By post-Ferguson standards, Premier League runners-up and Europa League finalists amounted to a good season for Manchester United in 2020/21.
Ultimately there were no trophies, but 'progress' has been the watchword for a few years now, and there have been signs of it both on and off the pitch. United’s trendline since the start of the 2018/19 season makes for pretty good viewing – attacking and defensive numbers have both been moving in the right direction for the most part.
However, a fairly distant second to a brilliant Manchester City is perhaps less impressive given the problems faced by United’s top four bedmates – a Liverpool without their three top choices at centre back, and a Chelsea that only really hit their stride after Tuchel arrived and turned them into champions of Europe. We are starting next season with the top end of the league in a very different spot; that makes United’s prospects less enticing.
Even with the problems endured by Chelsea and Liverpool, United were some way off the other Champions League qualifiers in terms of expected goal (xG) difference per game. Our numbers have them at +0.33 xGD per game to City’s +1.12, Liverpool’s +0.76, and Chelsea’s +0.70. Enough for fourth-best in the league, but still some way off the other three and you also have to factor in that United spent a good portion of the season chasing down 1-0 deficits, which could have inflated their underlying numbers slightly. All other things being equal, the gap between United and their top-four rivals could increase even more with their rivals starting in better spots than they were for much of last season.
There is, however, the small issue of the transfer window. Much will depend on whether or not the new signings can solve some of the issues that have plagued United for some time now. The over-reliance on the left side in build-up seems to be one of the issues United are trying to solve in this window with the signing of Jadon Sancho and fairly loud noises about Kieran Trippier. Likewise, the issue of depth at centre-back, with Eric Bailly’s injury proneness and Axel Tuanzebe’s youth, has been addressed in the way that issues of depth should be – by strengthening the first team. The addition of Eric Ramsay as a first-team coach who’ll be in charge of set-plays will also serve as music to the ears of this article's readership.
The Goalkeeper Dilemma(?)
One question United don’t appear to be trying to answer in the transfer window (despite Tom Heaton having a pretty stellar pre-season thus far) is: who is the first-choice keeper?
United spent most of last season with two keepers in direct competition for the first-choice berth. As always, competition for places should really fall into the ‘nice problem to have’ category. It's not something we often see when it comes to goalkeepers. From the outside, it seems that the most important factors in the decision are not things that can be measured directly: how comfortable the defenders are with the keeper, player morale, maintaining David de Gea's value for a future sale, the extent to which having to oust de Gea motivates Dean Henderson -- all of these human and managerial factors that football managers have to contend with and we don't have to worry about. There is not much in it from a playing point of view, so it’s all about the best way to manage the handover as far as I'm concerned.
Henderson was more aggressive in coming to get the ball, claiming the ball roughly 4% more than expected based on the crosses he faced. That is the only stark difference between Henderson and de Gea on these metrics -- they both had excellent positioning (measured by their average distance from a typical keeper location), distributed the ball similarly in terms of pass length and in finding free men, and were more-or-less average in terms of shot-stopping relative to post-shot xG. The minor differences here are pretty paltry relative to season-to-season variability.
Having the opportunity to bed Henderson in while also giving de Gea firm competition over the course of a season is a big luxury. The succession plan looked like it might be reaching completion towards the end of last season, so it will be interesting to see who is breathing down whose neck come the season's start.
Set-Pieces
United have clearly identified set-pieces as an area that can be improved. Eric Ramsay joins from Chelsea U23s as a young coach with a very good reputation and has reportedly been tasked with overhauling United’s set-pieces.
Their performance last season wasn’t terrible, but it was close to average, and set-pieces are an element of the game where focused work can pay real dividends in terms of goals and league standing.
What sort of impact could Ramsay have? A realistic objective is to get close to the top teams' set-piece efficiency of last season.
Let’s say United improve their xG/Set Piece from 0.0062 to 0.0082, and xG/Set Piece Conceded from 0.0064 to 0.0044, putting them about halfway between Everton and Man City on the scatter plot below. That’s an improvement of 0.2% of a goal for each set-piece conceded and taken. It doesn’t sound like much, but considering that teams take about 35 set-pieces in every game and concede about the same, improvements at both ends of the pitch would add up over a season.
That means United could (optimistically, but not fancifully) improve their xG difference per game by about 0.14 through set-pieces alone, a number that would make up about 38% of the gap between them and Chelsea. That kind of impact is really hard to achieve through player signings unless you’re spending big bucks on an upgrade to a current starter. This back-of-the-envelope calculation points towards a shrewd bit of business, as long as Ramsay can get United to somewhere near league-best.
Jadon Sancho
In signing Sancho, United have obtained one of the most exciting young prospects in world football, and in a position where they've been lacking for many years. According to StatsBomb's player similarity tool, the most similar players to Sancho in the top 5 leagues last season were Ousmane Dembélé, Christian Pulisic, Leroy Sané, Samuel Chukwueze, and United's own Marcus Rashford, who spent a good deal of time on the right last season. These are all quick, technically gifted wide players with an eye for goal.
Where Sancho stands out from these players (barring Dembélé) is in his creative output. 0.22 xG assisted from open play last season puts Sancho 19th of wingers in Europe's top 5 leagues. Having Erling Haaland to play those passes to is not a luxury Sancho will be afforded at United (unless...?) and, in their current incarnation, United are a bit less dominant in a typical league game than Dortmund. I'd expect his creative output to drop slightly, but he undoubtedly adds tremendous quality to an already exciting United attack.
