Monday, January 31, 2011

Shots and Conversion in the Big 4 Leagues: The 2010/11 Season To Date

Since my last report about conversion and shot records for various teams in the EPL, I've been wondering how things were looking so far this season on the shot frequency & conversion front. But instead of analyzing one league at a time, I thought it'd be interesting to compare several leagues together. This has the nice advantage of providing a greater number and range of teams to analyze.

So here we go. Below are average shot frequency and conversion ratios for all teams in the Bundesliga, the EPL, La Liga, and Serie A at the end of 2010 (that is, roughly halfway through the 2010/11 season). 

To put things in perspective, keep in mind that, across the leagues this season, the median number of shots taken per team/match has been 12.6, and the median conversion rate has been .27. This means that 50% of teams took more than 12.6 shots and had a conversion rate better than.27 (or 1 goal in 4 shots on target), while 50% of teams shot and converted less well. To help orient readers, I've superimposed a red line to separate the space into those where performance is better than 50% of teams and worse, on these two dimensions.

Here's what else we can see:

Sunday, January 30, 2011

Great Soccer Reads: Outcasts United and Pay As You Play

If you're a regular reader of this blog, you know by now that I'm a "Soccer Dad" - I coach my son's team of 10 and 11 year olds - as well as a huge fan of trying to generate genuine insights into the beautiful game by systematically exploiting evidence and fact. Why do I tell you? Well, it affects what I read for fun. So here are two book recommendations that deal with each of these passions of mine.

First, I just finished Outcasts United, a book that emerged from a series of stories Warren St. John wrote for the New York Times (here's an example).  Here's the book's description from the publisher's website: 
"Outcasts United is the story of a refugee soccer team, a remarkable woman coach and a small southern town turned upside down by the process of refugee resettlement. In the 1990s, that town, Clarkston, Georgia, became a resettlement center for refugees from war zones in Liberia, Congo, Sudan, Iraq and Afghanistan. The town also became home to Luma Mufleh, an American-educated Jordanian woman who founded a youth soccer team to help keep Clarkston’s boys off the streets. These boys named themselves the Fugees -- short for refugees. Outcasts United follows a pivotal season in the life of the Fugees, their families and their charismatic coach as they struggle to build new lives in a fading town overwhelmed by change. Theirs is a story about resilience, the power of one person to make a difference and the daunting challenge of creating community in a place where people seem to have little in common."
Nicole Bengiveno/The New York Times
Rumor has it that the book is being turned into a movie (the rights, at least, have been sold). And that's great; but don't wait for the movie. St.John tells a compelling story; it's also beautifully and competently written, and the book is a kind of testament to the power of human compassion and the love children all over the world have for the game. It's also an interesting story about how an American town copes with foreigners and immigrants in its midst - an especially novel development in the United States over the past decade, where levels of immigration were higher than they had been for almost a hundred years (today, about 13% of all people living in the U.S. legally were born in a foreign country). 

A minor side note: the book is also an interesting read for what it tells us about youth soccer in the U.S., with its predominant pay to play set-up (where kids have to pay - often quite a bit of money - to play on a soccer team). This tends to have the unfortunate side effect of giving the sport a middle- and upper-middle class demographic profile (with the exception of Latino immigration, of course).

The second book I'd recommend also has something to say about pay to play, but in a very different way. It's Pay As You Play: The True Price of Success in the Premier League Era by Paul Tomkins, Graeme Riley, and Gary Fulcher. For anyone interested in money and success in soccer, this is a great, great read. It's part data report - the authors did an amazing job collecting data on transfer prices for individual players during the Premier League era. But the book doesn't just tell us how much individual players have cost; instead, the authors find a way to make transfer costs equivalent across years to account for inflation through their Transfer Price Index. This allows us to compare the costs of players transferred in the 1990s to players in more recent years. Aside from reporting on the amazing data they have collected, the authors systematically look into the relationship between transfer spending and team success. Guess what?

Thursday, January 27, 2011

Dimensions of Defensive Production: Degrading Opponents' Shot Frequency and Accuracy in the Premier League

Following up on my post about shot creation versus shot conversion as measures of offensive production (see also the On Football post), here's a quick and dirty analysis of Premier League teams' ability to degrade opponents' shot frequency (measured by shots allowed) and shot accuracy (the ratio of shots on target to shots allowed).

We can think of these as two dimensions that aren't necessarily related. Teams can allow the other team lots of shots without letting them make accurate shots; or they can degrade their opponents' ability to create lots of chances while allowing a high level of accuracy on their opponents' part (say, allowing them to hit 3 of 4 shots). If it's the former, they may need to focus on improving midfielders' defensive capabilities to degrade the other side's ball delivery into good striking positions; if it's the latter, they may need to focus on defenders' ability to degrade chances that crop up.

