⚽️ How good are Xabi Alonso’s Bayer Leverkusen?
⚽️ How good are Xabi Alonso’s Bayer Leverkusen?
My Week in Sport(s): the Bundesliga champions, All-NBA selections and Nelly Korda
Welcome to My Week in Sport(s) — a regular newsletter from Plot the Ball.
In this edition:
⚽️ Xabi Alonso’s first league title as a manager, in context
🏀 Recapping this year’s All-NBA selections
⛳️ Nelly Korda, (still) on a tear
⚽️ How good are Xabi Alonso’s Bayer Leverkusen?
When we talk about sport, I think we can be led a bit astray if we conceive of ‘winning’ as a skill in itself — rather than as a product of definable technical skills, plus a healthy dose of ‘luck’ or variance.
In football, the fundamental, repeatable skills which good teams possess are creating high-value scoring chances at one end of the field and suppressing them at the other.
Where ‘luck’ comes into it — or so the theory goes — is in how frequently those opportunities are turned into goals.
This isn’t all good fortune, obviously. There are better and worse finishers of scoring opportunities than average, and better and worse shot-stoppers at the other end — even if it takes a while for both of those skillsets to emerge in the data.
But I like the construction — winning equals skill plus variance — as a way to start thinking about dividing up credit for a team’s success.
is always a voice of reason on this topic, and he framed it very nicely in a recent ESPN piece:“Measured by expected-goal differential, the top three (Premier League) teams are all above plus-40, while no one else in the league is north of plus-20. As for the bottom three, they're all at minus-28 or worse -- no other team in the competition sits below minus-20. But if we look at shot conversion, we stop learning about how the game works. Arsenal are third, Man City are fifth and Liverpool are 13th for the percentage of their shots that become goals. The bottom three consists of Everton, Sheffield United and Brighton.”
What happens if we apply this basic model to one of the most stunning footballing stories of the last few seasons?
After their heartbreaking loss to Atalanta in the Europa League Final this week, Bayer Leverkusen won’t go unbeaten in all competitions — but they have already concluded their Bundesliga campaign without losing a match.
It’s easy to conclude at a glance that they’ve benefited from some favourable bounces along the way: their goal difference of +65 in domestic play far exceeds their expected goal difference of +43.71.
But I was interested to find out how this compared to other recent title-winners in the major European men’s leagues. How does Leverkusen’s underlying performance level stack up against other recent champions? And did they benefit from good finishing luck — at both ends of the pitch — to an unusual degree?
To explore these questions, I’ve taken the squads which won each of the ‘Big Five’ leagues in each of the last seven seasons2 and ranked them according to four metrics:
The expected-goal value of the chances they created per game, on average;
The expected-goal value of the chances they conceded per game, on average;
The number of goals they scored per game in excess of the xG they created; and
The number of goals they conceded per game in excess of the xG they conceded.
From this analysis, it’s clear that Bayer Leverkusen are far from an unusually weak European league champion.
They created chances worth 2.2 xG per game this season, and conceded chances worth only 0.9 xG — ranking their attack 13th among this cohort of 35 recent title-winners, and their defence 14th.
Not all of these title-winning squads played under the same conditions, of course — and so a bit more context is necessary.
It’s certainly notable that, by the underlying numbers, Leverkusen’s attack was weaker than each of the Bayern Munich teams which won the Bundesliga between 2017-18 and 2022-23.
But only one of those Bayern teams suppressed chances more effectively than this year’s Leverkusen — and, in the aggregate, Xabi Alonso’s side come out as the fifth-strongest Bundesliga winner of the advanced-stats era by xG difference per game.
Did they get an unusual volume of favourable bounces, though?
From plotting the second pair of rankings, we can see that only seven champions outperformed their xG by more on a per-game basis than Leverkusen’s 0.5-goal margin — while they also benefited from opponents not converting chances into goals more than most other recent title-winners.
It’s worth saying here that the best teams in European club football do tend to consistently outperform their expected goal difference on the attacking end; Manchester City, for instance, have beaten xG by a considerable margin in each of their six title-winning seasons since 2017-183.
But I am skeptical that this outperformance by Leverkusen at that end of the field will be repeatable — purely because of the way it was spread across their squad.
It’s not the case that they have a number of superstar shot-getters who are also excellent finishers; in fact, their most effective striker — Victor Boniface — actually scored fewer goals than his xG implies he ‘should’ have this season.