There is always a question about whether a player will gel with a new team, but some of the personal issues players can face when moving to the Premier League are not really there for Sancho. He speaks the language, grew up with the league, knows lots of United players already, and knows Manchester. There won’t be much time to settle, coming straight off the back of the Euros, but I don’t see that being a problem despite the inevitable overreactions (including from me!) in the early stages of the season, good or bad. Right now, this looks a fantastic piece of business.
The left-hand side
Pre-Sancho, United's attack skewed towards the left-hand side. Here are their most over-represented pass clusters last season compared to the rest of the league:
Notice how similar the top-four teams look in terms of their most over-represented clusters. They all favour one side, and they all favour short-ish passes in the opposition half. There is not much getting away from this as a dominant side in the Premier League, and United clearly favoured the left flank.
This tendency is also borne out in United’s possession value. StatsBomb’s new possession value model, On-Ball Value (OBV), estimates the extent to which an action improves a team's expected goal difference over the next two possessions. Looking at the figure below, it's clear that United were strongest down the left (relative to the other teams in the league). That over-represented zone was bearing fruit.
The corresponding area on the right looks pretty barren by comparison, with United only producing at around league average from there. Sancho should help to turn some of those zones red, but that will, in turn, have an impact on usage – Bruno Fernandes will carry less of the creative burden, and the left side probably won’t be favoured quite so heavily. It’ll be interesting to see how this map has changed by the end of the new season.
The midfield
Another area of the pitch in which many fans and pundits believe United needs reinforcements is central midfield. The general rationale for this is that United’s attacking talent might be better suited to a system where both Paul Pogba and Fernandes play centrally in front of a single, more defensively minded midfielder. The typical concerns are that neither Scott McTominay nor Fred can cover for them on their own, Nemanja Matić is unable to play enough minutes, and Pogba’s defensive tendencies aren’t well-suited to playing in the midfield two in the current set-up.
Finding this miracle player to unlock the ‘Silva & de Bruyne’ style of play that I think people are envisaging is not an easy task, so I expect we’ll see something somewhat similar to last season in terms of the composition of the midfield. Matic can still offer a lot in short spells, but it’ll be Fred, McTominay and Pogba who mostly share the two midfield berths.
But what is the state of the ‘McFred double pivot’? Using Statsbomb 360 data, we can look at how often players break lines with their passing. We define a line-breaking pass as any pass that:
- successfully progresses the ball at least 10% closer to goal…
- …and either splits two opposition defenders or ends in the space behind them, provided that the defenders are less than 5m apart vertically.
We filter to passes that are completed and that break a line in the opposition half, then normalise per 90: the results make Matic’s lack of mobility and inability to accrue significant minutes all the more annoying. He ranked fourth in this metric among PL midfielders with >900 minutes last season, bettered only by Thiago, Mateo Kovačić, and Fernandinho. Fred was 13th of 73, McTominay 26th.
These numbers are fine but probably not quite at the level United need. That said, Paul Pogba’s presumably increased minutes this season should help substantially with progression from midfield (provided he sticks around, of course).
We can look at a simple model of pass availability to explore this a bit more. We start by drawing defensive cover shadows that look like this:
Attacking players who are not in a cover shadow are classed as ‘available’. If they are located 10% closer to the opposition goal than the ball, they are classed as a ‘progressive option’. Then we can look at the percentage of the time a player takes a progressive option when it’s available to get some idea of passing aggression.
This approach comes with many caveats, and a better way to do this would be with a 360-based possession value model combined with a 360-based expected pass model. However, we aren’t quite there yet, so for now, we’ll proxy expected passes with ‘availability’ and possession value with ‘progression’. Also, we’re only looking at the frame in which the pass was played, so it could be that players are turning down progressive passes before allowing the opposition to block lanes, which makes it look like the option was not there by the time the ball is released.
All that said, we can see that some of the criticism levelled at McTominay’s progressive passing is perhaps unwarranted – movement ahead of him gives him a decent number of options to play progressive passes, and the data suggests he attempts to do so at a reasonable frequency. He is in the company of the more aggressive progressive passers in the league from deeper in midfield when it comes to attempting progressive passes. However, his technical ability and accuracy let him down, and he’s not massively proficient at completing line-breaking passes or progressing the ball overall.
Parting thoughts
United have been making steady progress in their underlying numbers for a few years and have strengthened substantially in attack and defence in this window. The bookies envisage the return of an established top four, with United the least-favoured of those sides, an assessment that seems about right. The new signings and appointments should make up some of the fairly substantial gap between United and their top four rivals from last season, but probably not to the extent that we’ll see them in title contention come the end of the campaign.
Lastly, I can’t finish this article without mentioning the cultural impact that Marcus Rashford is having and at 23 years of age. Whatever happens this season, he is a beacon of light.
Want to read about another team? The rest of our Premier League season previews can be found here
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com
We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course
Follow us on Twitter in English and Spanish and also on LinkedIn
To move West Ham from 16th place in 2019-20 to 6th in 2020-21 has to rank alongside any of David Moyes' career achievements. His return to the club 18 months after leaving in the summer of 2018 was not greeted with great positivity, and it appeared to take the enforced pandemic shutdown to get the ship steering in the right direction. However, few teams can be seen to have benefited so greatly from playing in the current climate, and the way the club has evolved and improved in Moyes' tenure is a ringing endorsement towards the elusive attribute every manager would like--time.
