost every stat seems to be useful, though interestingly the distribution from the regression is a little less balanced than the crowdsourced intuition. In fact, Passing and Speed are by some way the most important stats when it comes to performance. I'd guess this was a slight overestimate however. It is possible that midfielders with high values here tend to get put on the wings more - where they should see more of the ball, and thus get higher performance. That could probably account for a few percent of each if it's happening regularly. The same effect in reverse could be shaving a couple of percentage points off the apparent value of Stamina, Tackling and Shooting, so although they all seem to be heavily overrated, bear that in mind. 

The basic Q formula infamously neglects Perception and Heading, and the regression agrees with intuition that these are actually worth somewhere in the region of 5% and 10% respectively. Of course, many managers might pick a player with low heading and high perception to take free kicks and corners, so when you take that performance bonus into account it's possible heading is actually a couple of percent more useful than the regression implies, and perception less important. I've had a working theory for a while that the transfer market has overcompensated a little and actually overvalues perception these days, so it's interesting that the regression also pegs perception as being less crucial than most managers believe. Finally, Age contributes to performance again, though apparently not as much as for defenders or goalkeepers.


 

Attackers

Yeah, I don't blame you if you skipped forward to read this one. What the heck is going on here?

 

Figure 7: Important Attributes for Attackers, estimated from three sources

 

OK, before addressing the stat that is staring you in the face and convincing you the analysis must have gone haywire, let's quickly go over the others, because there's actually a lot going on here that's different across the three methods. First, Attack is the only position where Stamina seems to be important. Attackers make a lot of headers - just check how many crosses, corners and free-kicks happen in a game - so if stamina represents a player's strength and helps in those aerial duels, it kind of makes sense that an attacker is going to rely on it more than a defender or midfielder. There are fewer attackers on the field, so each gets more aerial duels per game, and the result probably makes a bigger difference for their performance rating since it often ends in a shot on target or a goal, rather than just a clearance. And about those Headers - more important than Shooting? Really? Before running this analysis I was pretty sure that shooting was indeed worth twice as much as heading, because in match reports you get lines like "Striker Jim shoots!" about twice as often as "A great header by Striker Jim!". But if you've read the match reports carefully, you'll have noticed that crosses end up in "shots" from strikers quite a lot, not always headers as you might expect. So what if the word "shoot" is just a bit misleading, and chances from crosses are using the heading stat regardless of the test? Then you'd see something more like the 17%/20% split given by the regression. I'm not completely convinced that's the case yet, but it would be a big deal if it turned out heading abilities were actually more important than shooting.

Talking of more important than shooting, Perception seems to be strongly underrated for strikers. This may be a recent thing - the offside trap has become quite a popular and effective tactic in later versions of the sim, and so the need for high-perception strikers to stay onside has become correspondingly greater. On the other hand, the fact that Speed shows no relationship to performance is odd. The first conclusion I'd draw is that it means speed isn't helpful for breaking the offside trap - perhaps it used to be and just got downgraded in the current version of the sim. It should be important for getting to the ball first - in rebounds, long balls, misplaced passes - but perhaps these rely more heavily on perception, and/or strikers just aren't involved in middle-of-the-park speed duels to reach long balls as often as midfielders are. Finally, it could be mildly important for receiving certain passes such as through balls, but because set pieces have become so prevalent in the sim it's ended up not being as often-used as Spinner originally intended. Passing I think is easier to explain. Exactly how many times a match do your attackers get called on to make a meaningful pass? No, kick-offs don't count. I have never really understood why passing was rated as important for my forwards, virtually every time one of them gets the ball he has a crack at goal. It just doesn't happen often enough to be worth 9% of their value.

On the other hand, they're just tackling all the time, aren't they? Okay. Tackles are the most important stat for a striker. More important than perception, shooting, age, passing and stamina. Combined.

According to the regression.

So there's an easy 'out' here, especially if you're disinclined to trust the analysis - it could just be wrong. It could be noise. I might have made a mistake somewhere. But I'm not sure it's that easy to dismiss. First off - I could definitely have made a mistake somewhere, human error happens all the time. But I've double checked and triple checked and if I had made an error, it's not really clear why it would just give me a weird result for tackles in particular. Perhaps it's noise? Maybe just a handful of strikers happen to be huge overperformers, and completely coincidentally have high tackling. Well sure, except that the analysis is waaaay too overpowered to be that susceptible to noise or outliers. Because there are almost 1000 players - around 20,000 matches - that 33% is pretty solid. It could be out by 3 or 4%, sure, and I definitely think it's more likely to be an overestimate than an underestimate, but 30% off? Not a chance. Put it this way - the regression is 99.5% certain that the value is somewhere north of 25%. It isn't noise.

So if the effect is real, what could explain it? Well, I think we need to consider the possibility that the numbers are right, and that tackling really is critical for a striker. Now, obviously attackers aren't going around tackling all the time. So perhaps tackling is standing in for something else.


   

The Grand Unified Belizio Theory of Tackling / The Witterings Of A Madman

Delete as appropriate. Do you remember reading Bulldog's help guide when you started the game? You might recall a few lines of advice from Spinner, faithfully recorded on that page. One of them provides the quotation that begins this monster of a blog post. But it's another one that caught my eye:

 

"If you see certain players trying to perform "solo-raids" over and over, know that they are a bit special, and would do well with some extra speed and tackling to assist them in their attempts."

- Spinner

 

It's commonly understood that Spinner is referring here to players with high flair, a hidden attribute. The 'solo-raids' look a little like this in the match reports:

 

Dave Defender with the ball now, trying to outsmart Mike Midfielder.
This is great football! Running past him using his speed and technique!

 

Note that the match reports often give you clues (deliberately) about what attributes or teamstats are helping a player out in a given situation. And Spinner has often stressed that it's possible to learn a lot from the match reports. Here, we're specifically told that the player uses his "speed and technique" to dribble the ball past his opponent. Speed, sure, but we don't have an attribute for technique, right? That's why I'm wondering if the Spinner quote is important - it tells us that flair players benefit particularly from Speed and Tackling. So does tackling correspond to technique, or ball control? Well... I recently ran a quick analysis on my own flair players. The success rate of their dribbling was predicted by 3 factors: The number of times they attempted to dribble per match (which I take to be a measure of their underlying "flair" attribute), their speed... and their tackling. And tackling was more strongly predictive than speed. Small sample size of course, but interesting.

And remember. ManagerLeague uses only 8 visible player attributes. Every time someone suggests increasing this, Spinner is adamant that it isn't going to change - he likes the relative simplicity of it. So imagine you're Spinner, and you're on iteration 9 of your sim, which is 3 times as long and complex as when you started it. You're trying to make your match situations as realistic as you possibly can, but without introducing new stats. What's an elegant alternative? To recycle stats, and make them represent a broader range of skills. So if you want to make controlling the ball a key role for a striker, maybe you might use the otherwise useless tackling stat to code for that...

So that's the theory. Apologies if someone's already aired it, but I have never read it on the blogs or forums here:

Tackling = Ball Control.

If that's the case, it might be able to explain why attackers benefit so much from a higher value: Quite often, when they receive a cross or a pass, the striker miscontrols the ball. If performance ratings are heavily penalised for doing so, and of course get a big boost if instead the attacker gets a shot off, and control is being tested on every attcking reception, and only highlighted when it fails... that might just be enough to explain why tackling is so strongly predictive of performance.

One extra little bit of evidence in favour. Remember back in this early blog post, where I discussed how playing in certain positions might affect what attributes you improved? We concluded that although it was mostly random, players did seem to gain more in attributes that they should be using during a match. Well, if you look back to the distribution of attribute gains by position (the second table), which attribute do you think gains fastest in attackers?

 

- Belizio

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Contest! Rate Player Attributes and WIN! (13/03/2014 03:36)

Why the heck are you blogging when you should be SLAVING OVER THE NUMBERS MACHINE??

- All the managers still waiting for their analyses


Happy new season everybody! Some thanks are in order - I'm greatly enjoying indulging my nerdy side with this blog, and it is all the nicer to see the number of views I've had (thousands!!), and especially the kind messages. ManagerLeague sure has an active and friendly community! :-) And if you donated credits for some match analyses - or just out of generosity - then bonus thanks to you! And double bonus for those of you who are waiting patiently for me to get your analysis run...

Which brings me to the question at the top of this page. Which is a fair one. Well, I will certainly have the next few days pretty full up with analysing, so this blog might not get updated for a week (even though I have loads of great ideas! Including trying to figure out teamstats, which I find deeply mysterious). Plus I have to work, and cook my wife dinner because we're all modern like that. :-)
 
So, I thought I'd leave you guys with a competition to keep you occupied in the meantime. What I want to know is, for each of the four positions:
 

Exactly how important is each attribute to performance?

 

 

Introduction

OK, so we all know how important the attibutes are to Quality - for each position it is a simple weighted sum of attributes. Specifically:

 

  

 

You can find this in all the best guides. Very simply, the quality rating for each position relies mainly on the primary attribute, and to varying degrees on four others. So increasing a defender's tackling by +11 would increase his quality by +4, but increasing his perception by +11 wouldn't increase his quality at all. Nice and simple! Except, of course, that it isn't. As anyone who's played for a while knows, during a match players can end up using lots of different attributes that don't count towards quality. Your midfielders will perceive, your attackers will tackle, and every now and then your defenders will get a chance to shoot. Plus of course, Spinner has updated (i.e. complexified) the sim since initially coming up with those simple Q formulae, and has stressed before that actions are determined by realistic sets of attributes, rather than just a simple check against, for example, tackling. Plus if you actually sit down and read a match report - really read it - you'll notice that while there are a bunch of events that seem quite clearly tied to particular attributes ("He shoots!"), others like controlling the ball are more ambiguous, or refer to skills that don't seem to correspond neatly with a single attribute.

The upshot is that (for this and a whole bunch of other reasons) you shouldn't only think about a player's quality - that's referred to as "Q blindness". Instead, you should also take into account other apparently unimportant attributes, as well as factors like performance in previous matches. Anyone who's tried to sell a midfielder with low perception will know that it's highly valued, even if it doesn't contribute directly to quality.

So what is the real "value" of an attribute for a particular player? That's what we're all going to guess, and I'm going to try and measure. The person who guesses closest to what I measure wins 5 credits or, if they prefer, a Luxury analysis for Season 112. Each position is its own separate context so there are 4 credit-or-analysis prizes available. Finally, the person who gets closest overall without being top of any single position gets a 5-credit-or-analysis prize too, so that's 25 credits or sevaral hours of my life on the table. Needless to say, the kudos for demonstrating your deep knowledge of the sim will be the greatest prize of all. Some ground rules:

 

- We're going to define this value in terms of how well a player performs now. We don't care about his potential or resale value - basically if you had to pick players for one single crucial match, which attributes would you want to be high, and which wouldn't you care about? What do you think the match sim cares about?

