Analytic Review of 2020/21 - Part 1: The Overview
The first part of the WolvesAnalytics review of the season.
My overriding feeling on the last weekend of 2020/21 was, “bring on 6pm”. Absolutely horrible season in all regards, from significant and potentially life-changing injuries to the sad ending of one of the best managers this clubs has had for a long time.
Despite the overwhelming sadness of the end of the Nuno era, it’s time to take a slightly less emotional look at the season’s numbers.
Dropping to 13th on the final day of the season was disappointing, but about where we’d expect to be on the underlying numbers.
We had the 13th best expected goal difference throughout the season (essentially, the difference between the quality of chances created quality of chances given up). Brighton probably deserved a higher final position, and I’ll never believe that Newcastle was a better team. But 13th seems about right for us.
Throughout the season, we have given up up more chances than we have created. Aside from a run of games in January, we operating at negative expected goals per match. Put bluntly, teams who create fewer chances than they concede will lose more games than they win.
It doesn’t take a genius to see the issue that we’ve had this season - this is very much where the data passes the eye test.
We’ve struggled to score goals. Our attacking threat has been woeful.
This season, we seem to have given up shot quality for quantity. We have the 10th most shots in the league, but it takes us nearly 13 shots to score a goal. Only Sheffield United and Fulham score at a lower rate.
We struggled to create clear chances - we would expect to score a goal every 11 shots, the worst in the league. Only Newcastle and Southampton have had fewer shots from inside the six-yard box; only Leicester, Southampton and West Brom have taken a higher proportion of total shots from outside the penalty area.
As a very rough measure, between them, Raúl Jiménez, Diogo Jota and Matt Doherty took 37 six-yard box shots in 2018/19 and 30 in 2019/20. This season, Fábio Silva, Willian José, Pedro Neto, Adama Traoré, Daniel Podence and Nélson Semedo combined for 11 six-yard box shots. We very much lost that cutting edge upfront.
To put it another way, these six combined for 12 non-penalty goals. Last season, Raúl, Jota and Doherty scored 24. Very roughly speaking, one goal scored equals one point (look at the league table if you don’t believe me). An extra 12 points from an additional 12 goals moves us above Villa.
Obviously, two of those players are very young and have had good seasons. The performances of both Neto and Silva are not far away from the best under-21s in the league, and both will develop further. The issue this season was that they needed to be supporting players in attack, not crucial parts.
I will look a little more at the differences between formations in a separate post. Still, our chance creation across the course of the season has been very consistent, even if creating 1.04 goals worth of chances a match is low. Part of the issue was our finishing, particularly in the final third of the season, as we struggled to finish the opportunities we could create.
Fábio Silva and Leander Dendoncker are the main culprits here, each scoring 2.3 goals fewer than expected. There are obviously mitigating factors for both here. Silva getting chances on a part with his peers is fine for me: 0.35 NPxG per 90 compares favourably to Mason Greenwoods 0.37 NPxG.
And as for Dendoncker….
At the other end of the pitch, we were largely fine. We were down from previous seasons - this year, we conceded 52 goals from 46 xGA; last season, that figure was 40 conceded from 35 xGA.
We still gave up a lot of shots, but we’ve always done that. Our defensive strategy is essentially to sit deep and let the opposition have low-quality shots. Despite the changes in the back-4, we largely achieved that. In fact, defensively, we were relatively consistent throughout the season.
This season's worrying aspect is that we gave up far more goals than we did chances - this was consistent throughout the season.
Expected goals measure the quality of the chance, and so is not great for looking at goalkeeper quality. Think about a penalty - this would have an expected goals value of around 0.8, no matter whether the taker puts the ball wide, high into the top corner or Agüeros it down the middle. Expected goals measure the quality of the chance before the ball is struck.
A new measure has recently been introduced to better estimate goalkeepers - post-shot expected goals. This measures the quality of a shot - is it an arrow into the top corner, or is it meekly hit straight at the keeper. And so, this is a far better way (but still not perfect) to look at goalkeepers.
And unfortunately, Rui Patrício has not had a great season in terms of shot-stopping. Now, this is not to say that he needs to be replaced, but sometimes a keeper having an excellent season can make a world of difference to a season.
This has been a weird season, with Manchester United becoming the fourth team in history (third if you think of the Arsenal sides of 01/02 and 03/04 as the same team) to go through a season unbeaten away from home. And 51.2% of the points won in the league this season were by the away side.
A few teams bucked this trend - four teams won 55% or more of their points at home; Wolves, Southampton, West Brom and Sheffield United. I’m reasonably sure it’s not a coincidence that two of those sides were relegated.
At home, we were not bad. We only won 25 points for the 12th best home record overall, but that could have been very different. Our home expected goal difference was eighth in the league, demonstrating that we created more chances at Molineux than we gave up.
Again, our issue was in creating chances. We were only the 14th most potent attacking side at home but had a home defence worth of European qualification.
Away from home was a very different story. 20 points from 19 away games is fine most seasons, but not when everyone else is winning far more. And the underlying data is worse - on average, we gave up half a goals more chances away than we created.
Our attack away from home was nothing short of abysmal. Only Sheffield United created fewer chances and scored fewer goals on the road.
Defensively, we not too bad - solidly midtable. But suppose you create less than a goals worth of chances per match (as a reminder, only Sheffield United created fewer) and scoring even less. In that case, the defence needs to be exceptional. And it wasn’t.
Ultimately, 2020/21 will be remembered as a freak of a season. There is not much more to say than, 1) we didn’t create enough chances, 2) we didn’t score enough of the chances that we did create, and 3) we conceded more chances than previously, and 4) we gave up too many goals from those chances. Just one of those isn’t ideal; all four results in the manager losing their job.