That 26 point season-on-season improvement ranks third behind Leicester 2015-16 and Chelsea 2016-17 across the Premier League era:
It was no fluke either and is reflected in the shot metrics that powered the team:
- non-penalty xG rose from 1.14 to 1.34 per game
- non-penalty xG conceded declined from 1.49 to 1.17 per game
- ...causing xG to go from -0.35 per game to +0.17; half a goal gains are not to be undervalued, this team improved substantially
- they took more clear shots (1.5 up to 1.9 per game)
- they conceded fewer clear shots (2.6 down to 1.9 per game)
- the chopped a load off their set piece shot concessions (down from 3.4 to 2.6 per game, xG here halved from 0.31 to 0.17 per game)
...while stylistic play features were clear too:
- 32% of box entries were via a cross (most in league)
- led the league for metrics relating to pace and directness
- second lowest passes per defensive action
- least counterpressures and resultant regains
- closest shot distance (both non-penalty and open play)
This all paints a picture of an organised, somewhat reactive and patient team. They also knew which games represented their best chance of success. Eventual top four? A 0-1-7 WDL record. Rest of the league? A stellar19-7-4 record.
We discussed the role of long carries and shots and goals from defensive regains in last year's preview and these features generally persisted. West Ham kept their shape and waited for opportunities to break forward. Once they did they created chances at a decent volume for their key contributors, once more Michail Antonio benefited but also a resurgent Jesse Lingard:
The boost that Lingard brought to the back half of West Ham's campaign was both large and perhaps unexpected. Lingard, a player who has been noted to steer under his expected goals values in the past, scored eight non-penalty goals from an xG of around four to force his way all the way back into the England squad.It appears likely that he will be staying in Manchester this season and for West Ham losing not just the player but one in the form of his life is a significant blow.
As yet, the Hammers have been slow to enter the transfer market, with the main change from the start of last season the removal of their highly priced bench men. Sebastian Haller left for Ajax in the winter and Felipe Anderson was practically given away in a cheap deal to take him to Lazio. Neither convinced for West Ham in the aggregate and quickly fell down the pecking order under Moyes, but their removal, alongside that of Lingard, does leave something of a lack of depth in attacking positions. In general Moyes operated a small squad in 2020-21, and he talked about how he instilled an effective group ethos during difficult times in a column for the Times ahead of the Euros:
"At West Ham, during the club season, I made a conscious effort to make training more fun...to not keep it so heavy. I thought about the world the players were having to live in, there was no opportunity for them to leave the bubble (so) we had to make sure they came into work and felt they were with their friends. Footballers were challenged by lockdowns and the lack of "normal" life and found they were happy to get out of the house, pleased to be in training, and craved it from the point of view of structure and routine. So there was actually more training -- but we made sure that many days there were lighter training loads. (In general) unity and spirit helped "smaller" clubs navigate the difficult conditions and do well in many of the domestic leagues."
What has been ahead of and during this summer done is limited; Craig Dawson's loan was converted to a permanent deal back in April and Alphonse Areola recently arrived on loan from PSG after a solid season in London at Fulham:
With Fabiański now 36 years old, it makes sense to start thinking about succession and we will see who Moyes puts his faith in from week one. Areola is a more active and "modern" goalkeeper than Fabiański, but the Polish veteran has done little wrong in his West Ham tenure and may well see the position still as his own.
Warning Signs?
One of the hallmarks of Moyes' second spell has been his consistency in selection. In 2020-21, eight players played in more than 70% of the team's minutes and the core group he has relied upon is small. When Antonio missed time in the autumn of 2020, Haller stepped in, but there are a further string of players who look core--perhaps Declan Rice, Aaron Cresswell, Tomáš Souček and Vladimír Coufal. The team avoided injury problems last season and that consistency in selection looked to work to West Ham's benefit.
The main selection choices week to week look likely to be in the attacking midfield band. Jarrod Bowen, Manuel Lanzini, Pablo Fornals are all in the mix but could Saïd Benrahma's second season see better returns and omit the need to replace Lingard in the market?
He saw a lot of bench time in Premier League season one, and the variation of his profile compared to his time at Brentford was stark. Gone was the shot-happy drifting left-sided attacker, replaced by a more shot-shy, creative player fitting in across the attacking band. He was in competition with Pablo Fornals for a left-sided slot then Lingard became the key man in the centre and Benrahma never quite nailed down the starting space. It would be no surprise if Moyes values positional discipline and Benrahma doesn't quite offer exactly what he wants from his wide attackers, but he's certainly talented enough to come forward and contribute more, if empowered.
Setting realistic expectations for West Ham's season means without further investment, a replication of anything approaching last season's 65 points would represent the absolute best outcome this group could expect, and doing so, even from similarly small net positive metrics is fairly unlikely. There's also a hidden drift in the end of season form to note:
Throughout Moyes' tenure, West Ham's xG and goal differences have tracked fairly reliably. This is no given, and we can see divergence at the end of 2020-21, in which goal difference persisted significantly ahead of xG. Now, that was great to keep the results coming as the summer beckoned, but is less encouraging ahead of the new season; West Ham were at their worst in 2020-21 at season's end.