Using data from the 2009/10 season gives you the following picture (calculating each team's average over the course of the season).

Two things are noticeable about the pattern we see.

First, there seems to be a moderate positive relationship between opposing team frequency and accuracy; that is, teams that allow more shots are also more likely to allow accurate shots overall. However, there is considerable variation around this relationship, mainly allowing us to say that some teams are just weaker defensively (or stronger, depending on how you look at it). The teams that excelled defensively in 2009/10 were Chelsea and Liverpool (located he lower left hand corner): they allowed the fewest shots overall, and they had the best overall frequency/accuracy combination. In contrast, Burnley allowed their opponents to take lots of shots and were unable to prevent opposing teams from making them accurate.

Second, we can see that the upper left hand corner of the graph is mostly empty; this means that there are really no teams that allow opponents to take lots of shots but manage to degrade their accuracy significantly. Instead, we see teams that excel at degrading both frequency and accuracy of shots (Chelsea and Liverpool, and to some extent Man United). Then there are teams that allow relatively few shots but also allow opponents to shoot with higher levels of accuracy (Arsenal, and to some extent Wigan and Tottenham - I know this is debatable). And then there's everyone else clustered in the middle, allowing between 10 to 14 shots, and decent accuracy around .54.

Before we start taking these numbers from one season and one league too literally, it's clear that, in 2009/10, the best Premier League teams clustered in the lower left hand quadrant, suggesting that good defense paid nice dividends last year.

Creating Chances v. Taking Them: The Premier League in 2009/10

Ever since I started calculating shot to goal ratios as well as conversion and accuracy ratios of teams' offensive (and defensive) production, I've been concerned that these ratios do not tell the full story of what happens on the field. Conversion, you may remember, is the ratio of goals to shots on target - that is, how many of the high quality chances teams generated actually resulted in goals: how good teams are at finishing. Accuracy is the ratio of high quality shots to all shots teams took: how good teams are at generating high quality chances.

But I've always thought that there's a difference between Team A that finishes 1 of 5 chances and Team B that finishes 4 of 20. While both have a conversion ratio of .2, I'd much rather be playing on, coaching, and watching Team B, wouldn't you?! Turn it around: if you're coaching Team A, you know that your strikers are doing well converting their chances; but you are not generating sufficient opportunities for them. Hence, you may want to work on midfielders finding ways to get the ball in front of goal where we know the vast majority of high quality shots are taken.

So I was thrilled to see this clever post on the On Football blog, which took my concern one step further and actually did something with it, using MLS data. The author distinguishes between creating chances (shots) and converting them and plots these two dimensions of offensive production. What you see is that the best teams cluster in the upper right hand corner of such a graph: the best teams in MLS create lots of chances and they take them.

So can we generalize this pattern to other leagues? Since I had nothing better to do (well, I did, but I can't tell you about that), I grabbed the 2009/10 dataset for the Premier League and replicated On Football's graph for the EPL. Take a look at the average levels of creation and conversion by team for the 2009/10 season.
Four things stand out. First, just like in MLS, the top teams that year cluster in the upper right hand corner. Second, the struggling teams cluster in the lower left hand corner - they don't create many chances and they waste most of them. The rest of the league is somewhere in the lower right hand corner: these teams don't create as many chances as the top teams (in fact about as many as the bad teams), but achieving higher levels of finishing (with some, like Aston Villa or Stoke as high as Chelsea). Four, teams do not cluster in the upper left hand corner; there are basically no teams that create lots of chances and then don't take them.

Congrats to On Football's authors on moving things forward. Before too long, I'll also take a look at defensive production along these lines and perhaps across leagues. Stay tuned.

PS: One small note: conversion rates appear significantly higher in the English league than MLS.

Tuesday, January 25, 2011

Update: Accuracy Ratios Across the Leagues - The 2010/11 Season To Date

In case you were wondering how your favorite league compares on various metrics with the other big leagues of soccer, below is a quick graph of accuracy ratios (defined as the ratio of shots on target to all shots taken) across the Bundesliga, EPL, La Liga, and Serie A as of the end of the calendar year (roughly halfway through the season). Take a look.

The story is straightforward and familiar. Accuracy ratios are highest in the Premier League at over .5 (or 1 for every 2 shots taken is on target), and lowest in Serie A at .326 (or 1 in 3 shots). La Liga and the Bundesliga are identical and clearly more like Serie A than EPL in this regard at .364.