- We're going to ignore hidden attributes.

- Of course tactics will affect how often different attributes are used, so we're talking about the 'typical' game. Specifically the average of about 10,000 matches in a bunch of Q80 departments that I picked out semi-randomly for this analysis.

- And when I say we're looking at performance, I literally mean the performance measure the sim gives us in match reports. For better or worse, we're going to take this as a reasonable measure of how useful a player is during a game.

 

OK, so here's the contest. I've taken average season performance ratings for ~5000 players in some moderately strong departments, in a couple of different leagues. Between them these players have made over 90,000 league appearances: Hopefully this is enough data to smooth out some of the individual variation in things like hidden attributes. I've picked departments that are all within 2.5Q of Q80, and which have few bot or zombie teams in them (and of course I excluded any players from bot/zombie teams too). I've put all that data, together with each player's 8 attrbute values, into a big linear model for each position - GK, Def, Mid, Att - to find out how strongly each attribute's value predicts performance. So now I have a matrix very similar to the one above, but derived entirely by looking at the actual performance of players in games. Can you guess what it looks like?

I'll give you some clues. In some ways, it looks quite similar to the formula for quality. In other ways, it looks very, very different. All of the values are positive or zero, i.e. higher attributes are never associated with worse performance (makes sense, right?). Also, I'll tell you for free that Tk, Sh and He contribute 0 to a goalkeeper's performance, and Kp contributes zero to all outfield positions. Actually, because I gave you so much for free, I'm going to make you guess something extra: I'm including players ages as a variable. Do older players perform better than younger ones with the same attributes? And if so, how important is age compared to all the other stats?


 

How To Enter

To enter, just leave a comment on this blog with the proportion you think each attribute - and age! - contributes to performance. So if you thought that age had no effect, and the Q formula above was the true answer, you might type something like:

(Pos) (Kp) (Tk) (Pa) (Sh) (He) (Sp) (St) (Pe) (Age)

GK 36 0 9 0 0 18 9 27 0

Def 0 36 18 0 9 18 18 0

Mid 0 18 36 18 0 18 9 0 0

Att 0 0 9 36 18 18 0 18 0

 

So long as your formatting is understandable, it'll be considered a valid entry. Don't worry if percentages don't add up to 100% either, I just need the relative proportions. I'll give you all until League game 9 has been played, so that's 15:00 CET next Friday, March 21st. At that point I'll post & discuss the results of the analysis, measure how close everybody got (Euclidean distance, since you ask) and dole out some prizes :-)

Good luck!

 

- Belizio

 

 

Bonus technical details if you're interested (you don't need to know these to enter the contest):

- The analysis was a simple multiple linear regression on the 8 attributes plus age, predicting performance. A separate regression was run for each position. Certain attributes were excluded for each regression - the 3 outfield regressions did not include Kp, and the Goalkeeper regression did not include Sh, Tk or He. 

- I used average season performance but weighted each player's contribution to the model by number of appearances, so the analysis is equivalent to one run at the match level.

- All coefficients were assumed to be >=0, in other words increasing in an attribute (or age) should not reduce performance. This was achieved by removing attributes with small negative coefficients from the regression, and re-running with the reduced set - continuing as necessary until all coefficients were positive. In total 24 of 36 possible attributes had a positive effect on performance (a bonus clue for reading this far!).

- Because age is on a different scale (17-40ish) than attributes (20ish-99) I z-transformed everything (attributes/age/performance) before running the regression. That just means everything is on the same, intuitive, scale: So for example, if age is more important than shooting for some position, then being in the top 10% for age but average for shooting should give you better performance than being average for age and in the top 10% for shooting.

 

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Match Analytics For Your Team! (10/03/2014 00:24)

 

Win a free Analysis with the latest Free Wonderstar Cup!

- If you win a Wonderstar Player Cup, then as well as the trophy, cash prize and 1 credit, you will also get a free analysis for your team :-) You must be no higher than Q74 to enter as this is aimed at newer teams. Congratulations to Team Uruguay and FC Blatter, who have both won a cup and are eligible for a free analysis! The next Wonderstar Cup is now available, for Season 117 R30. Good luck!

  


 

Wonderstar Analytics offer personalised Match Analytics for your own team!

 

These are detailed statistical breakdowns of your team's performance in different areas; similar to the Match Analytics ideas I've been introducing in the previous posts (plus some other stats). These stats are drawn from your match reports, and can help you:

 

- Identify strengths and weaknesses in your team
- Develop new strategies that take advantage of your team's strengths
- Guide player recruitment
- Assess and improve your substitutions and strategy throughout the match
- See what strategies worked best for you this season, and how you won or lost individual games
- Guide your training; which teamstats are going to most efficiently boost your performance?
 

Plus I just think they're enjoyable to have, and make your team feel more three-dimensional. :-)

Full details below, in the FAQ, but here's a quick summary. For 10 credits you'll get an analysis on a set of your team's matches. Note that each analysis can include as many matches as you like from one season - for example one analysis could be all your competitive games in a season, and another could be just your league matches. The number of these will be quite limited each season, because they take some time to run, so they will be provided on a first come first served basis. Because they cost some credits, they won't be available to you unless you've bought credits from the game (because you won't be able to donate). Sorry but I don't have plans to change this just now, or allow friendlies as payment: I think of it as providing some extra value if you support the game! There are two ways to get an analysis without buying credits: Win a contest I run on my blog, or win one of the Wonderstar Cups.

I'll only be running 15-30 of these per season, which means they're quite limited. Buying an analysis once (or winning one) means all your future analyses will be just 5 credits.

Since launching at the end of Season 111, 36 teams have received analyses. These included 9 Champion's League and 4 SuperCup teams in Season 113, and this will go up to 9 and in Season 114: In fact, 14-out-of-16 top-division sides receiving analyses qualified for the International contests in Season 113! So far analysed teams have won 10 league titles and 9 league cups since getting their analysis (and have plenty of runner-up medals too), so congratulations to them! Season 114 is now open so if you want to join them get on it :-)

If you're interested, please send me a message in-game.

Details are covered in the FAQ below, in the meantime here are a few snapshots of the sort of thing you can expect to get (subject to change, which probably means improvement, and you also get a bunch of raw numbers to examine yourself):

 

 

 


 

FAQ

 

1. How much does it cost?

The service costs 10 credits for your first analysis, then 5 credits per analysis after that. Remember each analysis is run on multiple matches (as many as you like).

 

2. Which games can I include? Can I mix and match competitions?

Absolutely, it's entirely up to you which games you include. You can keep different types of games apart if you want to analyse them separately - e.g. SuperCup games in one analysis, and league games in another. Or, you could divide up your matches according to something else, like tactics - to see the effect of playing a 4-5-1 versus a 4-3-3 for example, or how your tactics affect the types of possession you get - or perhaps look at all the games a particular player took your free-kicks, versus those he didn't. Whatever is going to be interesting for you, or help you optimise your team. For example, I put all of my competitive games into the same analysis, and ran the analysis over multiple seasons to track my progress. That's a simple thing to do, but you can be as clever or inventive as you like here!

 

3. Can I analyse a different team?

No - you can only analyse your own side (you'll also see your opponents' data, but they won't be separated by team). I might make this available in the future if there is demand for it.

 

4. I can't make credit donations! Will you accept cash / food / friendly matches?

No, I'm afraid not! It's a deliberate choice to make payment only through credits. Buying credits means you're supporting ManagerLeague, but there's not a whole lot to do with them apart from playing friendlies (or avoiding taxes through slightly-shady credit deals for players...). Making these analyses available only through credit donations enriches the game for people who choose to pay for it, and provides an extra reason for supporting the game which I think is important. I don't have plans to change this just now, or allow friendlies as payment: I think of it as providing some extra value if you choose to support the game! The only exceptions are the occasional Wonderstar Cup, or contests I run on this blog from time to time, which are open to everyone.

 

5. But you have a donation limit, what happens when you hit that?

Then my analyses for the season are done :-) I can't afford to spend too many hours running these - I have a job! - so the donation cap gives me a nice limit of around 20-30 teams per season. This also means that if you get an analysis done, you know you have a competitive edge, since only a handful of other sides will have the kind of data on their team that you do. Generally speaking this will work on a first-come-first-served basis.

 

6. How many datasets can I get analysed?

You get one dataset per analysis, and each dataset can contain as many matches as you like. You can even run analyses on games from previous seasons if I have it available, which I will do if you've used the service before. So, for example, you could run an analysis on your matches partway through a season to see how you could improve, then run another at the end which includes all your games.

 

7. What exactly do I get?

You'll get a 7-page pdf document with figures and statistics about your team. These include: Key stats, Match Progession Graphs, Possession Analytics - which includes Conversion Rate Breakdowns for different match situations - and some additional stats to help you guide your teamstat training, like offside-trap and set-piece success. I have also added some more detail about how your conversion rates look broken down further (shooting accuracy, chance creation, save percentage) and a huge spreadsheet giving you details about every indivdual match you played - from starting tactics for both sides, match outcomes like performance, chances or shots on target, and a detailed breakdown of possession! As well as all this, you'll also get a whole bunch of raw numbers used to generate the figure, in a csv file or excel spreadheet, to do whatever you want with.

 

8. Will you tell me exactly what it all means?

I can provide some basic pointers, but I might be way off base! And anyway, figuring out your best strategy is the fun part, right? Some things are pretty easy to interpret anyway: If your offside trap is working very well, you might want to train up your free-kick stat more to take advantage. If you're not getting many corners, but you are converting them into goals at a high rate, you could try shooting from further out to increase the number you win during a game. Or if you see that you're converting more chances into goals when playing long passes out of defence than when you play short ones, maybe you should switch from continental to long-ball. There's so many other things you could use to fine-tune your team, so this is really where your own skill and imagination can give you an edge. I can't tell you exactly how to improve simply because I don't have access to the inner workings of the sim - but I can provide you with the data to figure things out yourself, which I think is more interesting.

All that said, I do enjoy talking tactics. So if you have any questions about your stats feel free to message me about them!