Where will West Ham end up? Bookmakers are currently projecting a firm mid-table finish and around 50 points and while that may seem slightly ungenerous, it's not hard to toss around their metrics, quietness in the transfer market and risk profile towards squad depth versus key injuries and feel that is in range. The coming month and transfers could move the needle here though and I'd probably have them good for a couple more wins than that, particularly if they acquire a useful back-up or alternative to Antonio.
2020-21 was a season in which the unique circumstances of low pressing pandemic-ball and a stable first team helped Moyes extract the best from a group of players that played with a clear identity. With crowds back in stadiums, it's hard to imagine that the slightly neutral aspect of 2020-21 will persist and some of the benefits accrued from a more passive strategic style may well decline. That does not mean that West Ham's future prognosis is particularly worrying, just that a small detour down into mid-table is the most likely outcome. This would still represent a huge overall improvement on what Moyes inherited and justify his retention, with a view to continuing to build and sustain within the league comfortably. To move upwards or match 2020-21 will probably take both more time and significant investment, and having been bitten by high-priced flops such as Haller and Anderson, West Ham's ownership may be content to swim in shallower transfer waters for now too.
Last season I felt that transfer backing could be the difference between lower mid-table security and something slightly more than that, and the equation is probably similar. Again West Ham have a choice to make, but this time they are making it from a higher vantage point, that of safety.
Want to read about another team? The rest of our Premier League season previews can be found here If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course Follow us on Twitter in English and Spanish and also on LinkedIn
For numerous reasons, Brighton will be one of the most interesting teams to watch next season. They enter year three of Graham Potter’s reign, playing football that has garnered praise from highly-decorated managers, and have several analytical storylines that need concluding.
Potter was brought in in 2019 to stabilise and then improve Brighton’s Premier League standing. They achieved the first objective with consecutive relegation-survivals of relative comfort. It’s the latter objective that the Albion hierarchy will now be looking to move towards and, despite back-to-back 41-point seasons, 2020/21 saw indications that Potter’s influence was beginning to turn Brighton into a team that can climb the table.
The topline, and I’ll bet this won’t be news to many of you, is that there was a dramatic improvement in Brighton’s underlying numbers last season. They added five goals of xG to their attacking output -- creating 44.8 non-penalty xG up from 39.7 in 2019/20 – but tightened up remarkably at the back, shaving 20 goals off their expected defensive output (32.5 xG Conceded in 2020/21 compared to 52.9 in 2019/20). The outcome was that Brighton had the fifth-best expected goal difference last season, making the most considerable improvement of all teams to compete in both campaigns.
The question is, how does a team improve their process so dramatically and yet fail to improve their points tally?
Both the attack and the defence have to take some responsibility here. They undershot their xG to the tune of ~11 goals in attack, whilst the defence conceded ~seven more than expected based on the chances they conceded. The result was a ~17 goal underperformance between their expected goals and actual goals, the largest discrepancy in the league.
There were numerous culprits up front. Neal Maupay faced most of the scrutiny for turning 9.9xG into five goals, but there were multiple contributors to the problem across the team: Alireza Jahanbakhsh and Pascal Groß combined for 47 shots, 4.1xG, and zero goals, for example. Collectively, Brighton turned 44.8xG into 37.6 post-shot xG, with post-shot xG only accounting for shots on target and factoring in the shot placement. There was plenty of banjo-swinging, but bovine derrières went largely unscathed at the Amex.
At the back, nearly all of their underperformance occurred in the first 12 games. Mat Ryan’s performances left a lot to be desired, and he was dropped for Premier League rookie Robert Sánchez, who proved to be an immediate upgrade on Ryan and finished the season with the best Shot Stopping % in the league: a measure of how many goals a goalkeeper saved/conceded based on the post-shot xG faced.
Sánchez’s introduction coincided with Brighton’s best spell of form in the season in early January, form that was the strongest indicator yet as to the potential level of Potter’s team. A run of 2-8-8 in their first 18 games returned 14 points, but Brighton then took 10 points from four games against Leeds, Fulham, Spurs and Liverpool, all without conceding a goal. The match immediately before that four-game run was a 1-0 defeat to Manchester City. Pep Guardiola said this in the aftermath:
“We were in front of the best English manager right now. You have to be a top side to play that way. They have good players, good build-up. Every pass makes sense. Their movement between the lines up front makes sense. Every player is in his position to get the ball and be free.”
What is it about the style of play Potter has implemented that prompted Guardiola to refer to them as a “top side”?
For starters, their approach in possession. Brighton generally play shorter passes in build-up, with fluid movement between players. Their Pace To Goal -- the average speed of build-up (in m/s) of possessions that end in shots – was the slowest (read: most patient) in the league last season, and the types of passes they play within those moves are those you’d more closely associate with a team competing towards the top of the table than one battling it out at the bottom:
Their most common pass types were those moving the ball around the defence in the early stages of build-up and those played from the “cutback zone” – the area of the pitch where high-quality chances are regularly created. Cluster #2 and #8 are the pass types we want to zoom in on to get a picture of how Brighton look to attack: short passes from central areas to create opportunities to cut the ball back from inside the box.
A short-passing possession game that’s effective at getting into areas for cutbacks? It’s no wonder Pep was enamoured.