The graph also shows that the variation around these averages (expressed by the standard deviation around the mean) is very similar across leagues at .15, indicating that, while accuracy levels are different across the leagues, there is very similar variation around these levels. In plain English, this means that teams in each league are similar in their accuracy ratios, and most EPL teams therefore shoot more accurately than most teams in the other leagues. So maybe this is why watching the Premier League is more fun?! Or is it?

Are Professional Players Entrepreneurs Or Employees?

Demba Ba's refusal to play for Hoffenheim and demand a transfer to England or Jefferson Farfan's late arrival from winter break in Schalke and reputed demand to leave the club or Darren Bent's push to sign with Aston Villa all have the whiff of spoiled players behaving badly and selfishly (the list could easily include a good number of other players). Are these players just soccer's divas, or is there something more systematic going on? That is, are players behaving ever more like spoiled children or like rational, profit-maximizing capitalists?

There has been a lot of talk these last few weeks, especially during this transfer window, about the changed and changing relationship between players and clubs. This is not entirely new and goes all the way back to the beginnings of professional sports. But it gets a whole lot more attention when you see egregious examples of selfishness (or self-interest, some might say). And many, like Schalke's coach Felix Magath, think that clubs are currently getting the short end of the stick. He said so very explicitly in a recent opinion piece in the Hamburg newspaper Abendblatt. Paraphrasing Magath slightly (while trying not to put words in his mouth), players have gone from being employees to being subcontractors (or perhaps entrepreneurs).

Magath is on to something. Ever since the Bosman ruling in 1995, which provided the opportunity for a freer movement of players across leagues and countries, we have seen several interesting trends that bear on the relationship between players and clubs: one, more movement across clubs, period; two, more movement across countries; and three, an arms race among the best clubs for the best players.

More movement of players across clubs and leagues suggests that the market for soccer players has become more efficient (though we can debate that, too). But it's the latter part about what this movement means for clubs and ultimately the industry of soccer that appears to worry Magath.

Here's what Magath actually said (translated from German):

Sunday, January 23, 2011

Red Cards: How Common Are They?

In earlier posts, I looked at the correlation between fouls and red and yellow cards in the Big 4 leagues as well as the distribution of yellows and reds in the 2009/10 season or the cost of receiving yellow cards, but I haven't spent much time talking about red cards in and of themselves. There's a simple reason for that: red cards are relatively rare events compared to other things that happen on the field, like shots, goals, or even yellow cards. But they are the ultimate punishment in soccer, and when they do happen, they can be very costly to teams and individual players.

But how rare are red cards, really? And do the big leagues look similar in terms of how common red cards are? So let's take a look, starting with the distribution of red cards per team and match for the big leagues of soccer during the 2005/6 to 2009-10 seasons - that is, for five full seasons.

Clearly and as you'd expect, red cards are very unusual: the vast majority of matches teams play do not involve any reds at all. Between 2005 and 2009, the percentage of matches without a red range from highs of over 91% in the Premier League and the Bundesliga (91.9 and 91.5%, respectively) to a somewhat lower 84.7% in Serie A and a slightly lower 82.4% in La Liga. The distribution also shows that matches with 1 red card are about twice as common in La Liga and Serie A than in the Bundesliga and the EPL (roughly 15% v. 8%). So the southern leagues Serie A and La Liga "lead" the northern leagues Bundesliga and Premier League in red cards given.

These patterns are completely consistent with calculations of fouls per match I reported in a post back in September, with one interesting exception:

Friday, January 21, 2011

Who's the Best in the Bundesliga? Ask the Professionals (Shh ... It's Dortmund)

It's one thing to have supporters or analysts tell you who the best players, teams, or leagues are. Depending on who you ask or which metric you use, you can have endless arguments about quality. But why not go directly to the source and ask the professionals themselves? After all, they have to play the game week in and week out and see things on and off the field that spectators often don't or can't.

That's thankfully what Kicker, the German sports magazine, does with some regularity. Here are some interesting results from Kicker's most recent survey of 286 Bundesliga players who were asked questions about how the first half of the season has gone.

So who do the Bundesliga Profis say are the best players, teams, and coaches, and which are the worst, so far this season?
Nuri Sahin

Best player:
1. Nuri Sahin (Borussia Dortmund) [32.6% voted for him]
2. Shinji Kagawa (Borussia Dortmund) [17.5%]
3. Bastian Schweinsteiger (Bayern Munich) [11.2%]

Best goalkeeper: Manuel Neuer (Schalke 04) [54.2%]

Best coach:
1. Jürgen Klopp (Borussia Dortmund) [32.2 %]
2. Thomas Tuchel (Mainz 05) [24.1%]
3. Mirko Slomka (Hannover 96) [22.7%]

No real surprises here. Given Borussia Dortmund's amazing first half of the season, Dortmund cleans up when it comes to best players and coaches. Perhaps slightly surprising is that Dortmund has two players as No. 1 and 2 on the best player list, with Sahin and Kagawa. And Jürgen Klopp's selection as the best coach is not surprising, either, given where the team stands at the moment. Finally, the fact that Manuel Neuer is voted best goalie is not a surprise, either, given that he has been in fantastic form.