Rather, eight different players exceeded their xG total by at least one goal in 2023-24 — and none of them produced more xG than Florian Wirtz’s total of 8.9.
While some of those players — like Álex Grimaldo, who scored 10 goals on 5.7 xG this year — do show some evidence of real finishing skill, benefitting from all of that upside at once feels unlikely to happen again to the same degree.
At the other end of the pitch, it is probably even less likely that goalkeeper Lukáš Hrádecký — who had conceded comfortably more career goals than he ‘should’ have in aggregate coming into 2023-24 — suddenly took a leap at age 33; we have noted before how volatile goalkeeper performance can be from year to year.
Heading into next season, then, there are some signs that regression is coming.
But the fact that they’ve managed to hold onto Alonso as manager — despite interest from two of his former clubs, Bayern and Liverpool — shouldn’t be underestimated.
Although looking at season-long aggregates is an efficient way to put team performances in context, in this particular case it obscures the considerable improvement in Leverkusen’s underlying numbers which Alonso oversaw as the season progressed4.
When teams go on long winning streaks, it’s easy to think that there’s something special they’re doing which isn’t being captured in the data. Wednesday’s Europa League Final was a sobering reminder for Leverkusen — and Alonso — that ‘invincibility’ in top-level sport is a myth.
But, over the long run, football teams more or less perform to the level of their underlying numbers — and the club should still be positive about their chances of repeating as Bundesliga champions, given the trajectory that their manager has them on.
Even if Alonso is destined to move on to bigger (and better-remunerated) things, this could well be the defining stretch of his career. As Rory Smith of the New York Times put it yesterday:
🏀 Run the Numbers
When we last looked at the geographical diversity of annual All-NBA selections — the groups of players deemed to be the elite of the men’s basketball elite each season — it was clear that the league had changed markedly over time:
The three All-NBA teams for 2023-24 were announced earlier this week, and only four ‘international’ players5 were selected — down from five last season.
The absence of one of last year’s selections from overseas6 was down solely to the league’s new awards-eligibility criterion, though. Joel Embiid — whose Box Plus/Minus of +11.6 points per 100 possessions was bettered only by Nikola Jokić among players on the court for as many minutes as he was — didn’t play the required 65 games this year.
Despite his inexperience, French rookie Victor Wembanyama received a number of votes for the second and third teams too — and might feel that he actually deserved even more consideration than he got7.
Using Basketball Reference’s BPM as a proxy, his on-court impact was roughly equal to that of first-team selection Jayson Tatum, second-team selection Anthony Davis and third-team selection Stephen Curry — and considerably higher than the Phoenix Suns’ Devin Booker and Kevin Durant, who also made the cut.
⛳️ Watch the Games
A 72-hole heater from Rose Zhang — who recently featured in this section herself — put an end to Nelly Korda’s quest for six straight LPGA Tour wins a couple of weeks ago.
Nonetheless, Korda — the world’s best women’s golfer, by miles — got herself back to the top of the leaderboard at the Mizuho Americas Open on Sunday to make it six victories in seven tournament starts.
Korda — like Zhang — is a natural ‘draw’ hitter, and in a great recent video for Golf Digest Luke Kerr-Dineen highlighted one of the aspects of her golf swing that aids her in achieving that shape8.
At the 15th hole last weekend, she uncorked a beauty of a second shot from 149 yards — with the ball sitting slightly above her feet — to set up her final birdie of the competition.
Korda stopped her high draw from the right edge of the fairway dead on the second bounce — and crucially matched the excellent approach of her playing partner, Hannah Green, who was eventually beaten by a single stroke.
You can watch a clip of this sequence here.
The next edition of My Week in Sport(s) will be published in two weeks, on Saturday June 8th.
Bayern Munich, by comparison, returned an actual goal difference of +49 on an expected goal difference of +53.0.
The period for which advanced statistics are available for these leagues on FBref, via Opta.
They also did so in 2019-20, when they finished second.
Seriously, look at this chart.
Using the same criteria as in last year’s piece (i.e. not including players, like Shai Gilgeous-Alexander, born in Canada — which has been home to multiple NBA teams over the years).
Embiid — who was born in Cameroon — has elected to represent Team USA in international competition.
Rather than focusing on ‘keeping her head still’ throughout her swing, she allows it to flow naturally with the rest of her body.