Only six teams reached the final third more often than Brighton last season. Only three completed more passes inside the penalty box. Only Leeds completed more cutbacks. Brighton controlled the territory effectively, but their slow build-up did come at the cost of some shot clarity: their opponents had an average of 3.6 defenders behind the location of the ball when Brighton shot from inside the box. That number was the highest in the league, as was the 39% rate of those shots being blocked.
While this may be problematic in attack, it’s a trait they use themselves in defence. On average, Brighton had the second-most defenders behind the location of the ball when their opponents took a footed shot from inside the box. That defensive organisation goes a long way to explaining why their defensive numbers were so good; they were able to limit the number of times their opponents could penetrate their area and reduce the quality of the chances their opponents took when they did pull the trigger. The shots Brighton conceded last season were taken from the third-furthest distance away from goal on average.
Current Squad
The big story is around the £50m departure of Ben White to Arsenal. A player that worked his way up through the Brighton youth system before spending time graduating through all three of the EFL divisions, White made his Premier League and England international breakthrough last season. The sale is undoubtedly a short-term blow to Brighton’s defensive ranks, but talk of a replacement is already underway. Defensive solidity will be the main attribute being searched for, helped by the fact that Brighton already possess one of the best in-possession defenders in the Premier League. Adam Webster made 7.7 carries and passes that moved the ball 25% closer to goal last season, more than any other centre back in the league.
Since promotion to the top tier, the Seagulls have developed a niche in the transfer market: no region is out of their reach. They’ve signed domestic up-and-comers (Adam Webster and Neal Maupay), domestic experience (Danny Welbeck and Adam Lallana), but have also been unafraid to sign from smaller European leagues (Leandro Trossard and Jakub Moder) or even continents further afield too (Alexis Mac Allister and Moisés Caicedo).
As a result, their squad is full of players that have the potential to be good Premier League players, signed for relatively modest fees. A good example of the strategy is mobile midfield maestro Yves Bissouma -- the other name being linked with a transfer to the league's upper echelons. His potential replacement could already be in the building, with Enock Mwepu the only bit of business Brighton have concluded so far besides Kjell Scherpen. Scherpen will be a 6’8” goalkeeping understudy to Sánchez, but it’s Mwepu who should make the stronger impact on the first XI after signing from Salzburg.
Transfers are glamorous, but Potter will be looking to players already within his current squad to step up and play a bigger role next season. Alexis Mac Allister, Jakub Moder and Steven Alzate all played 600-1200 minutes last season and could be set to make a more significant contribution this time around. Moder posted some impressive numbers while in Poland, so it’ll be interesting to see if he can adapt to the level of the Premier League, whilst Mac Allister showed signs of a player that could be a more regular starter in the 1000+ minutes he played last season.
Projection
It’ll be captivating to watch the development of this Brighton side under Graham Potter, but it’s far from a slight to say that we shouldn’t expect their metrics to be quite so strong this time around. Other teams should improve to overcome Brighton's position of fifth in the expected goal difference table and the loss of key players could hamper them slightly -- they’ll need to find a new talisman in defence after White’s departure, and the rumours of interest in Bissouma refuse to go away either.
All that said, their approach of signing players that are one-to-two years away from Premier League excellence (see the 2018 signing of Bissouma, Yves) could pay dividends again with plenty waiting in the wings to take their chance, under a coach in Potter that has proven he can bring these players into the team and improve them further. The positive metrics, the style of play, the development of players with potential; all these factors mean we should be feeling positive about Brighton ahead of the 2021/22 season. Steady progress and adding several points to the 41-point tally we’ve seen in the last couple of seasons should be the minimum aim.
Even more is not out of the question.
Want to read about another team?
The rest of our Premier League season previews can be found here
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com
We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course
Follow us on Twitter in English and Spanish and also on LinkedIn
Wolverhampton Wanderers hit the ground running following their promotion to the Premier League in 2018, achieving consecutive seventh-place finishes in their first two seasons back in the top flight. But things didn’t go quite as smoothly for them last time around, and they come into the 2021-22 season seeking to reestablish themselves as a top-half side.
After two seasons of positive goal differences and near-60-point hauls, Wolves finished down in 13th last season, with 45 points and a -16 goal difference. There was a pretty stark drop off in the underlying numbers, too, from expected goal (xG) differences of 10.84 and 11.51 in 2018-19 and 2019-20 respectively to a -2.86 difference in 2020-21.
In 2019-20, their performances were extremely controlled; last season, things started to fray at the edges. Their worst four results and three of their worst four individual match xG differences across the last three seasons came in 2020-21.
The downturn in their season-long xG difference was evenly distributed between attack and defence. They were roughly seven expected goals worse off at both ends.
In attack, their set piece production remained stable, but Wolves really struggled to create good chances in open play. They took a comparable number of shots season on season, but their open play xG production nevertheless plummeted from 0.9 xG per match in 2019-20 down to 0.68 xG per match in 2020-21. Only Sheffield United, West Bromwich Albion and Crystal Palace created less.
The culprit: the quality of their chances. Wolves had the worst average shot quality from open play in the league at 0.08 xG per shot. Over the course of the season, they actually created a league-sixth-high number of shots within 20 seconds of regaining possession -- the sort of transitional phases of play in which you’d expect them to be able to create better quality chances -- but even in those situations they generated the lowest quality opportunities in the league, right in line with their overall xG per shot.