And Dortmund also beats out everyone else when it comes to best new/young players, with Kagawa winning out in that category, closely followed by his teammate Mario Götze. Again, the professionals aren't telling us something we didn't already know, but it's nice to see the consistent performance of Dortmund rewarded by their colleagues across the league.

Best new player:
1. Shinji Kagawa (Borussia Dortmund) [26.5%]
2. Mario Götze (Borussia Dortmund) [24.8%]

Mainz beats out Hannover and Dortmund as the surprise team of the season; to me both Mainz and Hannover are clearly positive surprises so far. And it makes sense that Dortmund would "only" place third here, given that the team has been getting better every year since Klopp took over the manager's job.

Surprise team of the season:
1. Mainz 05 [33.6%]
2. Hannover 96 [26.9%]
3. Borussia Dortmund [24.8%]

Interestingly, the survey also asked about the losers of the season so far (the German term is "Absteiger", the word typically used for relegated teams). Some highlights (or lowlights, if you will):

Wednesday, January 19, 2011

Goal and Shot Ratios: The Big 4 Leagues Mid-Season

How many shots have teams needed to score goals this year? In an earlier post, I had reported goal and shot ratios for the Big 4 leagues overall. Here's where we stand mid-season for the leagues and by individual teams.

Recall that, over the long run, the goal to shot ratio tends to be around .111 - or 1 goal in 9 shots. Across the four big leagues, it's clear that Serie A has by far the lowest goal to shot ratio - that is, it takes Serie A teams systematically more shots to score goals than teams in the other leagues. So far this season, Serie A is at .085 or roughly 1 in 12 shots - a third lower than what is "normal" for the big leagues. In contrast, the other three leagues are around the historical average at .104 (La Liga), .117 (EPL), and .123 (Bundesliga). So spectators in the Bundesliga only have had to see their teams take 8 shots before scoring a goal, while those in Serie A have seen their teams take a full 50% more shots (12) before getting on the score board.

But what about individual teams and differences within leagues? Take a look.

Clearly, the overall distribution of goal to shot ratios - even when split up by teams - puts Serie A to the left (that is, the lower end) of the distribution of goal to shot ratios. In contrast, the other three leagues also have some low performing teams, but they spread out to the right (the higher end). Clearly, Serie A teams have been the least efficient ones across the big leagues - with exceptions at the high end of course. But even the league's most efficient team (Juventus with a .124 goal to shot ratio so far this year) is only as good as the average Bundesliga team on this score. It also appears to be the case that the low performing clubs in La Liga and Serie A are the lowest performers across the four leagues (with the exception of Wigan - sorry Roberto Martinez).

And finally, Barcelona is literally off the charts, and not just compared to the other La Liga teams: its goal to shot ratio of .194 translates to almost 1 goal in 5 shots taken. That's breathtaking - statistically speaking and otherwise.

Tuesday, January 18, 2011

The Distribution of Corners in the Big Leagues of Soccer: The 2010-11 Season To Date

Corner kicks are kind of like free kicks in the opponent's half: both are set pieces. They're also among the few moments in a match when teams have a real and predictable opportunity to score. And while I've been thinking about them for some time, I have not reported any systematic analyses of corners. Recently, one of the faithful readers of this blog asked about them, so I thought I'd start by taking a quick look at the frequency and distribution of corners so far this season in the four big leagues (Bundesliga, Premier League, Serie A, and La Liga). 

So, as of Dec.23, here's what the distribution of corners per match (and team) look like across the leagues.

What do we see?

Monday, January 17, 2011

London Map of Soccer Fandom

Take a look at this great map of football fandom in London from It's just too pretty and creative to not share. Enjoy!

Thanks to Cas M. from the Grondhopper Blog.

Saturday, January 15, 2011

Opponents' Accuracy As A Measure of Defensive Production: The EPL Season at the Halfway Mark

I have occasionally written about defensive production in terms of how many goals, shots, or shots on goal teams allow in a match. One of the best of these measures, so far as I can tell, is whether a team allows their opponents to place accurate shots on goal. If you think about it, allowing an opponent to shoot is one thing - anyone can sling the ball in the general direction of the other side - but only accurate shots have a chance of actually crossing the goal line. In this sense, it's a much finer measure of what a team allows its opponent to do - or turn it around: it's a finer measure of what an offensive team is able to do to the other side.