 

9. How accurate is the data?

Pretty accurate; I have spent some time fine-tuning the analyser and it's been based on over 50,000 lines of action from various matches. But of course there's always the chance your game will have featured some very odd event, so I'd put the accuracy at 98-99% overall. Certainly good enough to draw conclusions from. One thing that could affect the accuracy of your reports is if you have renamed players during the season, or named them something that appears elsewhere in the match reports (e.g. you call someone "off-side", or they share a name with an opponent's player). If that's the case be sure to let me know, it should usually be straightforward to adjust for this when I run the analysis, so long as I know all the names your players have had.

 

10. I'm sold! What do I need to do?

You'll need an email address I can send the pdf and data to. Assuming I've still got spots available for the current season, then you can donate the credits and I'll run the analysis for you. If I've hit my donation limit, or simply don't have the time to run any more analyses in the current season, I'll let you know before you try to donate, and give you priority for the next season. Finally, if your matches include some players who you've renamed or sacked, you'll need to give me their names (so that I can identify their actions as being for your team, not your opponents).  

 

 

- Belizio


  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Match Analytics Part 2: Possession Types (08/03/2014 04:16)

 "It's not the size that counts, it's what you do with it."

- Sniggering teenage boys everywhere

 

Long post today, with lots of graphs, so grab a coffee! In the previous two posts, we took a slightly more sophisticated look at match performance by separating possession from conversion rate. In doing so, we discovered that more possession doesn't always mean more goals in the ManagerLeague sim - switching tactics this season gave me reduced possession but more goals, and decreasing fitness during a match seems to reduce your ability to score but not the number of possessions you get. So, clearly, all possessions are not created equal.

One thing my match analyser does is extract the number of possessions each team gets, as well as the number and time of goals they score. But I also have it set it up to identify the types of possession each passage of play begins with, and whether or not that play ended in a goal. Let me show you what I mean:

Figure 1: Passages of play sorted by team and possession type for matches involving The Wonderstars, Season 109 (67 matches).
Numbers indicate number of possessions over the whole season, with goals resulting from those possessions in parentheses.
 
 

OK, so let's run through what I'm showing you in this graphic. Each time a passage of play begins, one team has the ball, and it will normally be in one of six possible situations:

1. A midfielder in possession in an advanced position (Attack). The text might say something like: "Charles van der Elst is in a good position to cross it now."

2. A player taking on and trying to run past the opposition (Dribble). The text might say: "Jurgen Hallaert with the ball now, trying to get past Jerome Vargas."

3. A short, simple pass from a defender to another specific player (Short) as the team build up patiently from the back. For example, "Ilie Cristea passes the ball to Mazin al-Hindi".

4. A longer pass, hit by the goalkeeper or a defender towards the midfield (Long): "Mauro Scifo passes the ball towards the midfield".

5. A Throw-in. This is pretty self-explanatory.

6. Ditto for Corners.

In addition to these normal possessions, each half begin with a Kick-off. Finally, teams can sometimes win possession midway through a passage of play, either by winning a free-kick or a throw, or by intercepting and Counter-attacking when opponents hit a stray pass or a weak shot. What the graphic above show you is the number of each possession type won by the Wonderstars (large green bars) and their opponents (large red bars) over the whole of Season 109. In the smaller bars we can see how many of these possessions resulted in goals for each team - they're designed so the area of each bar is proportional to the number of possessions or goals, and I've also added the actual numbers next to each bar. So what can we make of this? Well first off, if you read my Introduction to Match Analytics you might remember that Season 109 was a department-winning year for The Wonderstars, and they had 55% of the possession overall - so it's no surprise that you see generally bigger green possession bars than red ones. The distribution is clearly a little different too - The Wonderstars had 55% possession, but definitely more than 55% of the dribbles and short passes. On the other hand they had fewer long balls, counter-attacks and kick-offs than their opponents, while corners and throw-ins were roughly similar for both teams. A pie chart helps us see how reliant each team was on different types of possession:

Figure 2: Reliance on each possession type for The Wonderstars and their opponents, Season 109 (67 matches).
 

Well, kick-offs are easy to explain: Both sides will get equal numbers except if one team concedes more goals. So that difference is just due to The Wonderstars conceding fewer goals and thus having to kick off less. Counter-attacks is probably because The Wonderstars didn't often play with counter-attacks or offside traps, whereas some of their opponents did (resulting in more quick breaks and free kicks for offsides). The long ball / short ball distribution also makes sense because The Wonderstars like to think of themselves as a skillful passing side (it's actually because they have low speed, heading and perception but don't tell them that), and play continental style most of the time. Dribbles is quite interesting, but we definitely have a high-flair player or two in our ranks, so it seems that the sim gives them a chance to do their thing more often than players with lower flair.

Now let's take a similar look at Season 110, in which The Wonderstars got pretty much battered in a higher division, but still managed roughly 50% possession:

.

 Figure 3: Passages of play sorted by team and possession type for matches involving The Wonderstars, Season 110 (66 matches).
Numbers indicate number of possessions over the whole season, with goals resulting from those possessions in parentheses.
 
 

As possession was roughly equal, the larger bars show relatively little difference between the two sides. There's still some differences in how possession types are distributed, which are similar to the previous season: The Wonderstars were still building up from the back with short passes much more often than they were playing the long ball, and they also dribbled more frequently but were less likely to turn their opponents over and counter-attack. The problem for the Wonderstars was that they just didn't convert as many of those possessions into goals. In particular, you can see that despite similar numbers of possessions, The Wonderstars scored far less frequently from corners or attacking positions than their opponents, and in almost 500 possessions starting in defence - Short & Long in Figure 3 - managed a pitiful 4 goals. So let's take a look at this in more detail.


 

Strengths and Weaknesses

For every possession type, I calculated the conversion rate - so to get The Wonderstars' conversion rate for kick-offs for example, I divided the number of goals (5) by the number of kick-offs (170) the Wonderstars got. But I didn't just look at starting possessions - for this analysis I looked at every time The Wonderstars (or their opponents) got into one of those match situations. So if they started with the ball in defence, played it out continental-style to the wings, crossed it in and won a corner, that passage of play includes Short passing, Attack and Corner situations. This way, we can get a measure of how frequently overall the teams scored - whether directly or later in the passage of play - from different match situations. We can ask questions like, "If during a passage of play The Wonderstars were dribbling at their opponents, how often did that ultimately result in a goal?", and answer them with a percentage - basically a conversion rate, but only looking at chances involving that specific match situation. Note that all of these match situations therefore include more chances than just the starting possessions shown in Figures 1-3, with the exception of kick-offs which always occur at the beginning of a passage of play. It also means that the same passage of play - indeed the same goal - can get included in several different categories. Finally, we can also add free-kicks into the mix - although you never get given a free-kick to begin a passage of play, they're quite commonly earned halfway through by drawing a foul, or catching your opponent offside.

OK, here we go. Let's go back to that easy promotion year in Season 109:

Figure 4: Chance of scoring from different match situations for The Wonderstars and their opponents, Season 109 (67 matches).

 

Now we can see really clearly why The Wonderstars scored so many goals that season, despite 'only' 55% possession. Recall that overall conversion rates tended to be around 4-7%. Corners and free-kicks ended up in goals at a much higher rate though, a huge 11% of the time for The Wonderstars. The flair players in the team made a big difference too, with over 9% of their efforts resulting in goals, and crosses or passes from midfield were also converted pretty frequently into goals. Interestingly, even though The Wonderstars counter-attacked less frequently than their opponents, they were much more dangerous when they did so - perhaps because several of those counter-attacks would have been defensive free-kicks resulting from offsides or wrestling at corners. On the other hand, The Wonderstars didn't really score any more often than their opponents from defensive positions, regardless of whether they opted for Short or Long passing. Their advantage came from pretty specific situations, mostly set pieces and dribbles.

Now lets take a look at what happened when they moved up a division in the following season (110). Remember, possession was pretty even. Conversion was emphatically not:

 

Figure 5: Chance of scoring from different match situations for The Wonderstars and their opponents, Season 110 (66 matches).

 

Big difference! First the positives - The Wonderstars were just as dangerous as their higher-quality opponents from counter-attacks (though they had fewer), corners, throw-ins and kick-offs. The negatives though were a poorer conversion rate from midfield possessions (Attacks), and a catstrophic rate of return from defensive positions (Short / Long). Worst of all, two reliable sources of goals from the previous season - dribbles and free-kicks - were almost totally stifled by better defenders. But nonetheless, they survived. As discussed earlier, The Wonderstars brought in a striker and shifted to a 4-4-2 attacking style for this season, with the following results:

 

 Figure 6: Passages of play sorted by team and possession type for matches involving The Wonderstars, Season 111 (64 matches).
Numbers indicate number of possessions over the whole season, with goals resulting from those possessions in parentheses.
 
 

I've been getting less possession this season - note for example that my opponents started almost 100 times more often than me with the ball in midfield, and about 200 times more often in defence. I've outscored my opponents this season, though, so where is the difference coming from?

Figure 7: Chance of scoring from different match situations for The Wonderstars and their opponents, Season 111 (64 matches).

 

Generally speaking, there's an improvement across the board. But a few things stick out. I'm turning a respectable number of defensive possessions into goals now, and limiting my opponents much more when they get the ball in defence. But they also scored a bunch of times from kick-offs and throw-ins too, so overall we're pretty evenly matched when the ball starts in our own half - that's not what's deciding games most of the time. I was pretty vulnerable to the counter attack this year, conceding eleven goals that way (out of 83 total), and got fewer goals that way myself. So playing 4-4-2 and attacking more certainly has its drawbacks. But I made up for that with a big improvement in both my own attacks from midfield and in defending those of my opponent (perhaps improving attacking and defending teamstats might have helped here, as well as having that extra striker to aim for when passing or crossing). Plus my defending at corners is improving, and while I'm still vulnerable at free-kicks the difference is not as vast as it was last season. Teamstats ahoy! The really noticeable difference though is how much more effective my dribbling has been. My most frequent dribblers are defenders, with low shooting, so when they break through the defence they need to lay the ball off to a striker for me to score - if they have a shot themselves the result is usually pretty ugly. Having two strikers and an attacking pressure seems to mean that my defenders get support more often, and are able to make that key pass for the striker to finish the move.

And in the offseason I'll get to work on those free kicks...

Again, I hope this was enlightening or interesting for some of you, and feel free to ask questions or drop me some feedback in the comments below!

 

- Belizio

 

 

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Match Analytics Part 1: Stamina & Substitutions (07/03/2014 20:43)

"The match is not finished before the ref has blown his whistle. Remember what Solskjaer did!"

- Chairman

 

Before I get into some detail with Possession Analytics, here's another interesting thing you can do with very simple match statistics. I imagine most people reading this blog will be relatively experienced and will already know this, but for anyone new to the game:

 

SUBSTITUTIONS ARE CRITICAL!