It would be tempting to blame the sickening skull injury suffered by striker Raúl Jiménez away at Arsenal in late November and his subsequent absence for Wolves’ attacking issues. He was certainly a talismanic reference point at the centre of their attack that various experimentations with formations and personnel, including underwhelming January loan arrival Willian José, failed to adequately replace. Wolves were marginally better in terms of open play attacking output during the first nine matches of the season with him in the side, but not significantly enough so given the relatively small sample size to indicate his lack of availability from December onwards was the primary reason for their struggles.
In defence, Wolves gave up pretty much the same average quality of open play shot as in 2019-20 but just did so more often, conceding 27 more shots, 3.01 more xG and two more goals over the course of the campaign. Opponents found it easier to complete passes in the final metres of the pitch and created a higher volume of chances.
Wolves also also went from being one of the best teams in the league at defending set pieces in 2019-20 to a middle of the road outfit by the underlying numbers and one of the league’s worst in reality. Only five teams conceded more goals from set pieces than their total of 12.
The downturn in pretty much all phases of play left coach Nuno Espirito Santo scrambling for solutions. The back three that had been the permanent setup in each of the club’s first two top-flight campaigns occasionally gave way to a back four. Those long diagonal balls out to the wing-backs that had been such an identifiable feature of Wolves’ play began to fade from view. By the end of the season, it was little surprise that Nuno’s four-year spell at the helm came to a close with a mutual parting of ways.
What went wrong? Could it be that the players were tired with Nuno’s approach, tired with Nuno himself or maybe just... tired? A relatively small squad saw nine players take part in more than 75% of the available minutes -- a league high alongside Burnley.
Wolves were clearly not as good as they had been in 2019-20 but even still their goal difference and points haul weren’t fully representative of the quality of their displays. An underperformance of the underlying numbers at both ends of the pitch allied to a -3 difference on penalty goals (lurching from a +3 difference in 2019-20) saw a -2.86 xG difference turn into a -16 goal difference.
It isn’t the worst situation for Nuno’s replacement Bruno Lage to walk into. His only previous experience as a head coach -- he was assistant to Carlos Carvalhal at both Sheffield Wednesday and Swansea City -- came at Benfica in his native Portugal, where in early January 2019, he stepped up from his role as B team coach to replace the sacked Rui Vitória. What followed was a remarkable second half of the campaign that saw the club romp to the title with 18 wins, one draw and zero defeats.
Lage's side scored an average of 3.79 goals per match along the way, and had comfortably the best goal and xG differences of any team.
Benfica had played a 4-3-3 before Lage’s arrival but largely lined up in a 4-2-3-1 or 4-4-2 formation thereafter -- depending on the positioning of João Félix relative to the primary striker Haris Seferović -- something that continued into the 2019-20 season. Benfica again had the best xG difference in the league, but it was Porto who took the title by five points. Lage resigned five matches from the end of the campaign after a run of two wins, four draws and four defeats that had given Porto the upper hand.
While there have been a couple of deviations, the evidence of pre-season suggests that Lage will use the same formation at Molineux. While it would be unrealistic to expect him to play quite such dominant attacking football with a side who are not one of the primary powers of the Premier League, Wolves are nevertheless likely to employ a more front-foot approach than they did under Nuno, when they were one of the deepest defensive teams in the league.
Lage’s arrival aside, it has so far been a relatively quiet summer but one that has yielded what appear to be solid signings. Goalkeeper Rui Patrício departed to Roma to be replaced by another Portuguese custodian in José Sá, an over-performer of his post-shot xG numbers for Olympiakos in each of the last two Greek Super League seasons.
Rayan Aït Nouri has made his loan move from Angers permanent, while Wolves have also brought in Yerson Mosquera from Atlético Nacional in Colombia, a tall and aggressive young centre back who also doesn’t look too shabby with the ball at his feet.
But perhaps the most intriguing signing is that of wide forward Francisco Trincão, on loan from Barcelona. He burst onto the scene with Braga in Portugal during the second half of the 2019-20 season and was immediately snapped up by Barça in a €35 million deal. He received a smidgin over 1,200 minutes of action across all domestic and European competition last season but has been loaned out in search of a greater workload.
The 21-year-old would seem to perfectly fit the profile of wide forward that Wolves currently have on their books. Last season, Wolves were more reliant on carries to advance the ball forward inside the attacking half than any other Premier League side -- 35% of their distance advanced was achieved via carries -- and both Adama Traoré and Pedro Neto ranked in the league’s top 10 in terms of longer carries (>=10 metres) that led to shots, assists and key passes, as well as to direct entries or passes into the penalty area.
At Braga, Trincão ranked fourth in the Portuguese league in terms of longer carries that led to direct entries or passes into the penalty area and also produced 0.73 shots per 90 from longer carries -- a higher figure that either Traoré or Neto managed last season. If he can get somewhere close to the overall figure of ~2.5 shots per 90 he posted at both Braga and Barcelona, he could prove an astute addition.
It will be interesting to see how Lage chooses to align his various wide forwards given his seeming preference for a 4-4-2 formation. Traoré has more of the attributes of a winger so will likely occupy one of the wide midfield slots, but Neto, Trincão and Daniel Podence could all potentially be used as either wide midfielders or central forwards alongside one from Jiménez -- in action once again after his long layoff -- Fabio Silva or maybe even Patrick Cutrone, back from a pair of unsuccessful loans and scorer of the solitary goal in Wolves’ friendly win over Real Betis.
