Recently, there’s been a contentious discussion over at the Juventus SB Nation blog, regarding the best way to win points – whether that comes from attempting to score more or attempting to limit the number of goals conceded. The entire discussion started with the best way to win titles, which quickly turned into a discussion of winning points (as winning the most points is the only way to win titles). In an attempt to come to a conclusion as to whether there was a best way, I ended up with this analysis based on the results of Serie A matches going back 10 seasons.
In that, I found that so long as both aspects of a team are above average – that is, so long as they’re scoring roughly 2 goals and conceding roughly 1 – there isn’t a benefit to taking either path. There becomes a benefit to leaning one way or the other, as production in one area dips below average, but for the most part – especially among title contenders – there really isn’t a benefit. Due to there being a lack of benefit – given these parameters – I also posited that the Serie A title winners also being the team with the least goals conceded was merely a correlation – likely due to the footballing culture in Italy.
To test this theory, I decided to look at how teams across the five major leagues (EPL, Serie A, La Liga, Bundesliga, and Ligue 1) performed based on their goals forced and allowed – independent of each other. To do this, I looked at a weighted average of points won in matches based on goals allowed and goals forced. This shows how likely a team is to take all three points, based on the different scenarios – regardless of how they do on the other side of the ball.
As you can tell, there’s no serious difference between the leagues. The standard deviation for each goal scenario (both forced and allowed) is below 0.10 points, indicating that the value of goals is largely the same across the leagues, in terms of points. All this indicates, however, is that there’s largely not a difference between Serie A and the rest of Europe. It doesn’t tell us whether defense really does cause teams to win titles, or whether Italian champions are consistently the best defense for some other reason.
For this, I looked at the rate at which the best defense (defined, for this purpose, as the team that conceded the fewest goals throughout a season) also won the league titles, in other leagues.
As you can see here, the Bundesliga’s best defense also wins the title as a high rate, but no other league “predicts” a title winner at a rate better than 60%. Given the path towards winning points in all five leagues is roughly equal, and the lack of connection between the best defenses and title winning teams, it would seem any sort of relationship between style (i.e. most goals scored, least goals conceded) is more of a coincidence than anything. While there may be teams structuring their team with the belief that offense or defense wins titles, it seems as though any theory along these lines is off the mark.