 

The reason for this, again, is that ManagerLeague simulates the match event-by-event, rather than all in one go. Throughout the game your players randomly lose fitness (perhaps players with lower stamina or in certain positions lose more, but I don't actually know), and events towards the end of the match will be affected more and more by the relative fitness of the players left on the field: your star striker will be more likely to scuff his 85th-minute shot wide if he's been on the pitch the whole game. The best way to mitigate this is to be sure to use all three of your substitutions every match. I set a simple event in the 60th minute to just sub out my least fit defender, midfielder and attacker, regardless of their quality or how well they're performing, and replace them with some fresh legs. Here's the result:

 

Figure 1: Proportion of goals scored by The Wonderstars by match stage. Data from 191 matches, Seasons 109-111.

 

If you read the previous post, you'll know that I've been looking at the number of goals I scored relative to my opponents over the past three seasons. Here I've looked at all three seasons worth of data and recorded the time each goal was scored. That means I know how many goals I scored and conceded at different stages during a match, which is what I've plotted above in 15-minute increments (the exception is extra time, which includes the full 30mins but not penalty shootouts). Values above 50% mean I scored more than half of the goals during that stage (i.e., I scored more than I conceded).

We can see a couple of things straightaway. First off, my team seems to come quicker out the blocks than my opponents, scoring almost 60% of the early goals in a game, but then I'm pretty evenly matched with my opponents until around 60 minutes. At that point, I get increasingly dominant as the rest of the second half wears on, and if we end up in extra time I score nearly 2 goals for every one I concede. My guess is that the middle portion of the match is my "natural" level of scoring, but that two factors give me an advantage in the first 15mins and the last 30mins.

Let's take the first 15mins - it's too early for fitness to really be much of a factor, so what's changing around the 15th minute? Well, not my tactics. I pick a formation, style and pressure that I think is going to work for me and pretty much stick with it (see the previous post). So perhaps my opponents tactics are changing? Well, if you read the popular guides, several suggest a strategy of changing your tactics very early in the game. The reasoning behind this is that match summaries - which is what most people glance at if they're scouting an opponent - only show you the starting formation, style and pressure for each team, and you have to read the match report itself to see if the tactics change during the match. So many teams deliberately begin the match with tactics they don't really intend to use, just so it shows up on the match report and fools their future opponents. Then after a minute or 15 they set an event to switch to their "real" tactics. If you believe that particular tactics are inherently advantageous against other ones - things like "4-5-1 continental beats 4-3-3 mixed" - then this strategy should give you an advantage, as your opponent might set up to counter your "fake" strategy instead of your real one. I don't believe that by the way - though my mind can always be changed with data! - so I just adopt my strategy to fit my own team and pay relatively little attention to my opponents. But it's possible that some of my opponents are playing a weaker fake strategy or out-of-position players in the early part of the game (even if it's just for the first minute, remember minute 1 always has a passage of play from the kick-off), and that's what's giving me a relative advantage. I could be completely wrong about this and it might all be luck, but it's an interesting possibility to ponder, especially if you use that "fake tactics" strategy yourself.

The late-game advantage, on the other hand, begins from 60mins and gets stronger as the game wears on, and I don't think it's a coincidence that 60 minutes is when I make my substitutions. Although many of my opponents make substitutions of their own, some don't - or substitute players based on quality or performance instead - and so the fact that I always do means that on average, I tend to keep a greater proportion of my fitness in the later stages of a game. It's interesting to note, though, that possession doesn't seem to be affected:

 

  Figure 2: The Wonderstars possession by match stage. Data from 191 matches, Seasons 109-111.

 

So if my possession isn't changing, but my goal percentage has improved, then I must be getting a better conversion rate than my opponents, right? Correct! You've been reading this blog carefully. In the figure below, I've plotted both my conversion rate (green) and that of my opponents (red) separately. This lets us see what's going on in general with conversion rates, as well as how I compare to my opponents:

   Figure 3: Conversion rates by match stage. Data from 191 matches, Seasons 109-111.

 

Just as we suspected, the advantage I have over my opponents in the first 15mins and last 30mins seems to come from a superior conversion rate (the green line moves above the red line), instead of any change in possession. That suggests that fitness mostly affects individual player performance, and not the amount of possession (which instead comes from formations, quality and teamstats).

Notice too the overall trend for conversion rate, which could reveal a little about how the sim is working. First off, the weird blip in the first 15 minutes for my opponents may be a result of switching out of a fake formation. If they do so without subbing players, starting the game with a fake formation means at least one player will be out of position - so for example two attackers might be on the right side of midfield until the formation switches from 4-5-1 to 4-3-3. Being out of position tends to lead to bad player performance (yes yes, I know your midfielder has a higher shooting stat than your second striker, but resist the urge), which could cause the drop in conversion rate - one of those strikers goes to cross the ball and hoofs it right of play.

So, if you ignore that blip, you see pretty constant conversion rates for the first 30 minutes, but a dip as the first half wears on. Then for the first 15 minutes of the second half, things improve again. And after 60 minutes, conversion rates for my opponents start decreasing - and fall all the way down to a drastically-bad 3% by extra time. So it looks like fitness (hence player performance on things like passing and shooting, hence conversion rates) decreases throughout the match, with a little bump back up after half time thanks to the restorative power of orange slices. But because I always bring my least fit players off at 60 minutes and replace them with fresh legs, that is sufficient to push my conversion rate back up over 6% for most of the second half (before slipping back down again as we get into extra time). By keeping my conversion rate intact while my opponents' declines, I end up scoring more goals over that phase of the match than my team quality would cause you to expect - even though the quality of my lineup has decreased after bringing on three lesser players.

So always use your substitutes!

Finally, you could even go so far as to speculate that the least-fit-players-out-at-60-minutes approach is the optimal substitution strategy. Since players get a bit of a boost at half time anyway, it doesn't seem so advantageous to make the substitution after 45 minutes because you're not improving your team's average fitness by enough (note my opponents seem to gain more fitness after 45 minutes than I do, perhaps because some of them make substitutions at half time and I don't). On the other hand, if you leave it too late, you're only getting 15 minutes worth of fresh legs. And bringing off your least fit players maximises the effect; while it's tempting to leave your best performers on the field, if events are independent there's no guarantee they'll continue to perform strongly as their legs get heavy.

But anyway! Always use your substitutes! 

 

- Belizio

 

 

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Introducing Match Analytics (06/03/2014 04:45)

 "The simulator... is the single most important piece of the ML-puzzle."

- Thales

 

Gaining attributes on your players sure is addictive, but I think we've covered that enough for a while. Let's switch our attention to what I think is ManagerLeague's biggest strength - the match engine. I'm going to introduce a fairly simple analysis, and describe how it's successfully guided my transition into a higher division. You should probably think of this less as a "Chronicles of Team X" type post, and more of an illustration of how you can use even quite broad statistics to adjust and improve your strategy. 

The first thing to remember is that for a game this complex, ManagerLeague gives us surprisingly few statistics. We get to see goals, assists and appearances for league games (plus some cup goals), alongside a simple average of player rating in league games. This doesn't really do more than scratch the surface when it comes to assessing how well our team is performing, or why. Quality Blindess (judging a player or team by the Q-value and ignoring performance, teamstats or other attributes) is pretty well-known, and avoided by most experienced managers, but that doesn't mean looking at performance ratings tells you the full story either. The more detail you look at, the better your competitive advantage. 

A crucial step towards becoming a better manager in this game, then, is being able to figure out how your team is performing from match reports. That's because the match engine works from the bottom-up. What I mean by that is that results emerge in a relatively realistic manner from dozens or hundreds of events during a game, and specific attributes on individual players (combined with your tactics, teamstats, opponents, luck and probably a few other factors) determine whether those events go in your favour and lead to goals or not. Reverse-engineering all that data and working out how to improve your side is a pretty open-ended challenge, but an interesting one I think. Thales' Blog is a really interesting and enlightening place to start if you want to peek a little bit into the mechanics of the game. I'd treat most of it as conjecture rather than stone-tablet-fact, but I feel like I came out of reading it with a better understanding of how to read match reports.

The problem is, even if you learn to decipher from the match reports what's really going on under the surface, who the heck has time to read them all?


 

Possession vs Conversion

To solve this - and give myself a competitive edge - I've invested some time "up-front" automating the extraction of some key statistics from my match reports. This allows me to get an overview of how my team is performing in much more detail, without having to read through each report and try to spot trends over the course of a season.

As an example, in ManagerLeague a match is composed of passages of play, which each begin with one side in possession of the ball. Your possession - the proportion of times you begin with the ball - is a crucial factor in determining how many goals you score or concede, but this statistic isn't summarised for you (you'd have to go through each match report and count the number of times you started with the ball versus the times your opponent did). So, I wrote some script to extract this possession statistic - ignoring kickoffs, i.e. only 'earned' possession given to you by the simulator. I also pulled out the number of goals I scored compared to my opponents, and calculated a "conversion rate", which is the percentage of posessions resulting in a goal. So, Goals = Possessions x Conversion rate. Teasing those two factors apart should give me a better idea of where to strengthen than I'd get just by looking at mean performance or goal difference, since those simple statistics only tell me how I'm performing, not why.

So, averaging these across all my competitive games (including Player Cups) for season 109 gives me a nice visualisation of how my team compared against their opponents: 

 
 
Key Statistics for The Wonderstars (67 matches), USA Division 4; Season 109

 

To recap, that's my average possession (proportion of plays I began with the ball), my conversion rate, and the resulting number of goals I scored represented by the green bars. My oponents numbers are in the red bars. This gives me a little more information than just my league goals or average league performance - plus I can include whichever games I like, not just league matches. Overall I can see at a glance that during Season 109 I had around 55% possession, but also made the most of it - scoring nearly twice as many goals as my opponents in the 67 competitive matches I played that season. It's not too much of a surprise that the Wonderstars outscored other teams that season, as I was busy getting promoted from a bot-dominated division, so my toughest matches were mainly in the league cup and player cups. What's interesting to me is how much the advantage came from the conversion rate, rather than the amount of possession.

Now let's plug in my match reports from the following season, after I got promoted up to a considerably harder division:

Key Statistics for The Wonderstars (66 matches), USA Division 3; Season 110

 

For season 110 I figured to survive I needed some experience, so I invested my transfer kitty in a veteran midfielder and an upgrade on my goalkeeper. The extra midfielder also allowed me to switch to a defensive 4-5-1 formation to keep things tighter against my superior opponents. I was pretty heavily outscored in the league, finishing with 40pts rather than the 78pts from the previous season, but I narrowly avoided the relegation playoff spots and outperformed my team quality so I called the season a success.

