Why the Bundesliga had a title race
For the first time since 2009, the title race in Germany went on until the last round of play. Bayern won it in the end, of course, collecting their seventh consecutive title. Some of the underlying data suggests Bayern should have been partying at the Marienplatz a lot earlier than they did, however.
Bayern Munich amassed 78 points this season. By no means a small number, but it is a negative outlier when looking at the teams that championed the Bundesliga in the past decennium. The last time 78 points was enough to win the Bundesliga title was in the 2010/11 season, when Dortmund won the first of two titles with Jürgen Klopp at the helm. Bayern finished that season with Andries Jonker as head coach, Manuel Neuer was still Schalke’s goalkeeper and Michael Ballack played seventeen matches for Bayer Leverkusen. Sounds like a long time ago, doesn’t it?
Bayern did a lot better, except gathering points
In none of their last seven championship winning seasons, Bayern finished with a point total as low as 78. Surprisingly, Pep Guardiola’s middle season pops up as the second-lowest in this streak with 79 points. Bayern went over eighty points in all of the other seasons.
Whenever a sharp drop in points occurs, one would expect this to be caused by a drop in underlying numbers. This is not the case here. Compared to last season, Bayern took more shots per game, while the quality of those shots remained at the same excellent level. On the other end of the field, the shot volume fired at Bayern’s goal went down by a full shot per match, while the quality of these shots remained unchanged.
Somehow, despite generating slightly better offensive numbers and vastly better defensive numbers, Bayern ended up with six points less, going from 84 last season to 78 this season.
This is also clear when looking at the expected goal totals. The amount of goals a team is expected to score based on the quality of the shots they take. On offense, Bayern generated a whopping 86.5 expected goals, scoring 88, nothing to see there. On the defensive side however, their 26.3 expected goals conceded turned into 32 actual conceded goals. That’s a lot of goals to give up, especially in a close title race. Only five Bundesliga teams have have a bigger negative difference when comparing actual goals conceded and expected goals conceded.
All in all, there is solid evidence Kovač and his men performed as in an eighty-plus point season, which turned into a 78 point season because of variance. Basically, shit happened.
Dortmund mirrors Bayern
Now, for a proper race, you need two horses. Dortmund led the league as late as April, going into a direct confrontation with Bayern as proud Bundesliga leaders. After getting slaughtered at the Allianz ArenA, they surrendered the leading position to Bayern, never to get it back again.
All season long, Dortmund were hugely efficient with their chances, on a level that does not look likely to be sustainable in the future. Dortmund have scored 21.7 goals more than their expected goals total would suggest. According to our model, that makes them the biggest over performer of all top five European leagues. Take a look at Dortmund’s shot numbers compared to last season.
Despite these nuanced differences on both ends of the field – slightly worse offensively, slightly better defensively – Dortmund finished the season with 76 points, an improvement of 21 (!) points compared to last season.
Throw in two Bundesliga games Dortmund won in stoppage time leading up to the clash in Munich – at home to Wolfsburg and away at Hertha BSC – and ta-da! There’s your title race, carrying on until the final day of the season.
Dortmund’s underlying numbers are in line with just about making the top four, but their points total comes close to what the Bundesliga champions historically have.
The key is this: a 34-game season leaves a lot of room for variance, which is exactly what happened here. Bayern probably should have had more than 78 points, and Dortmund should have had less than 76. As long as seasons are being held over the span of 34 or 38 games, as is common in football, some teams may produce results that are not entirely reflective of their actual performance levels, with narratives understandably following results rather than the complex waters of underlying performance levels.
This is nothing new to those who are into football statistics. Something similar happened to the aforementioned Jürgen Klopp in his last season in Germany, when Dortmund finished seventh despite putting up solid underlying attacking and defensive numbers. However, the case of Dortmund manager Lucien Favre is a curious one, and one that deserves a closer look. We’ll be back with an article on that next week.
If you like this article, please consider a membership of Between the Posts.
This will grant you access to all our coverage of Europe’s top five leagues, the MLS, the Brasileiro and cup competitions like the Champions League, Europa League, Copa Libertadores, Copa America and many more.
Members will also get to explore our advanced team stats page. Finally, members are encouraged to post comments and interact with our writers.