Comparing Leagues by Points and Match Outcomes

I was curious about whether or not there was any difference in the relative frequencies of outright wins and draws in La Liga and the Premier League. So I looked at the total number of points earned in each league for entire seasons. A match with an outright winner contributes 3 points to the total whereas a drawn match contributes only 2 points. So a season with more draws would have lower total points than one with less. This is what the total points look like for the last 10 seasons:

SeasonLa Liga Total PtsPremier League Total Points
2004-20051,040 1,030
2005-20061,035 1,063
2006-20071,042 1,042
2007-20081,053 1,040
2008-20091,057 1,043
2009-20101,045 1,044
2010-20111,061 1,029
2011-20121,046 1,047
2012-20131,056 1,032
2013-20141,054 1,062
Total10,489 10,432

Over these 10 seasons, the total points earned by La Liga teams was 57 more than that earned by Premier League teams. Or to put it another way: a typical season in Spain seems to feature 6 more games with outright winners. 6 out of 380 (the total number of games played in both of the leagues) doesn't seem like much but looking at the breakdown by year it does seem as if the gap is pretty consistently in favor of the Spanish league.

Anyway, as soon as I was done doing this, I realized that I had overlooked the obvious: the number of wins/draws is available right there in the same league tables that I was using. I could just look at those numbers directly. I didn't feel very confident drawing any conclusions from the numbers so I plotted the graph below to see if the visualization would imply anything.

As far as I could tell, neither of the leagues seem to have any trend of note.

Another thing I was interested in was how the number of points earned was distributed amongst the 20 teams in each league. The plot below shows the mean number of points by position for the 10 years of data that I used.

Apart from the first three places, the La Liga curve does appear to be flatter and the lower-ranked teams seem to earn more points than their counterparts in the Premier League. Perhaps this is an argument for La Liga being more competitive. I'm going to ruminate on if this actually means anything.

Something else I saw when I was scanning the data is that on occasion some teams with low Goal Differences did surprising well. For example, in the 2004-2005 season Everton finished 4th having scored 1 goal less than they conceded over the course of the season. That seems like an unlikely way to earn 61 points. Maybe looking at Goal Difference for each ranking might reveal some difference between the leagues.

Well ... the lines do seem to cross more in the middle of the table and the numbers do seem to corroborate the idea that the teams in the lower rungs of La Liga might be more competitive than those in the English league. And that reminded me of something else I'd learned about recently: Pythagorean Expectation. Runs are just like goals right? This concept should transfer very easily. So I plotted that as well:

This looks very similar to the plot of Goal Differences. I'm mildly intrigued by the fact that the ordering of the leagues is flipped for 1st and 2nd places. I'll have to track down some literature and see what baseball people do with this statistic.


I'm not sure I'm willing to make any strong pronouncements on anything here. I wonder if there's anything further to the fact that the bottom teams in Spain seem to earn more points. Maybe I'll look into that further some other weekend.


I put the data I used on github. Here:
Premier League Historical Data
La Liga Historical Data