What's interesting though is that I had just as much possession as my opponents did that season (perhaps even a few more once you account for the fact that I was getting the ball back every time I picked it out of my own net...). Presumably the increased experience in my squad, and especially in midfield, helped me here, but I'm guessing the biggest factor was my use of five midfielders in tougher league games. Where I was overmatched was in the conversion of these opportunities into goals. The previous season I converted 6.7% of my own possessions and conceded on only 4.5% of my opponents', but this pattern pretty much reversed in the tougher league. The new keeper obviously wasn't enough on his own to stop my opponents' higher calibre attackers from scoring, while my top scorer from Season 109 struggled to adapt as a lone striker and scored only a handful of goals.

There were two big lessons I drew from this. First, despite having a lower team quality than most of my opponents, it didn't seem like I'd get too much benefit from continuing to play defensively - I get my fair share of the ball, so a tighter game would probably just lose me some points by increasing my draws at the expense of both wins and defeats. Notice, too, that my conversion rate decreased by more (-2.4%) than my opponents' rate increased (+1.6%). So while my new keeper hasn't turned out to be a world-beater, it made sense to persevere with him - and let my young defense continue to develop - and invest in my forward line instead. So, midway through the season I brought in a veteran striker, and though he fizzled a bit in that first year he postponed his retirement to come back for an extra season (I like to think he wanted to prove he wasn't just here for one last payday). My younger strikers had gained in quality and experience, so I also brought in a couple of young midfield prospects to boost my squad depth a bit, and returned from the season break with the same team, but a 4-4-2 formation and a more attacking mindset: 

Key Statistics for The Wonderstars (58 matches), USA Division 3; Season 111

 

The results this season have completely turned around: With three games to go I'm 17pts clear of the relegation playoff spots, and not yet out of the running for an unlikely promotion playoff. I'm conceding 0.2 fewer goals a game now, but the big difference is that I'm scoring 0.6 a game more than last season - I'm 3rd in the department for goals scored, and my veteran striker has finally justified his wages with 33 goals from 51 appearances in all competitions. So why such a dramatic improvement with the same group of players? Is it just a generic quality improvement? Well, dividing the data up like this shows us pretty clearly that it isn't. First off, my quality remains a couple of points below the department average, ranking me 11th. But I'm 17 points and +38 goal difference clear of 11th place this season. The key stats show a completely different story to last season - instead of retaining the ball, I'm now only getting 45% possession, which can probably be put down to the shift from 4-5-1 to 4-4-2. I'm more than making up for that drop though with a huge increase in my conversion rate (+2.7%, i.e. 60% greater chance of scoring when I start with the ball), as well as a decent improvement in my ability to repel attacks (-1.7% opponent conversion rate). I've had a mild increase in quality this season relative to my opponents, especially in my youthful defence, but that's not where most of my improvement has come from - it's the change in system that's made me far more efficient. Bottom line:

 

I was able to improve my strategy using possession and conversion statistics from match reports.

 

All that comes from a pretty simple analysis (possession and conversion rates). I hope you found it interesting, and I'd love to hear your thoughts in the comments! In the next post I'll be delving a little deeper by looking in more detail at what I'm doing with that possession.

 

- Belizio

 

 

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Gaining Part 5: Superstitions & Chance (03/03/2014 23:44)

"The root of all superstition is that men observe when a thing hits, but not when it misses.” 

- Francis Bacon
 

Just a quick post now, after the long one earlier today about how lineup age might affect player gains. I'm going to wrap up my look at player gaining by addressing the remaining two mantras on that list I posted when I started this blog. Not because they're especially convincing on their own, but because they're illustrative.

We've all been in the situation where we're playing a few friendlies and just not getting anything. We reload ml-reports and that big fat 0 is staring accusingly at us. Fine, it happens every now and again, but if you just played three friendlies in a row and got nothing for any of them?! Or you got two goalkeeping stats and nothing else from the last five games? Surely you must be doing something wrong, right? Time to change the formation, throw some different players in there quick, before you use up all your friendlies for nothing! As several guides I've read have suggested, perhaps you should even take a break and come back to the friendlies in fifteen minutes, or an hour.

Well, we can check this pretty straightforwardly. If this is true - that the way your team is setup is stopping you gain, or you just get onto a hot streak and start piling up the attributes - then the gains I earned on a match should predict the gains I earned on the next - because I keep the same formation and just rotate players out one by one as they lose fitness, and I often play the same sides consecutively because I'm playing sets. So, I checked the correlation between each match I played in my database and the one that came straight after. The sum effect of all those similar factors? Playing often against the same teams, with pretty much the same formation and players? Nothing. The R-squared is 0.0024. We can predict less than one quarter of one percent of the gains in that next match - the other 99.8% is effectively just luck. So just because you get no gains this match, doesn't mean you'll get nothing in the next one. Or that you'll get loads! You've got pretty much the same chance as you always do.

To understand why sometimes it seems that can't be the case, it's worth very quickly considering how statistics work. Gaining attributes is a random event with two properties: We know roughly how often we should gain an attribute, on average (e.g. we might expect a player to gain an attribute 10-15% of the time he plays a friendly) but we can't predict exactly when it's going to happen. When you have several players on your team this means you could get no gains in a match, or one, or two or several. Such a process is well described by a Poisson distribution. If you know the average rate at which events occur - e.g. 1.4 gains per friendly - then you can easily calculate the probability of getting zero gains, or of getting one, or 2... etc.

So even if you average 1.4 attributes per game, that doesn't mean you're always getting 1s and 2s. You should get +0 on about a quarter of your games, +1 on around a third, +2 on another quarter and really good gains of 3 or more on the rest.

Now, even though your average gaining rate is pretty good, and netting you close to 300 attributes a year from friendlies, you'll still be getting the dreaded +0 for fifty games a season! And you'll be getting two blanks in a row about 12 times on average. And three blanks a row should happen 3 or four times a season too. In fact, you're more likely than not to get four friendlies in a row with no gains at all at some point during the season. And as awful as it seems at the time, that doesn't mean you're actually gaining badly or doing anything wrong - you'll just get a +5 a couple of times, or a bunch of +4s to balance things out. This is just the way chance works, even when there's no difference between the games you get 0 for, and the ones you get +5 for. The problem is, you'll remember that run of zeros because it seems so improbable and meaningful - and you'll forget all the other times a zero was followed by some gains, because that doesn't seems so surprising. That's what we all do, it's just the way our brains are wired - to spot patterns. We're not so good at figuring out how unexpected and meaningful that pattern really is though, which is why a basic understanding of statistics - not a bunch of equations, just an appreciation for how often totally random events can seem meaningful - is really helpful to have, whatever your job. 

So don't put too much stock in how much you gain on a handful of games. Get nothing from that friendly set with team X, but +6 against team Y? Get no gains on your prize midfielder for ten matches? It's very tempting to conclude that playing against team Y is better for your players, or that you should avoid playing against team X or similar sides, or that you should swap your youth's position with that old player who gained +3 - but you just can't tell that from so little data. Those are the sorts of things you'll see from time to time, even when those two teams are identical, and your youth is actually a great gainer. Instead, you should look at hundreds of games, or hundreds of players - that will start to give you an idea of what really affects your gaining rate, and how much. Or if you don't have time to do that, just check my results!

Oh, and the pumpkin thing isn't true either.

 

- Belizio

 

 

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Gaining Part 4: Age, Gaining & Data (02/03/2014 04:45)

 "Surrounding young players with older ones probably helps their gains - but perhaps not that much."

- Belizio

 

I'm going to take a diversion from working my way through the list of gaining mantras to go back to the question of how much older players help your younger ones gain (I know you're all desperate to know whether the pumpkin thing is real, don't worry, I haven't forgotten). In a previous post, I wondered whether huge gains on young players were strongly influenced by the age of their teammates, or whether it was mostly just the fact that these young players were getting in close to 200 friendlies a year. I looked at the gaining rate of my own (medium-star, surrounded by all ages) young players, calculated roughly what I could expect from them if I simply made sure they played in as many games as possible, and concluded it was high enough that it probably accounted for most of the gains from "youth farm" teams.

That was based at least partly on a gut sense of what a "good" season's gain should look like, so today I'd like to run a quick analysis with a bit more data. As a jumping-off point, here's a nice blog post from (successful Dutch league-and-cup-winning manager) Getzome with a bunch of good advice for those of you who are relatively new to the game. Read it if you haven't already! Midway through the blog though, when discussing player gaining, Getzome stresses one single rule for improving your player gains:

"The key to success lies in just one thing: *drumroll drumroll*     -     Old Players."

- Getzome

 

I think this is exaggerated. I think success lies in a very wide range of factors, just one of which is the surrounding players' ages, and that this factor has less of an effect on gaining than reading most of the guides on the blogs would have you believe. In short, I think it's more possible to balance a competitive squad with decent player gains than is widely assumed.

Now just to be clear - my aim is not to suggest that some people are "wrong" for playing the game a certain way, or to pick on any particular guide (I'm just using Getzome's as a jumping-off point because he provides some nice data we can look at, and makes the typical argument that you read in most ManagerLeague guides). Nor am I arguing that age and experience have no effect on gaining. Instead, I see most guides - and therefore a majority of players - adapting identical strategies to build up their teams or cash, and I think this is at least partly because there's a prevailing wisdom that you "have to" do things a certain way if you're going to be competitive. So the aim here is not to tell you that you can't use the 3-young-players-surrounded-by-38-year-olds strategy if you want, but to point out that maybe you don't have to, and that there are legitimate alternative strategies available.


 

Comparing Strategies

To directly compare the effects of different age strategies, I've looked at my own gains for this season, along with Getzome's team (Baronie) and the other nineteen sides making up the current top-20 gainers in the team rankings. That gives us a range of teams who a) are all playing plenty of friendlies and gaining well, and b) are using different strategies. Now, obviously, the total number of gains earned by a team is not that informative on its own - the sides on top will tend to be ones with large squads, younger players, and those who have played lots of friendlies already this season. So, to really see how well each strategy is working I've looked at each team (after S.111 Round 19 but before the weekend cups) and noted the age, number of league games, number of friendlies and total attributes gained for each player on those teams. From this we can get a rough estimate of the number of attributes earned per game for each player*. Here's a plot of how many attributes per match were earned by players of different ages on these teams:

 

 

 Figure 1: Average number of attributes earned per match for different ages.
Data from 21 strongly gaining teams; Season 111 Rounds 1-19. 

