Pythagorean observations at end of 2012-13 European season

The 2012-13 European football season came to an end this weekend, at least in the top flight.  It won’t be officially over until the first of July, but it’s time to take a look back at the season and examine which teams were significant performers (or underperformers) relative to their statistical expectations.  And once again, we look at the Pythagorean tables for some insight.

Here are some of our observations from the season:

  • Bayern Munich is the most dominant side in Europe since Barcelona ’09.  Bayern stood out from other league champion sides in Europe by the combination of their exceptional performances and the small amount of luck surrounding them.  For most of the season, Bayern’s Pythagorean residual was at 0 or +1, which is very rare for a team with so large of a margin from second place.  They finished with +5, but the 25-point gap between themselves and Dortmund was real.  And now they’re the seventh team to clinch a league-cup-European Cup treble.  I’d say that such a team doesn’t come around very often, but given that we’ve had three Treble-winning sides since 2009, another one might be around the corner!
  • Manchester United and Barcelona significantly overperformed.  At the three-quarter mark both sides were performing at 10 and 14 points above expectations, respectively.  They finished the season with Pythagorean residuals of +12 and +13, which matches Twente’s +13 residual in the 2009-10 Eredivisie.  As I stated at the three-quarter review, United’s and Barça’s defenses weren’t that spectacular, and this is particularly apparent when you view Ben Mayhew’s defensive shot visualizations, but they benefited from sub-par performances from their competitors and high-scoring offenses (Van Persie for United, Messi for Barça).  Will they see a drop in performance as Twente did?  Both are facing transition periods in the next two seasons, but I expect Barcelona to remain on top during that time.
  • PSG and Juventus were also dominant in their leagues, but not on a Bayern level.  The Ligue 1 and Serie A winners had Pythagorean exponents of -1 and +3, respectively, right in that 3-point RMSE range of the Pythagorean.  One critique of the Pythagorean is that it tends to regress teams toward the mean, so teams at the top or bottom are performing much better or much worse than expected, but in both leagues the winners were expected to finish 9-12 points clear, which they did.  One explanation for that could lie in their defensive record.  If you consult Mayhew’s defensive visualizations you see that PSG and Juventus allowed fewer shots to be converted into goals than any other team in the Big Five.  We don’t know much about the nature of those shots — were PSG’s and Juventus’ shots/goals conceded ratios so high because opposing teams were poor shots, or because they were forced to take low-percentage shots, or because of an excellent ‘keeper?  Over a domestic season it might be important to have a high shots conceded to goals conceded ratio, but in European competition it might be important to allow very few shots.
  • How did PSV lose the Eredivisie?  PSV accomplished something I haven’t seen since Ajax in the 2009-10 season — they scored over 100 goals and had a goal difference of +60, yet they finished second.  And they weren’t pipped to the post; they lost by seven points and only goal difference kept them in second place.  Yet until the final weekend PSV were expected to finish in first place.  Only Twente’s 3-1 defeat of PSV combined with Ajax’s 1-0 win vaulted the champions over the Eindhoven side in the Pythagorean table.
  • Did Manchester City overperform this season?  It would be very tough for fans and the media to agree with that, but their Pythagorean residual of +6 does indicate that they performed very well for a team scoring 66 goals and conceding 34.  But the fact that they scored 20 fewer goals than Manchester United was the bigger issue.
  • Augsburg the least bad of the Bundesliga’s bottom three.  Augsburg’s expected point total would have sent them down to 2.Bundesliga but only the underperformances by Düsseldorf and Hoffenheim lifted them into safety.  And if Lewandowski had not been adjudged to have been in an offside position, Hoffenheim would have dropped down directly.  Greuther Fürth aside, relegation was a toss-up in this season’s Bundesliga.

Looking at the leagues on the ResultsPage app, the big overachievers for the season (other than Barcelona, Manchester United and Manchester City) are: Marseille (+14), Rayo Vallecano (+10), Hamburg (+9), Olympiacos (+8), Feyenoord (+7), Stuttgart (+7), Utrecht (+7), Groningen (+7), Tottenham Hotspur (+6), Galatasaray (+6), and Elazığspor (+6*).  (The asterisk means that the expected point total would have placed the team in the relegation zone.)  Big underachievers are Mersin (-11), AZ Alkmaar (-8), Liverpool (-8), Reading (-8), Orduspor (-7), Saint-Étienne (-7), Évian (-6), QPR (-6), Torino (-6), and Palermo (-6).

And of course, you can check all of this out yourself at ResultsPage, our league results and table application (and guinea pig).   We’re tracking the other major spring-to-fall leagues there, including MLS, the J-League, and the Brazilian championship.


Tags: , , , , ,