Speaker A
This is the Ligue 1 table from 2016-17. By almost any metric, Monaco were phenomenal.
Analysis of Lucien Favre's unique football tactics that led to consistent overperformance beyond expected goals models.
Key Takeaways
- Lucien Favre's teams consistently outperformed expected goals models through innovative tactics.
- A compact, narrow low block defense forces opponents into low-quality shots despite allowing possession.
- Possession is used strategically as a defensive weapon to control the game and create attacking opportunities.
- Favre's attacking style focused on patient build-up and exploiting space rather than direct crosses.
- Expected goals models have limitations and can miss critical tactical nuances that affect actual performance.
What the video covers
- Monaco's 2016-17 Ligue 1 season was remarkable, but Lucien Favre's Nice overperformed even more in points versus expected goals.
- Favre's teams, including Borussia Mönchengladbach, repeatedly exceeded statistical expectations in both attack and defense across multiple seasons.
- Favre's defensive style involved a compact low block that allowed opponents possession but restricted dangerous passes and shots.
- His teams allowed many passes in non-dangerous wide areas but made it extremely difficult for opponents to create clear goal-scoring chances.
- Expected goals (xG) models miss key context such as defensive pressure and shot quality, which Favre's tactics exploited.
- Offensively, Favre emphasized possession as a defensive tool, building from the back and exploiting space between opposition lines.
- Favre's teams attacked mainly through wide areas but preferred cutting inside rather than crossing, creating higher quality chances.
- His tactical approach was ahead of its time, resembling modern intricate build-up play and positional manipulation.
- Favre's style consistently generated better-than-expected chances and defensive solidity, explaining the statistical overperformance.
- The video highlights the limitations of xG models and the importance of tactical context in football analytics.
Chapters
- 00:00Introduction: Monaco and Nice's Overperformance
- 00:26Nice's Historic Point Overperformance
- 01:19Favre's Consistent Overperformance Across Teams
- 01:50Favre's Teams Competing Against Bigger Clubs
- 02:26Defensive Approach of Mönchengladbach
- 03:03Compact Low Block Defensive Shape
- 03:48Defensive Similarities at Nice
- 04:49Limitations of Expected Goals Models
- 05:51Favre's Defensive Plan and Contextual Factors
- 06:51Favre's Offensive Philosophy and Build-Up Play
Full Transcript — Download SRT & Markdown
Speaker A
They scored 31 more goals than their cumulative xG suggested, and finished with almost 17 more points than the models predicted.
Speaker A
A truly remarkable season. But if we expand this table, something even stranger starts to appear.
Speaker A
In third place, quietly putting together one of the biggest statistical over-performances we've ever seen, was Lucien Favre's Nice.
Speaker A
Nice finished the season with 24.53 more points than expected. Across the last 12 seasons in Europe's top 5 leagues, that's one of the biggest point over-performances ever recorded, narrowly behind Jürgen Klopp's record-breaking 2019-20 Liverpool side, and ahead of some of the
Speaker A
best seasons from the biggest clubs in Europe. It would be incredibly easy to point at this season and say Nice were lucky, and there are plenty of one-off seasons where teams over-perform or under-perform these statistical models.
Speaker A
But for Lucien Favre, this wasn't an isolated event. Under Favre, Borussia Mönchengladbach had already produced two of the Bundesliga's biggest attacking over-performances between 2012 and 2014, while also producing two of its biggest defensive over-performances.
Speaker A
Put these together and Favre's Gladbach account for three of the four biggest combined over-performances in the Bundesliga across that period.
Speaker A
One individual season might be luck, but four times across two different teams starts to look a lot more interesting.
Speaker A
It suggests that Favre had found a playstyle that allowed his teams to consistently outperform what these models thought they were capable of.
Speaker A
But here is where it gets even stranger, because if you watched these teams play, they don't look like the type of teams that should be breaking these statistical models.
Speaker A
They rarely pressed aggressively, they deliberately slowed the game down, and they weren't even particularly prolific at creating chances.
Speaker A
And yet, Favre's teams in this period kept finishing near the top of the table, qualifying for the Champions League and competing with teams with much bigger resources.
Speaker A
So what exactly was going on? How could a team that deliberately played slowly, didn't press the opponents and often allowed them to attack, be one of the most consistent over-performers in the modern data era?
Speaker A
And more importantly, what was Favre doing that these statistical models just couldn't see? Spoiler alert, his playstyle was way ahead of its time.
Speaker A
Ok, so for this analysis to make sense, we need to ground it in the statistics, because the deeper you go, the weirder it gets.