 

Here we can see of course that younger players tend to gain faster than older ones, as we already know. Your youngsters should be getting an attribute for roughly every five or 6 competitive games they play (or roughly every 8-10 friendlies) while your over 30s probably need to play twice as many games for each attribute. The important thing about this is that we can now subtract the appropriate rate from each player's actual attributes/game to see how well they're doing compared to what we'd expect, given their age and the number of games they've played in. Let's call this the Adjusted Gaining Rate

Also, for each team we can figure out the average age of the lineups they actually played (not just the average squad age, but the age of the players on the field), because we know how many matches each of their player appeared in. The average lineup age should be a decent measure of the extent to which that team adopted the "lots of old players" strategy. So, we should see higher Adjusted Gaining Rates per player for older lineups (but lower actual gains, since they're playing fewer young players in friendlies). Which is what we see, albeit with quite a bit of variation:

 

 Figure 2: Adjusted Gaining Rate for lineups of different ages. 

 

That said, there are two teams in this dataset that look a bit different to all the others - the ones in the bottom left corner. Those sides have an average starting lineup of below 20, and have played zero players above the age of 30 this season. They also have pretty terrible gaining rates on their players. That's not to say their teams haven't racked up big gains - they're in the top-20 so far this season - but that's only because they have large squads, filled with young players, and have played most of their friendlies already, and these things are masking the fact that their gains are way below what you would expect from their players. So a pretty solid conclusion you can draw from that is that you're not going to get huge gains on individual players if you surround them with 20-year-olds.

OK, let's not do that. But how old do we really need our squad to be? Let's instead look just at the sides with at least some 31+ year-olds in their squad. Suddenly, the pattern isn't quite as conclusive:

 

Figure 3: Adjusted Gaining Rate for lineups of different ages (but with at least some players over 30). 

 

That's a tiny correlation of just 0.07, way below the size you'd need to be confident there was really an effect. You hear from almost every gaining guide that you should aim for an average lineup age of 28 or more. Now, of the top gaining sides this season, only a couple are getting close to that kind of average age, and with pretty mixed results. So:

Although you definitely need at least some older players to get decent gains, there appears to be far less difference between an average age of 22 and one of 28.

 

Now, (Champion's-League-League-Cup-and-8!!-times-English-League-Champion) Crazy Lion has a very detailed blog post on gaining which you should read. Interestingly, in point 32 Crazy Lion suggests that:

"Spinner maintains that having one older player per line will help with gains just as much as filling your squad with older players"

- Crazy Lion

 

I think that leaves open the possibility that older and older squads lead to diminishing returns. It's definitely possible that 3 youngsters surrounded by 38-year-olds will gain slightly better than 5 surrounded by 31-year-olds, but I don't think it is certain. More importantly, if it's true, the size of the boost seems to be quite limited. If you enjoy pouring all your efforts into a very small handful of players each year and striving for the highest gains you can on those guys then go for it! It can't hurt to follow that strategy to its extreme. But if you're more interested in keeping your players, growing organically, and maintaining a competitive squad over a longer time, I think that's also possible. Perhaps it's more of a challenge, but that's why we play the game, right? And your team can always rise again if a strategy you try doesn't work out, so why not mix things up and try doing things a different way? Do whatever makes you enjoy the game most, and don't stress out about following a prescribed route to success!

One last thing. When we talk about gaining, we always talk about the effect on a single player - how do we maximise his gains? It's useful to remember that we're always trading off between gains on one player and those on the others, especially with a strategy that fills up most of our squad with older players. Of course it's true that +6Q on three players is better than +3Q on six players, but is it better than +5Q on six players? +5Q on ten players? That's a call you have to make based on your circumstances. So remember, the strongest effect that you see in this dataset is that maximising gain on a few players reduces your gains on everyone else:

 

:

Figure 4: Average gains per player, per match, for lineups of different ages.

 

To put this in perspective, here's a back of the envelope calculation. That difference in Figure 4 means somewhere around 100 extra attributes per season for a team with a few 31-year-olds and an average lineup age 22, compared to a team with average lineup age 28. The boost in adjusted gaining rate for focusing on very few youths, and surrounding them with an old lineup - i.e. the trend in Figure 3 - is worth something like 3-6 attributes per season in total. Getting a single extra attribute onto one of your 6 chosen youth players with this strategy is probably 'costing' something like 20 attributes in lost opportunity for other young players. Something to think about.

I hope this post has been interesting, and I'd love to hear what you think in the comments :-) Meanwhile, may your gains be fruitful, whatever strategy you choose!

 

- Belizio

 


 

*Details for those interested: Players can earn through league, cup, player cup or friendly games, plus through normal training and camps, but looking a player's page will only show you how many league and friendly games they played. So I assumed each team played in roughly 4 cup games and 16 player cup games this season, i.e. they won 50% of their knockout matches. Each player was assumed to appear in cups and player cups at the same rate as they appeared in league and friendly games respectively, so a player featuring in 75% of league games and 25% of friendlies was assumed to also have about 3 cup (75% x 4) and 4 player cup (25% x 16) appearances. I assumed every player took part in both of this season's camps, and however many trainings they had fitness available for (for this I assumed stamina of 70-79, losses of 2 fitness per match, and worked out how many trainings they needed to rest in to make the fitness match up over the season so far). I then weighted each type of match and training according to my own data over the previous 3 seasons, i.e. Camps > Matches > Friendlies > Training, with the weight of a competitive match set at 1 - which means all the numbers that follow are "attributes earned per match", but also take into account attributes earned other ways. One attribute per match is roughly equivalent to 0.75 attributes per friendly, 0.45 attributes per training, or 1.9 attributes per camp.

 

 

 

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Gaining Part 3: Gains & Opposition Quality (27/02/2014 06:01)

"Don't be ridiculous, we're way too good to share the same field as your rag-tag bunch of footballing lepers."

- Every team who has ever rejected my friendly challenge, in my head at least

 

For those managers with the time to play them, most player gains come from friendlies. The most credit-efficient way of getting to 200 per season is to take 100 open challenges, and play 50 'sets' - that is, a home and away tie against another team. Once I started buying credits and found myself with enough to spare, I went over to the forum and started challenging people who'd requested sets. To my surprise - and, OK, indignation - most of these were ignored or rejected.

Flashbacks to high school. Didn't the forum say that teams asking for challenges would accept them, and challenge you back? And weren't these teams still posting messages asking for friendlies ten minutes later? Why was I being so cruelly rejected from the dance? Could it be that my lower-division side was failing to fill their stadium and costing them some money? (No, by the way - even I hit the maximum income for friendlies and my stadium's only just reached 20,000). Well, after a while I realised my self-consciously ridiculous team name probably wasn't the reason, and stopped taking it personally. It seems instead that player gains are affected by the quality of the opposition you face, so most of these sides were waiting for someone with a higher Q rating to challenge them.

In a comment on the previous post, King-Eric highlighted an old quote from Spinner which seems to confirm many of the factors involved in calculating player gains during a match:

"There are so many factors involved in this "gaining-bit", each of them alone has almost no impact. Ok, a stronger opponent might increase the chance of gaining marginally. However, against a strong opponent, your team is likely to perform "less good" due to being tackled, intercepted etc, which is another factor in the gaining-thing. Along with minutes on the pitch, individual performance, potential, which stat is selected for gaining, how high the stat is, how much the stat was used etc etc etc etc...." 

- Spinner

 

So let's take that as true. Better opposition quality means better gaining. We knew this already! Our gaining strategies are optimised! We can all go home now, right? Not so fast.

It might be true that stronger opponents give better gains. But look at all the other factors that are true - attribute levels, age, stars, minutes on the pitch, performance, age of surrounding players, position... and a healthy dollop of luck. As Spinner himself notes, the fact that there are so many factors means that the influence any single factor has on your gaining is small. It's also easy to forget that every time you make a decision, or do things a certain way, it has other consequences - missed opportunities for example, or unexpected impacts on other factors. If you just blindly restrict yourself to higher-quality opponents in friendlies (or play your defenders in goal...) you may dilute or even reduce your gains through lower team or individual performance. And restricting yourself to only certain opponents means it could take longer to play all your friendlies, requiring a greater investment of time in the game. All this is to say:

 

It's useful to know whether something has an effect - but it's more important to know how big the effect is.

 

That could be the mantra of this blog.


 

Opponent Strength: To Steamroller Or Be Steamrollered?

So let's take a look at opposition quality in particular. This is an interesting case because as noted above, if you take the Spinner quote at his word there are several different factors working in opposite directions:

1) Better opponent quality should lead to better gains. But...

2) Better opponents also leads to poorer performance from your own players, and less time on the ball. Which leads to poorer gains. But you could counter this by having...

3) Higher-quality players on your own team. But that means fewer gains..

Etcetera. Let's decide we can't really control the quality of our own players - we know who we want to gain, and we just want to maximise how many attributes they get. So which should we aim for? Easier opponents, or harder ones? And is the overall effect big enough to worry about selecting our opponents carefully, or is it so small that we'd really be best off playing whoever challenges us first and using the time saved to go out in the sunshine?*

To answer this question, I've kept a track of my last 342 competitive and friendly games, recording the type of game, the opponent, their quality, and the number of attributes my players earned for each. I can now look at the attributes, and determine how strongly related they are to the type of match and the quality of the opposition. I've isolated the effect of each factor in a couple of ways - partial correlations and multiple linear regression - and they give pretty much the same conclusions.

It won't surprise anyone to know that friendlies earn you fewer attributes than player cups, which in turn earn you fewer than competitive league and cup fixtures. The effect isn't really that dramatic: Friendlies only gained 27% fewer attributes than competitive games, while player cups were in between. That's the reason that friendlies still make up the bulk of most teams gains in a season - you play so many, and they're nearly as good as playing a real match. But even this modest difference is very strongly statistically significant, with p-values less that 0.0001 for both of the analyses I ran. That means there's less than a 1 in 10,000 chance of seeing such a difference just through random chance, and means we can be very confident both that friendlies really do earn fewer attributes, and that our analysis is powerful enough to detect quite small patterns in the data very clearly. So what happens when we look at the effect of opponent quality in the same way?

Answer: We see virtually no effect. Very slightly more attributes were gained against higher-quality teams: For example, playing against a Q90 team rather than a Q79 one earned about 7% more attributes. To put it another way, the quality of your opposition determined about half of one percent of the match gains, while the other 99.5% was determined by all those other factors we talked about. It's not a big effect. What's more, we can't even be sure this effect is real. Unlike the difference between playing a friendly or league match, the significance of the trend was marginal at best, with p-values of 0.18 to 0.19 depending on the analysis. In other words, even if quality had zero overall effect, we'd see this pattern about 1 time in 5. Not exactly something you'd stake your house on.

