European Goalkeeper Stats 2010-11

The following is a list of (barely) advanced Goalkeeper stats for the four major European leagues for season 2010-11.

European Goalkeeper stats 2011-12

This post details some (barely) advanced statistics for goalkeepers from the 2011-12 European football season.

European Goalkeeper Stats 2012-13

The following post is essentially a list of numbers, from which you can make what you will. It details each goalkeeper's individual performance over the course of the season - and, by extension, the performance of each club's defence over the season past.

Click past the byline to see the (rather lengthy) statistics.

New goalkeeper stats available

After literally an two and a half hours of data collection and limited analysis, we've manage to update Balanced Sports' European Goalkeeper Stats page.  It now includes all of 2011-12 and stats from the first half of 2012-13.

Thanks must go to Blogger, whose algorithms for some reason stopped screwing up the spreadsheet inserts.

Debate away!

Updated Goalkeeper stats available

After some happy-fun-time data entry, we can now present you with a fully updated list of European goalkeeper stats, including all players who started a game in the four major European leagues this season.  You can get there by clicking the link below.

European Goalkeeper stats, 2011-12 season

Balanced Sports also offers other collections of stats, like Key Man analysis and Invidivual/Team efficiency measures.  Feel free to take a look at them as well.

A statistical breakdown of Premier League goalkeeping

Last year, we asked what would  happen if we applied hockey stats to goalkeepers in the English Premiership.  The results were interesting, if expected - decreased save percentages and often reduced GAA across the board.

What was more interesting was the amount of variation between goalkeepers on the same team: where David Stockdale rode an early wave of form (and a Mark Schwarzer injury) into the England squad yet wasn't able to repel the big Australian when he came back to full fitness.

This year, save percentages have decreased ever so slightly (from a league average 0.694 to 0.691) while GAA has seen a commensurate tiny increase (1.396 to 1.416).  The following tables show all the Premiership's goalkeepers arranged by GAA.  The red line marks the approximate league averages.

(c) Balanced Sports

This second table orders each goalkeeper according to their relative save percentage.   "Backup" Lindegaard again tops the league, with a save percentage slightly greater than Chelsea second-teamer Hilario and England's Joe Hart.

(c) Balanced Sports.  Please not the error in Michael Vorm's GAA in this table.  It is correct in the table above.

These numbers mean little without full context - Wojciech Szczesny's percentages still suffer greatly from Arsenal's 8-2 drubbing at Old Trafford, while Ali Al Habsi of Wigan and Tim Howard of Everton have perhaps been their club's best performers this season.  Therefore, it bears reminding that these numbers don't reflect the quality of the goalkeeper, but the quality of the defence in front of them.


For all the raw goalkeeper stats you can lay your hands on, visit our Goalkeeper Stats page.

The need for Advanced Football Stats

Balanced Sports has been published on Soccerlens, suggesting the need for easily-accessible Sabermetric football data:

The Need for Advanced Football Stats

Europe's top Goalkeepers - by the numbers

Although the quality of a goalkeeper cannot solely be judged by statistics, they can provide an insight into how well they perform. They need to perform sweeping duties much more than their NHL counterparts and often are left horribly out of position by lapses in skill or concentration by the defenders they are attempting to organise.

Amongst numerous possibilities, the two major statistical categories that NHL goalkeepers are measured by are GAA (goals allowed average) and Save percentage. These stats, though accessible aren't kept as a matter of course by football leagues throughout Europe simply because they don't really provide an accurate representation of a goalkeeper's duties - for example, Ben Foster was by most accounts Birmingham City's best player last season yet the club was relegated. They do, however, allow some analysis as clankers (like Rob Green's worst nightmare) as even the occasional mishap can affect a player's numbers.

The following numbers were compiled using raw data from ESPN Soccernet. For full details, you can visit our Goalkeeper Stats page.

Europe's top ten goalkeepers by Save percentage:

League	Team	Player	Games	Save %	GAA
Serie A	AC Milan	Christian Abbiati	35	0.853	0.54
La Liga	Barcelona	Victor Valdes	32	0.826	0.50
Serie A	Inter Milan	Julio Cesar	25	0.806	0.84
Bundesliga	Hannover	Ron-Robert Zieler	15	0.804	0.73
Serie A	Chievo Verona	Stefano Sorrentino	37	0.795	1.05
La Liga	Valencia	Cesar	15	0.781	0.93
La Liga	Malaga	Wilfredo Caballero	15	0.774	1.27
EPL	Man City	Joe Hart	38	0.771	0.87
EPL	Birmingham City	Ben Foster	38	0.765	1.45
Bundesliga	B. Dortmund	Roman Weidenfeller	33	0.764	0.64
Average	Cagliari's	Michael Agazzi		0.712

And the top ten by GAA:

League	Team	Player	Games	Save %	GAA
La Liga	Barcelona	Victor Valdes	32	0.826	0.50
Serie A	AC Milan	Christian Abbiati	35	0.853	0.54
Bundesliga	Borussia Dortmund	Roman Weidenfeller	33	0.764	0.64
Bundesliga	Hannover	Ron-Robert Zieler	15	0.804	0.73
Bundesliga	Mainz	Christian Wetklo	24	0.762	0.83
Serie A	Inter Milan	Julio Cesar	25	0.806	0.84
EPL	Manchester City	Joe Hart	38	0.771	0.87
EPL	Chelsea	Petr Cech	38	0.750	0.87
La Liga	Real Madrid	Iker Casillas	35	0.717	0.91
La Liga	Valencia	Cesar	15	0.781	0.93
Average	Levante's	Gustavo Munua			1.35

There's significant overlap between the two lists - 'keepers in teams like Manchester City played a staid, solid brand of football limiting opponents' scoring chances before counterstriking themselves leading - with his considerable skill - to Joe Hart's desirable numbers.