Speaker A
Let's start with their defensive approach. The first thing you notice when looking at Mönchengladbach in this period is that they're not concerned at all with what the opponents want to do with the ball.
Speaker A
In the 2014-15 season, they allowed the opponents the most passes in the opponent's own third, the second most passes in the middle third, and also allowed the opponents the most amount of passes in their own defensive third.
Speaker A
They would sit in a low 4-4-2 shape, and watch as the opponents slowly circulated possession from side to side and forwards towards their box.
Speaker A
Rarely, if ever, in any rush to regain possession. But there was a clear structure underneath the surface.
Speaker A
This low block was extremely compact. Passes forward and into the centre were difficult to make, and they weren't just narrow vertically, they were also extremely narrow horizontally, with the fullbacks rarely starting further out wide than the box.
Speaker A
This meant that the opposition's wingers had all the time they needed, but once they wanted to push forward, this is when this defensive shape starts to make more sense.
Speaker A
In this graph, 0 is comparable to league average in each area. You can see that Gladbach go from nearly two standard deviations easier to pass against further up the pitch, to about 0.7 standard deviations on the tough side as opponents
Speaker A
close in on goal. So this shows us exactly what they're doing. Once the opponents get into the box, there is absolutely no room to manoeuvre.
Speaker A
It's one of the most compact low blocks you could imagine, and one that allows nearly no passing options in the box.
Speaker A
So naturally this forced the majority of their opponents to shoot from outside the box.
Speaker A
But because they had so many players packed in the way, this also resulted in Gladbach having incredible blocking metrics.
Speaker A
The same principles can be seen at Nice as well. While Favre had generally speaking moved on from his 4-4-2 defensive block, and the team had more of a focus on pressing slightly further up the pitch, their defensive game
Speaker A
was very similar to what we saw previously. They were compact in the centre, quick to close off dangerous passing lanes, but also giving the wingers time and space out wide.
Speaker A
Only once again to find themselves with no options once they attempted to attack forward.
Speaker A
This however did mean that the opponents could usually comfortably get the ball into the danger zone, only to then find themselves completely surrounded.
Speaker A
And here is where this overperformance in xG starts to make much more sense. xG is probably the default metric of the modern game, but it is not a perfect metric by any means.
Speaker A
All it tells you is that from a specific position, based on all chances created from that specific position, what the likelihood of a goal is going to be.
Speaker A
But there is a lot of context missing, and while these models are definitely improving, they often don't tell you how many defenders were in the way, whether the player shooting is weaker footed, or whether the player shooting is your top
Speaker A
scorer or a centre back who's just run 80 metres and is out of breath.
Speaker A
And all these things are incredibly important to understand whether this player actually has a good chance of scoring or not.
Speaker A
And this added context makes all the difference when understanding Favre's defensive plan. Because while teams were able to get the ball into dangerous positions and with a relatively high scoring probability, what the models didn't account for was just how many players
Speaker A
were in the way of the shot, and they rarely, if ever, allowed a player to shoot with a clear sight of goal.
Speaker A
So that gives us some insight into their defensive overperformance. But what about in attack?
Speaker A
What did Favre do with the ball that allowed his teams to attack in a way that was so much better than what these models could predict?
Speaker A
Lucien Favre was heavily influenced by Johan Cruyff and his vision for football. He understood that possession could also be a defensive weapon, so his teams were deliberately in possession, and he wanted to keep the ball until specific opportunities opened up.
Speaker A
From 2012 to 2014, Gladbach had the lowest ratio for passes in their own third compared to passes in the opponent's third, essentially meaning that all their possession was deep on the pitch.
Speaker A
But tactically this was actually an extremely clever setup, and one that was quite ahead of its time.
Speaker A
In fact, Favre was consistently playing out from the back, even when build-up looked like this, with no players allowed in the box from a goal kick.
Speaker A
Build-up didn't really explode in popularity for another 4 or 5 years, but his teams were already comfortable under pressure in this area, and would play the ball around the back while dragging the defending team forward.
Speaker A
Once they held the ball further up the pitch, this would create a big gap between the midfield and defensive line that they wanted to exploit in different ways.
Speaker A
They rarely played directly through the middle, but once they were ready to attack, they would shift players out wide and attack down the flanks.
Speaker A
However, once the ball was in these positions out wide, another counterintuitive thing would happen, and rather than crossing into the box, they would then cut back into the centre and look to play right through the middle.
Speaker A
In fact, no other Bundesliga club played fewer crosses, but 20 of their assists came from passes of 14 metres or more, which was far above th
Speaker A
So while they rarely got forward, when they did they instantly had the space to create a chance.