My guess, then, is that since opposition quality has a negative effect on your own player and team performance, these two things roughly cancel out, and you gain about as many attributes against a Q65 team as a Q95 one. The story might change as we add more data, but for now I'd conclude:

 

Don't worry too much about the quality of your opponent. Any differences in gains are so small you probably won't even notice them. Go outdoors and enjoy the sunshine.

 

I'll add a small caveat. If you're a very small team, with a stadium capacity below 15-20,000, you'll probably earn a little extra money from playing a higher division side. If your stadium's much bigger though, you should be maxing out your income against virtually any opposition. And for stronger teams, it's possible you'll pick up more injuries against weaker sides, since lower tackling stats should lead to more fouls (on the other hand you get more practice building up hidden free-kick stats). Having said that, my thankfully rare friendly injuries have all come from better teams so far, so I think it's mainly just luck.

Finally, though, there's one much bigger reason why we should all relax and accept challenges from smaller teams: It's just polite, and you'll be helping some new player get involved and addicted to the game. Remember that was you, once upon a time! :-)

As always, let me know what you think in the comments! Next post I'll look at whether there's a 'form' element to gaining. Should you keep playing friendlies when those gains are rolling in?

 

- Belizio

 

 

*The author lives in California. Your climate may differ.

 

 

Click Here to return to the Wonderstar Analytics home page

 

Read the full entry
Gaining Part 2: Positions and Gaining (26/02/2014 19:41)

 “You are what you practice most.” 

- Richard Carlson

 

In the previous post I talked a little about player gains, and whether you could really improve them by surrounding your players with older ones. Today I'm going to look at how to control the type of attributes your players gain, rather than just the number.

For training and player camps, you can control this directly by choosing the attributes each player should focus on. But in matches, attributes scatter randomly across your players. Well, perhaps not randomly - those goalkeeping gains on your strikers must surely be some kind of cruel joke meted out by Spinner and his diabolical dice.

Ask around in the forums though, and most people believe that attribute gains aren't completely random. Instead, it's generally thought that different positions will boost different attributes - put your defender in midfield for example, and he should improve his passing. Play your midfielders up front, and their shooting should improve. It makes sense that a player's improvement should depend on the role he plays in a game, and Spinner's 15,000 lines of code are surely complex enough to include this sort of detail. So how true is it?

Once again, I'm going to dig into the data I've collected on my own players - just under 1200 match gains from the last 2.5 seasons, plus the number of friendlies and competitive games it took them to gain these. That means I know the average number of attributes gained per match for each player. What's more, I know this for each type of attribute separately - so for example I know that Danny Vanderhaeghe gained eight tackling in season 109, at a rate of +0.055 per match. Armed with all these numbers, we can use them to test whether the position a player is on the field affects the distribution of attribute gains they receive.


 

Measuring Gains

OK, so how do we measure the rate at which attributes are improving in different positions? Well first off, I used my regression model from the previous post to factor out the effect of age - in other words, I worked out the rate at which each player should be gaining attributes in a match, given their age, and subtracted that from what they actually earned. So now for every attribute on every player, for 3 different seasons, I know how much more or less they gained on that attribute per match than you would expect for someone their age. To give you an idea of what I'm talking about, here's a snapshot for season 109:

Season 109 attribute gains per match, compared to the average for a player of the same age. Darker shading means better gaining.

 

So, for example, we can see that Danny Vanderhaeghe's +8 tackling over the course of that season was a better return than someone of his age should expect from the number of games he played: A typical 22-year-old should gain each attribute at a rate of +0.024 per match, so Danny's rate of +0.055/match during season 109 means his return is +0.031 better (0.055 minus 0.024). I also ran a quick correlation to see whether attributes tended to gain more slowly if they started out higher - this came out nonsignificant and extremely small, accounting for less than half a percent of the total variation we see in attribute gains. This is probably because not many of my players have high enough attributes yet for their gaining to slow down. The potential of the players also didn't have any significant effect on the gains. So, having checked for and removed the usual factors that determine gaining, we can now determine whether all the differences in the table above are just random, or whether they're being influenced by the positions each player is in. 


 

Effects of Position

We know what position each of those players was in for their games because I rarely if ever play them out of position. This means we can test out the idea that playing someone in defence helps their tackling, or that strikers earn shooting and heading stats faster than goalkeepers do. To do this, I averaged together all the extra attributes earned by players in a given position, and put them in the table below. To make things easier to interpret, each number is shown as a percentage of the total expected gain. So the defenders (green) Tk value of +20% means that tackling is gaining 20% faster in defenders than you might expect if all their attributes gained at the same rate.

Distribution of match attribute gains for different positions, relative to expected gains. Darker shading means better gaining.

 

So how do we interpret this? Well, just naively glancing at the pattern, we can spot a few things we would expect if position was really having an effect on gains. For example, attackers gain more shooting than other positions do, and attackers and defenders both gain heading a bit faster too. Passing increases a bit better for outfield players than goalkeepers, though not really any faster for midfielders. But there's a few things that look wrong, too: Why would tackling be gaining so much faster for goalkeepers and attackers than midfielders and defenders? And why does playing in goal help your stamina, but not your goalkeeping stat?

Certainly the data are pretty noisy. And we know already that position can't be the whole story, or we wouldn't keep getting keeping stat increases for our outfield players. But is there still a small, but real, advantage to be gained by playing in a particular position? A simple way to test this is by running some ANOVAs on the data. I'll skip over the statistical details, but essentially, this approach looks at the sizes of the effects in the table above, as well as the amount of variation across players, and gives us an estimate of whether a given pattern is reliable - consistent across players - or just down to random chance.

So, let's look at the pattern for each position, and use the ANOVA to tell us whether it's consistent across players. For goalkeepers, the ANOVA comes back very non-significant, indicating that those big numbers we see aren't really that reliable, but are instead just because we didn't have very many goalkeepers to look at in our sample. Defenders show the same story, although we have more data to look at we still can't really be certain that the pattern in the green row is very reliable. But the situation changes when we look at midfielders and goalkeepers: The distribution of gains across both of those rows seems to be pretty consistent across all of our players and seasons. In other words, it seems like maybe midfielders do gain tackling faster than they gain perception, and that strikers gain heading more than stamina.

And what's more, not only are these consistent patterns starting to emerge from our data, but they kind of make sense too. That's because the distribution of gains for midfielders and strikers roughly follow the stats that are important for those positions - and therefore the stats they might be using in a match. In other words, the pattern of gains for each position above shows a positive correlation with the amount each of the attributes contributes to quality. To use attackers as an example, heading, shooting and speed all contribute quite a lot to the attacker's quality, and they also increase a bit faster than normal during matches. Meanwhile goalkeeping, passing and stamina are a bit less important to quality, and don't increase so quickly. Overall, the mantra that you can influence the attributes gained by changing the position of your players seems to be true:

 

Position on the pitch does seem to affect the type of attributes players gain. 

 

Having said that, I'd like to revisit this in another thousand attributes or so just to see if the patterns are still there. The effects of position also seem to be quite small - gaining a stat 20-40% faster sounds like a lot, but that really just means one extra per season. So I wouldn't drive yourself crazy micromanaging your gains too much using this strategy, that's a sure-fire way to invite retribution from the gods of chance and be showered with goalkeeping stats.

Nonetheless, if you take this at face value, then go ahead and play your future attacking midfielders up front. Can't hurt. But make damn sure to put your young defender in goal for every last friendly...     


 

Any feedback or questions, please feel free to comment! In the next post I'll take a look at opposition quality. Can you gain faster by playing against higher quality teams? And if so, just how much?

 

- Belizio

 

  

Click Here to return to the Wonderstar Analytics home page

Read the full entry
Gaining Part 1: Oldies & Player Gains (26/02/2014 05:37)

"If a black cat crosses your path, it signifies that the animal is going somewhere."

- Groucho Marx

 Click Here to return to the Wonderstar Analytics home page

I thought I'd better jump straight in and tackle the big one. I've seen more theories and superstition and frustration over player gains than anything else here (OK, teamstats run them close...). Even winning matches in Managerleague seems to take second place to getting that extra perception stat on your young midfielder. The forums are dominated by gaining contests. Champion's League teams sell all their decent players at the start of the season and throw their youth players in to get torn apart - and maybe, just maybe, gain some quality to make the humiliation worth it.

I totally understand. In the previous post, I touched on how the game's depth keeps us challenged and prevents us getting bored. But I think it's the pace of the game, and especially the opportunities for gaining, which get us really addicted. Who hasn't sat and refreshed frantically to see whether their training camp has bumped their striker up a quality point? Or jumped straight onto ML-Reports after a league match to check what they earned? Or squeezed one last friendly in before training, convinced this is the one that's going to give you 4 great attribute gains and make up for your run of blanks? Yeah, me too. That frequent, satisfying little dopamine rush keeps us coming back.

That said, I don't ever see myself selling all my players and starting from scratch with an army of identical 17-year-old clones, or buying a young prospect with the sole intention of selling him a few weeks later. I enjoy building up my squad organically, and forming an attachment with my players. I'll take good old Danny Vanderhaeghe, my limited but industrious midfielder, over any five-star teenager using my team as a stepping stone to greater things. Ultimately, I expect I'll hit a ceiling around Q88 in division two, but so long as Danny's still puffing away in midfield and outperforming his moderate quality level I'll be satisfied. So I guess keep that in mind - this article isn't really from the perspective of someone looking for monster gains on six players surrounded by a hollowed-out husk of a team sitting in division 5.


 

OK OK, I get it. You don't want to be a youth farm. But how do I get my players to gain?

Make the appropriate sacrifices to the gaining gods and your Q67 youth player might just become the lynchpin of your side for years to come. Anger them and you'll end up with a midfield full of exceptional goalkeepers.

But what are the appropriate sacrifices? Nose around on the forums and you'll find dozens of confident mantras:

 

"Make sure you surround your young players with 36-year old goalkeepers, they won't steal all the gains."

"Play your midfielders up front to boost their shooting skills."

"If you're not getting any gains in your friendlies, take a break and try again in an hour."

"Only play against Q90+ teams, you'll get more gains against them."

"Never play on a full moon, all of your gains will disappear and turn back into pumpkins in the morning."

 

Spoiler alert: I ignore all of these. Nevertheless, over the last three and a half seasons I have gained almost 1800 attributes, ranking 110th, 4th and 27th at the last 3 season ends (as I write this we're in the top ten for this season). And though I once found a pumpkin in my house, I think my wife must have bought it because there were no attributes missing from my players.