So how have Premier League teams been doing on this score so far this year? First, the average accuracy score for opponents is .53 (or roughly 1 in 2 shots), with a range from a low of .48 to a high of .59. 

There are some expected and some surprising patterns - take a look.

What is expected is that some of the top teams like Man U and Man City lead the league in preventing their opponents from getting off accurate shots when they make it in front of their goal. At the same time, what is unexpected is that Arsenal is among the teams least able to prevent their opponents from taking accurate shots. But what is expected is that West Ham, Wolves, and Blackburn are in their vicinity among the teams that allow opponents to take accurate shots. Unexpected is Liverpool's, Fulham's, and Wigan's relative defensive prowess - who would have thought the Latics do well in degrading their opponents' shooting precision?

As with offensive production, a factor that may underlie these patterns is differences across home and away play. So let's take a look at that; below are opponents' accuracy measures when teams play away (top panel) or home (bottom panel).

The first thing to notice is that teams defend better at home than away, as evidenced by the range of accuracy ratios in the bottom panel (home defense) extending to the left of the top panel (away defense). This is consistent with a home-team effect documented with regard to great a variety of performance measures.

Perhaps surprisingly, given their historically awful season so far, Liverpool leads the EPL in preventing opponents from making accurate shots when playing at home (allowing 4 of 10 shots to find the target), followed with some distance by Man U. And Roberto Mancini's Man City are the kings of "accuracy prevention" when playing away from home (an opponent accuracy ratio of .45), and with some room to spare relative to the second best team (Everton). 

On the other side of the ledger, Blackburn and Chelsea have been least able to prevent opponents from making accurate shots when playing on the road, allowing over 6 of 10 shots to be placed well. So Blackburn's ability to allow only 6 of 31 shots on goal on Saturday at Stamford Bridge for a rate of .19 is a huge improvement over the club's performance this season. With 34% of possession, the final result could have been much worse than the 2-0 it turned out to be.

When playing at home, Stoke, Arsenal, and Blackpool have allowed their opponents to give their goalkeepers plenty of work to do. Arsenal is a bit of a surprise here, suggesting that Wenger is on to something when he is looking to reinforce his central defense in the transfer window. But of course, what we need to keep in mind is that a high ratio of accurate shots for opponents is one thing if we are talking about an average of 9.6 total shots Arsenal opponents take when playing at the Emirates, or an average of 12 for Stoke or 17.6 for Blackpool.

Can't wait to see how the second half of the season develops - and on both ends of the pitch.

Thursday, January 13, 2011

Value For Money? Ticket Prices and Goals in the Bundesliga

It seems that people are always looking to find value for money, and never more so than in tough economic times. Football's no different, so to help us look for value in football from a supporter's perspective, I dug up some interesting data from, the very helpful portal with stadium-related statistics. Turns out, the good folks at Stadionwelt (which translates to "stadium world"), put together a set of numbers that allow us to calculate how much fans have to shell out to see their team score a goal or two. How much, you may be asking yourself, does it cost to see a goal?

While Bundesliga ticket prices are generally quite affordable for the average fan, there is still quite  abit of league-wide variation in how much you can expect to shell out to see what some think is the best league in the world. On average, Hamburg and Cologne matches are most expensive, with average single tickets priced at €38.75 and €38.50, respectively, while Mainz and Frankfurt (€24.75 and €24.25, respectively, are a real bargain in comparison). So, given this variation, and given variation in teams' scoring prowess, how much does it cost fans to witness a goal when they go to the stadium? Below, there's first the price of a goal - calculated as by the average ticket price per match (in euros) times the number of games played in the stadium divided by the number of goals scored. So it's simply the relationship between the average ticket price in a club's stadium and the number of goals those in attendance got to see. To make things easy to digest, I then put these numbers in a bar chart, ranking "goal prices" from most expensive (worst value) to least expensive (best value. Here's what you get.

Cologne's truly dedicated fans are willing to pay the most to see the occasional goal. In Cologne you have to pay almost 35 euros to see your team score. And the difference to second most expensive Moenchengladbach - another club with long-suffering fans, especially this season - is a cool €10. Contrast that with Hoffenheim, Hannover, Kaiserslautern, Dortmund, Leverkusen, Bremen, and Wolfsburg, all of which serve up a goal for only about €15 a piece. These numbers seem to suggest that value for money is better when fans go to see good teams play.