Patricio aside, there have been no major outgoings to date, although Rúben Neves continues to be linked with a departure. We’ll have to wait and see how things shake out in the last month of the market, where Wolves will probably still add to their own defensive and midfield options.
Much of how Wolves’ season will go would seem to hinge on how well the squad can adapt to a more attacking approach. While there can be a reasonable fear that last season represented the start of a downward curve that might see them in genuine relegation trouble if things don’t immediately click, it seems more likely that Wolves will enjoy a solid (and perhaps even entertaining) campaign that nudges them back towards the top half of the table.
Want to read about another team? The rest of our Premier League season previews can be found here
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com
We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course
Follow us on Twitter in English and Spanish and also on LinkedIn
The latest destination on our traverse around the Big 5 European Leagues sees us land in England. The Premier League pendulum could be swinging back across the North-West, it's halcyon days if you wear claret-and-blue, whilst a couple of yo-yos appear to be unwinding at the bottom.
It's been four seasons since the top of the Premier League table has been this close after 22 games. There was a 25-point gap between 1st and 4th at this stage in 2019/20, it's just seven points in this campaign. Both Liverpool and Manchester City have taken it in turns to race ahead of the pack in recent seasons - some title races have been processions as early as January - but, for now, things remain congested.
With heavy recency bias, the current prevailing narrative is that Liverpool may well have ceded their title defence already. We all know how unrelentingly good they were last season (they were 21-1-0 at this stage) and maintaining those otherworldly standards was always going to be tricky - and that's even before you factor in the injury to Virgil van Dijk (and Joe Gomez, and Joel Matip...) as well as the issues in negotiating a pandemic-condensed fixture list.
It'll raise eyebrows, but there’s virtually nothing between Liverpool’s metrics last season and this. Their attack is identical, creating 1.8 xG per 90 and, ok, shots have risen from 15.4 to 15.6, whilst their defensive metrics have also seen a slight improvement, knocking 9 shots per game down to 7.7, and xG conceded from 1.0 xG to 0.9 xG.
The metrics may be the same, but obviously the results are not. Why the big difference in outcomes?
More detailed analysis on their title win can be found in their season preview by James Yorke, but it boils down to this. Liverpool were extremely good at taking the lead in games last season, scoring first in 16 consecutive wins between November and February, for example. Not only that, they were also very good at doubling their lead once they were ahead. And what happens when you’re 2-0 ahead? You stop attacking as much and your opponents start attacking you more. Your chance creation decreases, your opponents chance creation increases.
This would’ve kept a ceiling on their metrics last season, and therefore those metrics may not actually be a true reflection of how good a team they really were.
This season, they’ve spent far less time ahead and haven’t opened the scoring as regularly, meaning they’ve had to keep their foot on the accelerator more often and their opponents have sat back to try and soak up the pressure - conditions under which you’d expect the metrics to skew more in Liverpool’s favour. So, whilst season-to-season the numbers may look the same, they're the product of very different game scenarios.
Another factor is that Liverpool are scoring slightly behind expectation in both attack and defence in 2020/21. Turns out this team is from Planet Earth after all, despite their near-perfect showing last time out suggesting otherwise.
For all that’s gone wrong at Liverpool, relatively speaking, let’s talk about what’s gone right at Manchester City. The pendulum swung away from City last season as they finished a distant second, but right now it’s looking like it might be their time again. City are top, with the best goal difference, a game in hand, and have just won nine games on the spin.
Which means that their start to the season is already an imperceptible spec in the rear-view mirror. Just 12 points from their opening eight fixtures suggested City might fall short again, but that reaction now looks like it may've been premature, both with hindsight and when you consider they had to face Liverpool, Leicester, Spurs and Arsenal in a tricky set of opening fixtures.
A run of 11-2-0 from their next stretch has seen them muscle their way to the front.
What may've been underplayed is that this is a different Manchester City team to the relentlessly attacking one we've been accustomed to in recent years. Their insatiable desire for goals has been replaced by an insatiable hunger for clean sheets. They're creating less in attack compared to previous seasons, but this is also the tightest they've been defensively in a long time. The trendline below shows the single-game xG values for each fixture and illustrates how consistent they've been at the back. They've conceded more than 1.0 xG on just one occasion.
They're successfully limiting the number of shots their opponents are able to generate, but it's mostly a very noticeable decline in the quality of chances that City are giving up that has driven this improvement - this is the lowest xG per shot conceded that City have recorded in four and a half seasons of StatsBomb data.
The title appears to be City's to lose at this stage, but the red half of Manchester are very much in this race as well. Manchester United are 2nd in the league and are also top scorers with 46 goals. The entire United squad appear to have to magic finishing touch right now with the exception of Anthony Martial and, perhaps surprisingly, Mason Greenwood. Greenwood blazed his way to 10 goals from 3.6 xG last season, but it's just 1 goal from 2.5 xG this. Fortunately, his teammates have taken on the mantle to leave United as a whole ahead of expectation.
Bruno Fernandes has continued his excellent form since signing in January last year, serving up chances for the United attack with the regularity that Marcus Rashford serves half-term dinners. Fernandes has been the creative cog and link player between the midfield and attack that had been missing from the United side for years, and his performances have put him firmly in the conversation for the end-of-season player of the year award.