Now, just getting a bunch of attributes each season isn't necessarily that incredible - they're spread out over quite a few players, so not so useful if I was trying to sell Q80 18-year-olds every year, and my team is pretty young and low-quality, so I don't have many players hitting their maximum quality yet. Besides, out of the 40,000 teams in these rankings only a small proportion of them will be active and playing 200 friendlies a year. But even if I'm only in the top 10%, my point is that I can't be doing that terribly by breaking the rules above. Am I just getting some unusual luck? Or are these rules not as firm as they seem?


 

Data to the rescue!

I'm pretty sure I'm not alone in this, but I've been keeping track of some of my team's data. For the last 2.5 seasons, I have a record of all the attributes each of my players has gained in friendlies or competitive matches, as well as the number of friendlies/matches they've played. That gives us about 1200 attribute gains to look at. I've also got a list of roughly 250 matches where I've recorded the quality of the team I've faced, the type of match (player-cup, friendly, league etc.) and the number of attributes my players gained. Armed with this, let's take a look at some of the claims above in the next series of blog posts.

First, does surrounding your young players with older ones help? I think Spinner has confirmed that player gains are not zero-sum: they're worked out individually, so you don't have to worry about your older players "sucking up" gains that would otherwise have gone to someone younger. More interestingly, though, he's also told us that it helps to have some experienced players on the pitch alongside your youths. The only question is how much does it help, and is it worth it to spend valuable game time on those useless older players? To a certain extent this is a question of priorities - mine is to build and maintain a team organically, without selling many players young. If you just need those three valuable youngsters to gain and you don't care about the rest of your side then your strategy is going to be different. But let's take a look at some of the data.

Remember how I said I kept track of each player's match gains, as well as the number of games they played in? Well this means I can work out how many attributes per game each of my players earns on average. Looking at my younger players, an average gain of just under 0.2 attributes per match is typical (or 0.14 per friendly). In other words, if my 18-year-old plays in 5 league games or 7 friendlies, he should gain one attribute on average. Maybe none, maybe a couple, but over the course of a season that's the average gaining rate from playing games.

Now, suppose I wanted to make sure this guy earned as many attributes per season as possible - exactly like a youth farm team, except I'm going to keep playing all my usual 20-25 year olds alongside him. I could play him for 15mins in 200 friendlies, and let's say 65 competitive games in the league, cups and player cups: That comes out at 40 attributes per season on average. Even if I'm having to rest him on every training, he's got 5 camps to add a couple more attributes, so let's call it 42 per season. Maybe 30 one season, 55 the next, depending on the dice. That means that even with the moderate and low star players I have, surrounded by other young and medium-aged players, I could still be getting quality gains of 5-7 for an individual player simply by making sure I play him.

Now, this means that in order to justify using up some of those matchtime slots for old players, a youth farm team surrounding his player with decrepit goalkeepers has to see much better gains than 42 attributes per season. Not just on one player a season: He needs his chosen players to be putting on +50 on average. It's possible that's the case, and certainly a few players every year will get those kinds of gains, but my sense is that they're more the exception than the rule. Huge gains come mostly from playing huge numbers of matches - the average gain per match doesn't seem to be affected that much by the ages of the other players on the field.

So to address the first strategy from the list above:

 

Surrounding young players with older ones probably helps their gains - but perhaps not that much.

 

Does it sound like I'm on the fence? Well, put it this way: I'm not going to be in the market for 35-year-old benchwarmers any time soon. I think this mantra is overrated. I'd love to hear what you think though - has this made you think differently about your gains, or do you reckon I'm underestimating the influence of experience? Let me know in the comments!


 

In the next post, I'll be tackling the question of how position affects your attribute gains. Should you be putting your midfielders in attack to help improve their shooting?

 

- Belizio

 

Click Here to return to the Wonderstar Analytics home page

 

 

Read the full entry
Welcome to Wonderstar Analytics (26/02/2014 04:46)

 Welcome to the blog!


The original first post is below, and remains to serve as a sort of introduction. But from now on, this post is going to serve as a sort of homepage, with updated links to everything I post. Hopefully you'll find it useful, and thanks for reading!

 


 

DONATIONS

 

Some kind managers have thrown a few credits my way in appreciation of the blog. Thanks! If you'd like to do so too, I greatly appreciate it - but I'd like to be able to give you something back too!

 

So rather than just donate, why not get a Team Analysis for ten credits?

Or, if you don't want/need one of those, sometimes I run more in-depth analyses that end up as a pdf, rather than on the blog (recent examples include a bigger and more detailed analysis of player attribute effects, and a more complex gaining analysis that measures the effect of stars across a player's career). These usually get sent round to managers who've helped with the data/analysis ideas, or who have spontaneously donated to the blog.

Or, in the future I might make little excel widgets to use with your team. Again, if you've spontaneously donated you might get first dibs on one of these.

Finally, if you want your name in lights, why not sponsor one of the manager awards I calculate?

 

If you'd like to donate for any of those reasons, or just plain generosity, you can do so HERE. Thanks!

 


 

 LATEST POSTS

The most recently updated posts are listed here!

 

UPDATED! The ManagerLeague Manager League!

A list of the top managers in Div1, created by summing the Manager Ratings from Season 112-115. 

 

UPDATED! The ManagerLeague League of Leagues!

A ranking of the 33 leagues in the game, based on a rolling 5-season weighted average of their league ratings. 

 

NEW! League of the Season S114 

An alternative award, given to the Div 1 manager who has most outperformed his or her team strength.

 

NEW! League of the Season S115 

The latest version updated for Season 115.

 

NEW! Manager of the Season S114 

An alternative award, given to the Div 1 manager who has most outperformed his or her team strength.

 

NEW! Manager of the Season S115 

The latest version updated for Season 115.

 

The Gaining Pie!

 A quarter of a million attribute gains carefully sifted through to make one delicious pie of data. How important is age to gaining? Quality? Friendlies? Find out here!

 



ARCHIVED POSTS

This section will be kept a bit more organised, with similar topics together.

 

Personalised Match Analytics - A service I offer for credit-buyers: A detailed statistical breakdown of performance, to help refine tactics (or just for interest!).

 

--- GAINING ---

Gaining Part 1 - A brief, data-light discussion of gaining.

Gaining Part 2 - Does the position you play in affect what attributes you gain? Yes. Well. Kind of.

Gaining Part 3 - A look at opposition quality: Should you be playing friendlies against strong teams? Or sides you can beat easily?

Gaining Part 4 - A more in-depth, data-rich discussion of gaining, and just how much the average lineup age really helps.

Gaining Part 5 - A quick post about interpreting statistics, and a confirmation that pumpkins have nothing to do with ManagerLeague.

Gaining Part 6 - A quarter of a million attribute gains carefully sifted through to make one delicious pie of data, and a recipe for gaining. 

 

--- STRATEGY & TACTICS ---

Introducing Match Analytics - Starting to look beyond results and performance: Separating possession from conversion rates.

Match Analytics Part 1: Stamina & Substitutions - A case study of match dynamics, emphasising the importance of substituting your less fit players.

Match Analytics Part 2: Possession Types - Delving deeper into the sim. How do conversion rates change for different possessions?

 

--- PLAYER ATTRIBUTES AND CONTROVERSY ---

Player Attribute Rating Contest - The contest is now over, but this is still a useful introduction to the next few (highly controversial!) posts. And see also this thread for some other entries and discussion.

Contest Results - The winners of the attribute rating contest. 

Player Attributes - In which I explain why tackling is a striker's most important attribute, to a mixture of sage nodding, skepticism and outright disbelief - recorded in these two forum posts.

Caveats and Details - Some additional important notes about interpreting the Player Attributes analysis that didn't fit in the original article (apparently there's a word limit!).

 

--- AWARDS AND RECORDS ---

The ManagerLeague Manager League! - A list of the top managers in Div1, created by summing the Manager Ratings from Season 112 onwards.

Manager of the Season S112 - An alternative award, given to the Div 1 manager who has most outperformed his or her team strength.

Manager of the Season S113 - The latest version updated for Season 113.

Manager of the Season S114 - The latest version updated for Season 114.

Manager of the Season S115 - The latest version updated for Season 115.

---

The ManagerLeague League of Leagues! - An updated list of the strongest leagues, created by summing the last 5 seasons of League Ratings.

League of the Season S112 - A table of all the different leagues, based on how well their teams perform internationally, and how exciting the top division is.

League of the Season S113 - The latest version updated for Season 113.

League of the Season S114 - The latest version updated for Season 114.

 

League of the Season S115 - The latest version updated for Season 115.

---

Wonderstar Player Awards Season 112-113 - Alternative Player awards for Seasons 112-113. 

Champions League Preview/Review S113 - A case study in why you should never tryto predict the most unpredictable competition in ManagerLeague.

 

 

 

 


"The greatest trick Spinner ever pulled was convincing the world he didn't exist keeping the rules to himself."

 

Managerleague is... deep. When I started playing - alongside a half dozen other online football games - it was just to kill a few hours. As I looked at the paltry 8 attributes, the simplistic formulae for player quality and the paucity of positions I was pretty confident this wasn't going to hold my interest for long.

Given how wrong I was, I guess you ought to take everything else in this blog with a large dose of salt :-) Long after those other games stopped sending me plaintive reminder emails and gave up on me, I'm as committed as ever to this one. So why is that?

Well, lots of things, obviously. But mainly, there's the creeping sense as you play this game that it's way more complex than you realised. That complexity arises first and foremost from the fact that the mechanics are largely unknown. Having spent a little time perusing the forums, the manual, and all the help guides out there one key thing struck me:


The number of mechanics asserted by managers far outstrips the number Spinner has told us about.


Essentially, people are learning by playing, and their experience lead them to conclude that things work a certain way. Or, when they begin, an older player tells them things work a certain way and that's what they remember. Sometimes they're right, sometimes they're part right, and sometime they're dead wrong. That's great - it adds a whole extra layer to the game. Not only are you trying to manage your team, but you're also trying to figure out the system. The better you are at doing that, the greater your advantage. More than any similar game out there, Managerleague rewards the savviest managers. 

That brings me to the point of this blog. There's a ton of received wisdom about the game, and being a data scientist it always nags at me when someone makes a claim without supporting evidence... I've been collecting plenty of data in the few months I've been playing and reached some conclusions of my own. Of course, you should remember that pretty much every blog is just one person's view (though one that in this case is hopefully backed up by evidence). You should play your own way, and make your own decisions! However you enoy playing the game is by definition the 'right' way. That said, hopefully you'll find it interesting!  


- Belizio

 

 

Read the full entry