EPL 2011-12: Data analysis from the season that was (Part 1)

An open drive by the Manchester City Football Club to crowd-source ideas on the use of season data for predictions and insights has been a boost for the use of analytics in sports. As someone who follows football but is not that much passionate about the sport, I was able to come up with a few preliminary insights on the season that was – 2011 EPL, thanks to the complete Opta Sports data set that MCFC Analytics provided access to as a part of their MCFC Analytics drive.

 

Let us start with the eye-ball catchers. The ones that even pedestrian football followers can relate to – The Strikers! We know that Robin van Persie scored the maximum no. of goals in the season. But was he the most successful converter of shots to goals? No! The scatter shown below pits players against each other in terms of their shots to goals conversion.

It is significant to note that Papiss Demba Cissé has the best conversion rate scoring 13 goals off 35 shots at a staggering 37%.

To their credit, RVP and Rooney have decent conversion rates of 21% and 22.5% respectively considering they scored a lot more goals. It is also interesting to note that Torres and Suarez were not trolled for no reason. Their conversion rates are a meagre 12% and 10% respectively.

The non-open play goals come next. Though goals are often scored by most teams when on open play, some of the crucial game changers come via corners, penalties and the likes. Here’s a visual to show how the top 5 teams scored such goals.

Points worth noting are that Manchester City are the strongest from corners. Manchester United scored more than twice the number of goals via penalties than their nearest competitors. No wonder supporters of other clubs consider referees to be United-friendly! Chelsea have been pretty good by scoring through set plays.

Do fouls matter when it comes to topping the table? A close look doesn’t show so. Here’s a stack of top teams and Swansea (who’ ve been the neatest in terms of fouls).

All the top teams have conceded more fouls than Swansea (check the blue baseline in the graph) and have won a lot less fouls than Swansea! (The rankings of teams in net fouls won is shown in the bracket by the side of teams’ names). The topping clubs from Manchester: City and United, ranked 13th and 9th are among the most deviant from Swansea – showing neat play is not all that significant. The yellow line indicates the number of penalties each of these teams have won in the season by fouls committed by opposition. This clearly shows what I said above. The yellow line shoots up more than twice its height at the United Skyscraper!

What’s football without the Goal-keepers? Here is a chart depicting the number of saves the top goalies have made during the 2011-12 EPL season.

It shouldn’t come as a surprise that Wanderers’ Hennessey has top honours of saving the most shots. Being the goalkeeper of a weak team is a huge responsibility, ain’t it?! Goalkeepers of top teams are somewhere in the middle among those towers. The red line running across the towers indicates the % of saves each of those goalkeepers has made from inside the box. The squares for top teams are highlighted. One look at it shows, among keepers of top teams, Arsenal’s Szczesny had to bear the brunt of saving a much higher % of goals from inside the box than his counterparts at Tottenham, Chelsea and the Manchester Clubs.

That leaves us with the midfielders. How well have teams tackled their way through in the league last year? Here’s a look of a scatter of tackles made to tackles won by defenders/midfielders with more than 40 tackles in the season.

The likes of Lescott, Bosingwa, Formica and Fellaini come up trumps in tackle percentage whereas a relentless tackler like Cabaye hasn’t been all that successful with his tackles. Stoke City’s Whelan has been the inefficient of prime tacklers and at 66% success rate, Arsenal’s Ramsey has not done all that well. One major highlight for City this season was the combination of Lescott, Kompany & Toure all of who had tackling success of more than 80%: 91%, 80% and 82% respectively.

Some of the more interesting patterns emerged as I tried pitting style of play of teams against different teams and looked at Home & Away patterns. More of that will follow as I progress further. In the meanwhile, comments and suggestions for angles to look at are welcome! To get access to the full data-set for yourself , follow the instructions on Manchester City’s Analytics webpage here.