Speaker A
It's a similar story at Nice as well. That season their entry tally into the final third was only 653, the second worst in the league.
Speaker A
Yet 31.24% of their entries resulted in a shot from inside the box, which was the second best in the league.
Speaker A
That places them on the complete other end of the spectrum compared to the other top teams in the league like Monaco and PSG.
Speaker A
Usually by staying deep on the pitch and circulating possession they could often drag the midfield line forward, and as soon as the right gap opened up they would break forward quickly, freeing space between the lines for these players
Speaker A
to then attack through the centre. For example, this goal from Mario Balotelli, who led the line for Nice in 2016.
Speaker A
Nice hold onto the ball and break forward on the flank, and now they have all the space in the centre to cut inside and find Balotelli in a perfect position to shoot.
Speaker A
And this brings us to the final piece of the puzzle. Favre's attacks weren't necessarily creating more chances than the model predicted, but they were creating better chances that the numbers could understand.
Speaker A
By keeping possession for long periods, stretching the opposition from side to side and then attacking through the centre when the right gap appeared, Nice were able to shoot in situations where there were very few defenders between the ball and the goal.
Speaker A
Their shots had an average of just 2.2 defenders in the way, the lowest figure in the league.
Speaker A
At the other end, their opponents against Nice spaced 2.8 defenders on average, the highest in the league.
Speaker A
So while the location and characteristics of these shots might have looked relatively ordinary to an XG model, the actual context of the chance was completely different.
Speaker A
And well, this was and still is a fundamental limit of expected goals. Football is wonderful because no two chances from the same position are exactly the same.
Speaker A
It can vary enormously depending on how close the defender is, the speed of the assist, or who's even actually taking the shot.
Speaker A
And Favre's teams were consistently manipulating these factors. They were allowing opponents to shoot in relatively dangerous areas, but these shots were incredibly difficult to execute because of the amount of players in the way.
Speaker A
And as a mirror image and attack, they were so good at creating a clean opportunity that the chance was actually easier to convert than what XG predicted.
Speaker A
So did Lucian Favre actually break XG? Well, yes, in my opinion he did exactly that.
Speaker A
He showed us all that just shooting from one location doesn't automatically mean that that chance has the same statistical value as another shot from that specific location.
Speaker A
Now this was over 10 years ago and a lot of the more recent expected goals models do factor in a lot more of this nuance, like defenders and body positioning and the type of shot that's taking place.
Speaker A
But XG is still far from perfect and in my opinion will never be able to truly capture everything that's going on when the player is taking a shot.
Speaker A
What it does do well, however, is to give us a benchmark that we can use to compare and contrast different players and different teams.
Speaker A
And one thing that stands out from this analysis is that Favre's tactics were ahead of his time.
Speaker A
He developed a tactical system that gave the opponents no space to shoot, yet at the same time allowed his players to be in the perfect position to finish off the chance.
Speaker A
And his tactical style was very reminiscent of what a lot of modern teams want to do today, with intricate build up play and dragging defenders out of position to then attack quickly when the defense isn't set up correctly.
Speaker A
Whether this system would work in the modern game, I'm not so sure, because at the end of the day, while Favre did overperform statistically, these teams weren't the highest scoring or even the best defenses in the league.
Speaker A
And maybe what Monaco did that season is a better approach. So maybe next time I make a video on them, because that team was absolutely stacked.
Speaker A
Okay, thank you so much for watching and I really hope you enjoyed this video, and this video is actually based on a chapter from this book called Football Hackers by Christoph Biermann, who I met very briefly in Berlin a
Speaker A
few months ago, and he mentioned this book to me so I picked it up.
Speaker A
And I really enjoyed it, and it's a little bit about the development of data in the modern game and it gives you good insight into why football looks the way it does today.
Speaker A
The only thing I'd say is maybe a little bit outdated, because it was written in 2019 and a lot of things have changed since then, but still if you're curious in knowing more about how these models developed, how expected goals came to be,
Speaker A
then this is a great recommendation that I will give you and the link is in the description down below.
Speaker A
Thanks for watching!
Topics:Lucien Favreexpected goalsxGfootball tacticsBorussia MönchengladbachOGC NiceLigue 1Bundesligafootball analyticsdefensive strategy
![[YTP] JORDAN’S LAST STAND (SUPERCUT) — Transcript](https://i.ytimg.com/vi/ioyaZ39gV5I/maxresdefault.jpg)
![[YTP] OOPS, ALL RATS! (SUPERCUT) — Transcript](https://i.ytimg.com/vi/Wl-xtUEGjOI/maxresdefault.jpg)
