StatsBombIQ StatsBomb Data
November 2, 2018

Milan Are Finally (Almost) Back

By Mike Goodman

AC Milan have spent a long time in the proverbial wilderness. The storied Italian club hasn’t won a league title since the 2010-11 season. They’ve only won two total since the turn of the century. They haven’t qualified for the Champions League since 2012-13. The fact that they finished sixth the last two seasons is actually a sign of improvement for a club that spend the three years before that mired between seventh and tenth. Currently, however, Milan sit in fourth place, ahead of Lazio on goal difference, and it’s just possible that they might have finally turned the corner.

The place to start, as always, is with their underlying numbers. Milan’s are quite good. Their expected goal difference of 0.59 per match is the fourth best in Serie A. They trail only Juventus and Napoli, the league’s two dominant teams, and, strangely enough, Atalanta who are mired in 15th place despite an xG difference of 0.64 per match (a brief aside here to say that Atalanta’s misfortune is truly impressive, and can’t all be laid at xG’s feet either. Even their plain old regular goal difference of positive five is the seventh best in the league. Sometimes when things don’t go your way, they really don’t go your way).

Breaking it down further, Milan’s attack is clocking in at 1.60 xG per match, the fourth best total in the league, and their defense allows exactly 1.00 xG, sixth best in Serie A. This paints a picture of a team with a pretty good attack paired with a defense that’s just good enough. The story here is also one of shot volume, not shot quality. They take 18.40 shots per game, the third most in the league, and allows 11.60, the fourth fewest.

An initial look at their shot chart might show cause for concern. There are not exactly a lot of premium attempts here.

That’s an awful lot of blue. And Milan’s xG per shot really isn’t anything to write home about at 0.09. It’s smack dab in the middle of the Serie A, 10th best. But,nobody in Italy really takes great shots. Not a single team has over 0.11 xG per shot. Everybody is cranking the ball from distance, or crossing it in for the wing and hoping for the best. The teams that excel just manage to pile on the volume advantage. Milan’s shot selection may not give them an advantage, but it doesn’t particularly disadvantage them against their rivals either. It just makes them normal.

Given that Milan are wracking up those shots, it doesn’t hurt to have Gonzalo Higuain leading the line. Sure Higuain is 30, and he’s not going to be the secret to Milan’s long term success. He’s also mostly a continuation of a troubling Milan trend of paying for older players rather than shelling out for contributors that still have their best days in front of them. But Higuain remains very very good. He’s ninth in Serie A with 0.46 xG per 90 minutes (and four of the eight players above him play for Napoli or Juventus).

Outside of Higuain, however, the team’s lack of ability to create good shots becomes clear. The players that take the most shots like Suso and Hakan Calhanoglu are different from the ones who take the best shots, mostly just Diego Laxalt.

Defensively Milan are a very interesting side. They’re one of the most conservative teams in the league. Opponents complete 83% of their passes against Milan, only Chievo, Frosinone, and Parma, otherwise known as by far the worst two teams in Italy and Parma, are easier to pass against. The average distance from their own goal of the defensive actions they undertake is 43.16 yards, the third most conservative average in the league. Their PPDA (passes allowed per defensive action, a measure of how easy it is for opponents to complete passes in their own half) is 13.13, the sixth highest (or most permissive) in Italy. When the opposition has possession, Milan don’t worry about taking it away, they just worry about making sure they get men in the right places behind the ball to not get broken down.

The one wrinkle, which their defensive action heat map makes clear, is that they do have a brief counterpress installed to muddy up the works around their opponent’s penalty area. In effect, when they lose the ball, they make sure they have enough defensive presence to stop counters before they start and give the team time to set up their own defense.

Putting the two halves of Milan approach together can make for an odd dynamic. Milan’s matches have a lot of passing. They play 587 passes per match, only Juventus, Inter and Napoli play more. On the other hand they allow 529 passes a total that’s more than all but four other teams. Only Napoli play matches with more passes in them than Milan do.

In effect, Milan’s bet is that they can turn all the passes they play into shots at a much faster rate than their opponents can. Win that battle by enough and the fact that those shots are basically low value doesn’t matter. One of them is bound to go in eventually. That’s a bet that has paid off so far this season, and there’s no reason to think it won’t continue.

Interestingly, while there’s reason to believe Milan will continue to be successful, there’s also reason to think that their matches won’t be quite so high scoring going forward. Their goal difference of positive six is almost exactly in line with their expected goal difference, but they have both scored, and conceded, four goals above expectations.

Young goalkeeper Gianluigi Donnarumma has quietly had a fairly nightmarish start to his season. He’s conceded 12 non-penalty goals so far. StatsBomb has a brand spanking new post-shot expected goal model that’s specifically calibrated to look at only the shots keepers face (in other words, shots on target). Donnarumma has only faced 30 of them, and in his defense they’ve been a tough group of shots. But still 12 conceded is a good deal more than the 9.75 xG the model suggests Donnarumma should have given up. His expected save percentage is at 67.5%, his actually save percentage is currently languishing at 60%. There’a  a lot that’s still unknown about keeping ability, and just how much keepers can and will differentiate from an expected save percentage over the long run is an open question. But, for now, what this information tells us is that Milan’s young, highly touted keeper has been a poor shot stopper so far this season.

Assuming this all normalizes, what we’d expect to see from Milan over the rest of the season is a competent, but somewhat more boring side. They’ll likely score fewer goals, but they’ll also likely concede fewer. That’s more than enough to keep them in the race for fourth place. Given where this side has been over the previous five years that’s an invaluable step in the right direction.

Article by Mike Goodman