But is it also the case that good teams provide equally good value for money, and that bad home teams' goals are equally expensive? The quick answer is no. Take a look at a simple scatterplot of the total number of goals scored by home teams and the average ticket price in that stadium (for the technical geeks among you, I've superimposed a regression line).

There is a positive correlation between total goals scored by the home team and average ticket prices, as shown by the regression line (the Pearson correlation is a moderate and positive .29). So, when you pay more, you also see more goals. Moreover, this trend line would be much steeper, and the correlation therefore much stronger, if we eliminated the "expensive" but underperforming 1.FC Koeln (whereas, with M'Gladbach you get what you pay for).

But the more interesting point about the pattern of data shown in the graph is to see which teams over- or under-perform relative to ticket prices. To make this more obvious, I have labeled a few of the teams. Teams below the regression line are more expensive than the average - you have to pay more to see goals - and teams above the regression line are a bargain - you see more goals than you pay for league-wide. As you can see, Cologne is truly woeful - way below the regression line - and their fans are paying through the nose to see their team score. But even Frankfurt with its relatively affordable ticket prices isn't scoring as much as they should, given what they charge. In contrast, Bayern and Stuttgart (despite the former's relatively mediocre season and the latter's terrible season) are giving their fans their money's worth, outperforming the league in goals scored relative to average single-day ticket prices. But perhaps the best value for money is Schalke 04 in the upper left corner of the graph: at Schalke, you pay some of the lowest prices, but you get to see more goals. Now, that's value for money.*

* These data do not show the relationship between goals and season ticket prices. I may get to that in a future post.

Wednesday, January 12, 2011

Conversion As A Measure of Offensive Production: The EPL Season at the Halfway Mark

Now that the transfer window is open, lots of managers are looking to strengthen their squads. Managers like Everton's David Moyes seem typical; he's looking to sign a striker, having repeatedly bemoaned his team's lack of ability to finish in front of goal, most recently after the away draw at West Ham. Alex McLeish, too, is looking to strengthen Birmingham City's front line to convert good offensive play into actual goals scored. Even Carlo Ancelotti is talking about needing some help up front for Chelsea, where Nicolas Anelka has been having a slow season and Didier Drogba is not in great form.

One way for the numerically-minded fan to see if these managers are correct in assessing their team's weakness as the lack of finishing is to take a look at a team's conversion rate for the first half of the season. Conversion rates are conceptualized as a team's ability to turn accurate shots into goals; it is measured as the ratio of goals scored to shots on target. I think this is a particularly useful measure of offensive production because it goes beyond simply counting how many times a team got anywhere near the other team's goal or just got a shot off - many of these incidents never pose a threat to the other team. Instead, this measure indicates how threatening shots on goal were - were they mostly weak and badly aimed balls, or was there more to them? To help with this, I have calculated teams' conversion rates as a key metric of a team's offensive production. Take a look.

The average conversion rate in the EPL this season is .227 (or about in 1 goal in 4.5 accurate shots). But clearly, there is also wide variation in conversion rates, as I have reported in previous posts. This season is no exception. Conversion rates range from a low of slightly above .1 (1 goal per 10 shots on target) for Wigan to slightly over .3 (1 in 3) for Blackburn and Newcastle. These latter teams have done exceptionally well once they find themselves in front of goal and are able to get an accurate shot off. The Top 5 teams in conversion so far this season are (in order): Blackburn, Newcastle, Sunderland, West Brom, and Man U and Stoke. The Bottom 5 are Wigan, Fulham, Everton, Aston Villa, and West Ham. So, based on these numbers, Moyes is right to look for reinforcements, as would be the managers of some of these other teams. In contrast, Birmingham's .25 conversion rate is actually slightly above average. So for Birmingham, the problem there does not seem to be conversion but getting the ball in front of goal. Evidence for this is that Birmingham's average number of shots taken per match (including inaccurate ones) is 9 so far this season, while the EPL average is 12.4.

So does it matter to teams' conversion rates if they are playing at home or away? The answer to this question is slightly surprising. It actually does, but in unexpected ways (at least to me). Take a look at the following graph, which lists conversion rates per match for home and away matches.

While average conversion rates are essentially identical at home and away (.230 v. 224, respectively), the data show that there is greater variation in conversion rates when teams teams play away from home than when they play at home. That is, the lowest and highest conversion rates can be found in the away panel of the graph. Aston Villa and West Ham's woeful away performance of .099 are the lowest conversion rates in the Premier League this season, but West Brom's .38 conversion rate, achieved when playing away from home is the best in the Premier League this season. Part of this has to do with the fact that West Brom have scored about as many goals at home as they have away (1.375 v. 1.444 - slightly more on the road, actually), but they have needed many more shots on goal to get there at home than on the road (15.1 v. 10.4). And Everton's woes are more pronounced at home than on the road, interestingly, while the teams that need the most help in this department clearly include Wigan, and Fulham.

