Lamine Yamal is so good he defies advanced stats

If you didn’t watch last night’s UEFA Champions League semifinal between Barcelona and Inter Milan, you should probably catch up on the highlights. And even if you did watch the match, you should watch these again, because they’re some of the best highlights you’re going to see at any level all season.

(UK version embedded, US version here, sorry to other countries that UEFA is weird).

I, like everyone else, thoroughly enjoyed the performance of Lamine Yamal. He scored a cracking goal, set up several more chances, and hit the crossbar with an audacious chip in the 87th minute.

Ball-knowers like Alex Barker and non-ball knowers like Piers Morgan’s son were in total agreement with their hyperbolic superlatives. No one could contain themselves. We were watching a genius.

I woke up today, made my tea and went for my morning walk thinking about Yamal’s performance, and what I might want to say about him if he was the subject of today’s newsletter. When I sat down at my computer and fired up the StatsBomb platform to look into the numbers behind last night’s game, I was greeted by an utterly horrifying sight.

Folks, the possession value models say that we do not know ball. None of us. We are all stupid. Lamine Yamal was Barcelona’s worst player!!! He was credited with -0.19 OBV, with his on-ball contributions apparently harming Barcelona’s chances of scoring goals.

You might be wondering how that’s possible, and if there was some kind of data collection error. I don’t blame you, I let out some expletives when I saw it. But it’s not that crazy. For all of Barcelona’s dominance — 72% possession and 622 completed passes — they generated a very average 1.33 xG on the night. Their 0.06 xG per shot was also pretty meh; both numbers are well below Barcelona’s competition average.

Yamal’s goal was a world class finish on a low percentage chance, and the own goal came via a lucky bounce off a low xG rip by Raphinha. The Ferran Torres goal was the only big chance Barcelona had all night.

My absolute favorite moment of the match is actually one of the best examples of how Yamal might have been given a negative OBV on the night. Here he is at the moment he decides to attempt a chip of the keeper in the 87th minute. It’s one of the most audacious shots I’ve ever seen, and he only missed scoring the winner by an inch.

A possession value model like OBV (or Karun Singh’s xT, or American Soccer Analysis’ G+,) will have assessed this shot attempt as a poor decision, and one that significantly decreased Barcelona’s chances of scoring a goal from this move, given the positioning of Yamal’s teammates and the Inter defenders. The best thing he could have done, from a pure numbers standpoint, is probably play a quick return pass to Torres’ feet. Second best would be to try dribbling into a better position. Third best would be a cross towards Raphinha at the back post, though that instantly vaults to No. 1 if he actually finds his teammate’s feet perfectly. 4th is what he actually did, which would not have changed even if he scored.

None of this is to say that Yamal is overrated, or that people who think he was awesome last night or wrong, or that trying to score insane goals when he could have passed should be coached out of him. I don’t believe any of that. One, oh my god just watch the games and see how good he is, you absolute nerd. Two, he’s 17 years old. Three, we have plenty of evidence that he does actually make the statistically optimal passing decision more often than not.

This is just to say that sometimes the stuff we absolutely love to see, the most enjoyable football imaginable, is so extremely far from the most objectively optimal football.

I could have written this same post about some of Lionel Messi and Cristiano Ronaldo’s most iconic performances. These players were ruthlessly efficient statistical marvels who generated the highest possible value scoring chances for themselves and their teammates 9 games out of every 10, consistently for an entire decade. But the games we remember and will tell our grandchildren about are the ones where they made “bad” decisions and scored insane bangers out of absolutely nothing.

Saying that this was actually a below average performance for Lamine Yamal is not hating, it’s praise. That’s how good this dude is. He’s so good that even when he makes poor decisions, he’s a serious threat to score and assist. His technique on those strategically sub-optimal shots, passes and dribble attempts is so out of this world that people who really know their stuff can’t tell the difference between those ones and the optimal ones.

If you enjoyed this newsletter, we’d appreciate it if you would forward it to a friend. If you’re that friend, welcome! You can subscribe to The Transfer Flow here. We also have a podcast where we go in depth on transfer news and rumours every week. We’re on YouTube here, and you can subscribe on Apple Podcasts or Spotify by searching for “The Transfer Flow Podcast.”