What this overlap suggests is either down years or the early signs of the demise for two of the best custodians in recent memory, Iker Casillas and Petr Cech. Though both had above-average Save Percentages (in Casillas' case, only barely), those numbers don't correlate well with their GAA.

GKs such as Valdes and Abbiati have excellent GAA and Save Percentage numbers. To have a great GAA could reflect on team's possession statistics (ie. Valdes), a team's defensive solidity and finally, the ability of the goalkeeper. A low save percentage means either a defense prone to playing their keeper out of position, or that keeper not stopping as many shots.

The opposite was Ben Foster, who made a million and one saves - actually, 179, fifteen more than the next most called-upon 'keeper, the aforementioned Green - yet let through slightly more than average goals per game. His save percentage is high despite his facing heaps of shots, rather than perhaps youngster (and Football Manager stud) Ron-Robert Zieler of Hannover, who only faced three shots per game.

Goalkeeping Stats across Europe - an analysis

101 goalkeepers made over ten starts in Europe's four major leagues during season 2010-12. Another fifty-six made at least one start in either the English Premiership, La Liga, the Bundesliga or Serie A. So after thirty eight games (thirty-four in Germany), which goalkeeper is best?

Of course evaluating custodians purely by stats is a fraught and inaccurate business. Goalkeepers, by their demeanour and wits bring more to a team than simply stopping shots, as per NHL netminders. They also command and organise their defence as well as arrange their players in best positions to defend set pieces. Also, they are often not at fault for the goals conceded, being beaten by lackadaisical or poor defending.

That's not to say that quantification of a goalkeeper's performance is useless. For one thing, we can evaluate definitively the differences across leagues.

LEAGUE	Total Starts	Conceded	Saves	Saves/Start	Save %	GAA
Serie A	760	953	2560	3.36	0.729	1.25
La Liga	760	1042	2645	3.48	0.717	1.37
EPL	760	1057	2403	3.17	0.695	1.39
Bundesliga	612	890	1925	3.13	0.684	1.45

GAA = Goals Allowed Average. Note the term "Start" has been used rather than "Games". Due to the data available, we weren't able to evaluate when goalkeepers conceded after coming on as a substitute. This also means that all the goals scored in a league throughout the course of the season may not be included in this table.

Perhaps as a result of a slower game tempo, of a reliance on a star individual, miserly defence or even plain and simple goalkeeping ability, Serie A produced consistently the best goalkeeper metrics across the board. That is, Goalkeepers in Italy conceded less goals per game, had a higher save percentage and produced a high level of saves per start.

Italian football is notorious for it's slower pace - unless you're Napoli or Udinese this season - and infamous for low goalscoring. That isn't necessarily borne out here - while the GAA was lower than in the other three major leagues, the increased Save Percentage counteracts that. This could also be a tactical measure, where Italian defenders are more "pigeon-holed" as defensive stoppers, eschewing forward play in favour of stolid defence.

Most surprising was the relative decrease in saves across the water in England. The Premiership produced the second-lowest save percentage and the second-highest GAA. Again, tactical considerations could be keeping these averages low or simply that the offensive play is superior. This could both back up and dispel the popular preconception that the Premier League is the most exciting and balanced in Europe, so while it's interesting data it's hardly definitive. While this table takes into account goalkeeper performance, it is we have said, impossible to divorce that totally from a defence's (in)ability.

How would these numbers look then, if we removed the backups? Most clubs have one primary goalkeeper and at least one creditable custodian in reserve - take for example Italian Champions AC Milan who fielded Christian Abbiati thirty-five times during 2010-11 and Marco Amelia during the remaining three matches. The following table has removed all the players with less than ten league starts. The numbers don't change much at all:

LEAGUE	Total Starts	Conceded	Saves	Saves/Start	Save %	GAA
Serie A	732	912	2469	3.37	0.730	1.25
La Liga	698	938	2432	3.48	0.722	1.34
EPL	714	988	2283	3.20	0.698	1.38
Bundesliga	553	795	1758	3.18	0.689	1.44

The greatest difference between the two is a minor increase in German Saves per Start and Save Percentage, while Spanish Save Percentage increased mildly as well. The GAAs, though changed in all leagues except Serie A, were minor adjustments. While it was a nice thought that a league's overall goalkeeping quality could indicate how crucial defences were to the numbers, it is unproveable with the above results.

It's perhaps telling that the greatest difference in the table above comes from Germany & La Liga's goalkeeping starts: during the course of the Spanish season, forty-three players got a start between the sticks. Thirty-eight Goalkeepers got starts in the Bundesliga (forty-two, if adjusted for a twenty-team league), thirty-nine Italian league players and thirty-seven in the EPL. While the vast majority of each team's matches were played by the starting GK (over 90% in all leagues), Germany had the most trust in their backups.

This means that backup keepers, whether for reasons of injury, form or rotation claimed 9.93% of the available starts in Germany while in Italy, starters accounted for an enormous 96.3% of the available starts.

In part two, we'll go through Europe's best and brightest goalkeepers by save percentage, Goals Allowed average and by league. In the meantime, you can find all the raw data here at Balanced Sports' Goalkeeper Stats page.