More to come on defensive production soon. Enjoy the second half of the season.

Saturday, January 8, 2011

The Biggest Draws in 2010-11: Matchday Attendance in European Leagues

Which are the best soccer teams or leagues in the world? Of course we can see which teams win more matches, and for leagues, UEFA's league coefficient allows us to answer that question with a fair amount of precision. Less systematic but popular ways of determining league or team quality involves looking at international head-to-head competition, or evaluating which leagues attract the best players and managers. All are tried and true ways of answering the perennial question of who is best.

Here's another, and slightly different, way of comparing leagues and teams: how much excitement do they generate among supporters? If "better" football - however we define quality - is more enjoyable and therefore valuable to supporters, we can measure the excitement of supporters by quantifying how willing they are to part with their hard-earned cash to see a club play. (As an aside, this approach should also have intuitive appeal to those who think soccer is a business.)

So here are some interesting numbers on matchday attendance I came across a few days ago, from a source called Stadionwelt. They show average attendance for home matches for the Top 30 draws in Europe so far this season (on average, teams played between 8 and 10 games at home). I took these numbers and turned them into a bar chart for easy reading and comparison (the x-axis shows attendance in thousands).

So what do the numbers tell us?

Thursday, January 6, 2011

Becks By the Numbers: Quantifying the Beckham Effect on MLS and the LA Galaxy

I'm not typically a big fan of trying to quantify an individual player's contribution to a team's success absent lots of information about the team the player plays on or plays against - that is, lots and lots of good statistical control and plenty of observations. But I thought these journalistic efforts by Leander Schaerlaeckens for ESPN (and Lawrence Donegan for The Guardian) were well worth sharing. I found them particularly apropos in light of Tottenhams' keen interest to sign David Beckham to a short-term loan deal before the start to the new MLS season on March 15.

Among the more interesting findings about Beckham's contributions to his team and the league:

Games played
Regular season games: 48 of 106: 45.3%
Minutes on the field: 3,855 of 9,540: 40.4%

Winning percentage
With Beckham: 18-18-12: .500
Without Beckham: 26-17-16: .576

Goals scored
With Beckham: 1.69
Without Beckham: 1.27

Goals conceded
With Beckham: 1.73
Without Beckham: 1.12

Galaxy Attendance
Pre-Beckham (2006): 20,813
Beckham era: 20,008 (2008), 20,416 (2009), 21,436 (2010)
Pre-Beckham: 19,929 (2006)
Beckham era: > 28,000 (2007, 2008), 18,525 (2009).

MLS Attendance
Since 2007, the league's average attendance is down 95 fans per game, from 16,770 to 16,675.

So what are we to make of these numbers? Looking only at the cold, hard facts, the evidence is decidedly mixed - at a minimum, they don't allow anyone to claim that the Beckham experiment has been an unadulterated success. Beckham has played fewer than half the games and minutes the LA Galaxy has been on the field, and it's not clear that they have performed better when he has been active. And while the Galaxy were a real draw when Beckham first started playing for them, this, too, has tapered off since.

But before we jump to the conclusion that having Beckham join the Galaxy and the league has been a failure, it's important to remember that there are some things you can't quantify as easily; things, that you may not see in these numbers.

For one, as Zach Slaton from A Beautiful Numbers Game pointed out to me, numbers related to team performance with Beckham in/out of the game are going to be skewed by the awful early years he experienced on the Galaxy, and his absence during the two best years. Clearly, the numbers penalize him for what was truly awful team management those first two years (but if you believe Grant Wahl's reporting in The Beckham Experiment, this may have been Beckham's own doing, at least in part). But thankfully, Bruce Arena, and his ability to build a viable defensive core, was the key to turning an awful team into a perennial championship contender.

There's another thing that may not be obvious. As Schaerlaeckens and Donegan point out, it's worth asking whether the Beckham transfer has made MLS a more credible soccer league or has made the Galaxy a better team. My hunch is that the answer is a qualified yes, and a definite yes if you ask Thierry Henry or Arsene Wenger (or Bruce Arena), if only because of the knock-on effect that having a Beckham on your team has on the rest of the squad (through his presence at training and with the team, he makes others work harder, be more professional, and play better) or the league (his presence makes it more likely that other top players not completely past their prime will think seriously about a transfer to an MLS team). In fact, I believe that Beckham's continued ability to attract the interest of some of the best clubs in Europe, even if only for a few weeks' worth of playing time, is actually a good thing. Not only does it speak to Beckham's own ability and desire to play at the highest level or his global and local appeal as a player and soccer symbol, but it also means that there's some pretty good football being played in America. And if we know nothing else, it's clear that we haven't seen the last of Becks.

Wednesday, January 5, 2011

Elections, FIFA Style: Sepp Blatter's Campaign

He's off and running. Looks like Sepp Blatter has kicked off his reelection campaign for the FIFA Presidency to be held in June. Only, it's unclear what kind of entity FIFA really is. It's not a country (its reach is global), it's not a state (it doesn't have an army), it's not a democracy (despite the elections to the body's Presidency), and it's not the Vatican (despite the lovely idea of a "year end message", in which the FIFA chief boldly states that he looks "forward to meeting the representatives of the member associations there in person, addressing their concerns and, together, setting a strategic course for the future.") It's sort of like an INGO - an International Non-Governmental Organization, except that it doesn't meet the World Bank's definitional standard of pursuing "activities to relieve suffering, promote the interests of the poor, protect the environment, provide basic social services, or undertake community development" (in fact, you could argue the opposite in the case of some of its activities, like the World Cup in South Africa - but that's a rant for another day).

In any case, it's going to be a busy spring election campaign for Sepp, filled with the hardships of staying in unfamiliar hotels and eating exotic foods. The campaign to garner a majority of the 208 votes at stake at the June 1 meeting in Zurich will take Blatter literally all around the world. His quest for a fourth four-year term starts off in Doha, Qatar in early January for the vote rich 46-nation Asian Football Confederation; then later that month, the campaign moves on to Pago Pago, American Samoa where the smaller Oceania body will hold its meeting (only 11 votes). In February, he'll trek to the the critical African confederation meeting in Khartoum, Sudan - critical because of the 50+ votes at stake, and an important constituency in Blatter's previous election (when he promised to bring the World Cup to Africa - a promise he kept, after all). March brings Blatter to more familiar environs in Paris, where UEFA meets that month, before Blatter's reelection effort concludes at meetings of the South American (CONMEBOL) and Central and North American (CONCACAF) football federations by the end of May.

And wouldn't you know it? Blatter has identified his number one campaign issue that he hopes will persuade electors to grant him another term: corruption! After an intense year filled with accupation of bribery and the actual banning of two FIFA ExCom members from voting on the 2018 and 2022 World Cup bids, it seems that Blatter wants to become FIFA's Mr. Clean. In an interview with the Swiss SonntagsZeitung Blatter boldly declared that he would personally ensure that there will be no (more?) corruption in FIFA. He is working on a "project" that he will present at June 1 FIFA Congress that will create a new committee tasked with ensuring "compliance" (not clear with regard to what or how, but, hey it's a campaign slogan, not actual policymaking).

So far, it's not much of a race. As in the last "race", so far there are no challengers. If any are planning to step forward, they have to hurry up since they have to be nominated by a FIFA member nation at least two months before the vote. So it could be smooth sailing for Blatter, but that's not guaranteed.

The campaign issue Blatter has chosen has some inherent risks: it's could reveal him as a hypocrite, and that's never a good thing. Just yesterday, Bayern Munich's President Uli Hoeness - a powerful man in European soccer - made no secret of his distaste for Blatter's leadership of FIFA in an interview with Germany's Sport Bild newspaper. Hoeness' criticism? It's pretty simple: Blatter has lost control, FIFA decisionmaking is corrupt, and Blatter cannot be counted on to clean up the mess he helped create. The solution?* Have UEFA throw its weight around and make FIFA more transparent and accountable. It would be an easy trip for Hoeness to help clean up FIFA - it's only about 200 miles from Munich to Zurich.

* This should sound familiar, as I've noted this in a previous post.

Monday, January 3, 2011

The Worst Defensive Teams in the Big Leagues: The 2010-11 Season To Date

Which teams are the worst (and best) defensive teams in the big leagues of European soccer? Let's take a look at defensive performance so far this year - since it's roughly the halfway mark everywhere, these numbers should be more or less comparable across teams and leagues (assuming that the leagues are similar in important ways, which we know to be sorta true from previous analyses I've posted here ...).

I'm measuring defensive production in a very straightforward way by calculating the average number of goals conceded per match (as of Dec.23). Below you can see the number of goals conceded at home (graph 1) and away (graph 2). Take a look.

As in offensive production, there is considerable variation across clubs, from fewer than .5 goals conceded per home match to almost 2.5 per match (the average is 1.12 across all teams).

Here are the extremes: Only four teams have conceded more than 2 goals per match at home, on average (so about twice what the average team conceded). These include