Leicester are third and at the speed they’re picking up points are rightly considered contenders, though they’re coming off less impressive metrics than their competition. Their expected goal difference is just above zero without penalties, but that comes with the caveat that they haven’t had to be as productive in open play due to their rate of winning spot-kicks, which at a 1-in-every-2-games rate is currently the highest in the Premier League and also the highest since the 2016/17 iteration of Leicester City.
What Leicester do have in their favour, and the secondary explanation for their strong points tally despite the average metrics, is their record when taking the lead. The Foxes have gone in front 15 times in matches this season and conceded equalising goals on just two occasions. They’re at their best when ahead in games as seen in the positive-game-state shot maps below. They still create more than the opposition despite having a lead to hold onto. It’s also notable that six of their ten penalties have been won when they're ahead, allowing them to settle nerves and secure victories.
A collection of over and underachievers exist in the smörgåsbord between 5th-10th. Even now, ignoring pre-season expectations, if I asked you to put Arsenal, Aston Villa, Chelsea, Spurs, and West Ham in order of where you think they lie in the xG difference table, I bet it wouldn’t be this:
There’s also a deserved word to be said on Leeds too. People can’t seem to stop talking about Marcelo Bielsa's side this season and it’s not hard to understand why: they've come up from the Championship, changed nothing, and the result is the fourth-best attack and the second-worst defence by xG. Leeds games are box office; their press like a 90-minute swarm of angry Yorkshire bees, with a regular flow of chances at both ends of the pitch. Leeds' PPDA of 5.9 is a Premier League low in the last four seasons.
Lastly, the bottom of the table where three teams find themselves cut adrift of the pack. Just three points separate Fulham, West Brom, and Sheffield United between 18th-20th, though that gap was larger before Sheffield United picked up three wins in their last five.
There was talk of the Blades’ possibly becoming the Premier League’s worst ever side before that renaissance, but it’s newly-promoted West Brom who would be the strongest candidates for that dishonour on current showing, had they not already surpassed 2007/08 Derby County’s points tally. The Baggies currently have the worst xG difference per game and the worst xG conceded of the past four and a half seasons of StatsBomb data. And as the ancient football proverb goes: what yo’s up, must eventually yo down.
Join us at StatsBomb Evolve on March 17th 2021 to discover how we're going to change the football industry, including the launch of our new data product, StatsBomb 360. To find out more, click here.
We are five matchdays into a Premier League season that has already provided more than its fair share of incident and excitement. Here are some emerging trends. Goals, Goals, Goals Goals have been flying in from all angles, producing a series of seven-goal thrillers, five-goal thrashings and 3-3 draws that have enlivened the early running. With penalties included, there have been 3.58 goals per match. The total of 172 is 16 more than any other 48-game stretch in our data set (starting from the 2016-17 season). Will it continue? Probably not. Either through player adaptation or a more relaxed interpretation of the rules, the current rate of a penalty goal for every two matches is likely to eventually settle down to somewhere nearer the rate of one every three matches that is now pretty much standard across the major European leagues. It is also notable that non-penalty goals are running abnormally far ahead of expected goals (xG). The total of 147 is over 130% of the xG sum of 112.42. The furthest ahead that goals ran of xG in any consecutive 48-match sample last season was 111.35%. From 2016-17 onwards, the highest mark prior to the start of this season had been just over 120%. There are plenty of larger swings in smaller sample sizes, but in one of this size there has never previously been a swing of much over 20% at either end of the scale. All of which suggests that things will begin to slow down a bit as the campaign progresses. Is Anyone Good? It’s a fair question to ask. Liverpool and Manchester City were the clear pre-season favourites to occupy the top two positions, but both of them have already suffered heavy defeats -- City 2-5 at home to Leicester; Liverpool 2-7 away at Aston Villa. Manchester United have lost twice in four matches, including a 1-6 thrashing by Tottenham Hotspur; Chelsea are still just as open as they were last season and have carried a negative xG difference through their opening five fixtures. We could go on. Mikel Arteta has now had nearly a full season of matches on the Arsenal bench and while they do look more structurally sound to the eye, their numbers are still middling, and their shot volume is still poor. Leicester are fourth but have mediocre metrics. Only Aston Villa and Everton have traversed the opening weeks of the season undefeated. Spurs have very good metrics, but as with everything at this stage, it is too early to tell if that really means anything. Particularly so given that their numbers were so poor last season, and with the boost of that 6-1 win over United (3.07 to 0.16 on the non-penalty xG) to account for. West Ham have also looked good, as have Brighton by the numbers but less so by results. It will take at least another 5-10 matches for everything to shake out and give us a clearer idea of which teams are genuinely good and bad, maybe longer given the strange circumstances of the season. Once we get to that point it wouldn’t be at all surprising to see Liverpool and City again emerge as the league’s most impressive teams. Even with that Villa defeat on their record (a 3.91 to 1.77 defeat on xG), Liverpool have been so dominant in their other matches that they still have an overwhelming positive shot and xG difference. Their system is now arguably strong enough to absorb Virgil van Dijk’s absence, particularly once Alisson returns in goal. City’s attacking output hasn’t been great to date, but Pep Guardiola has done plenty of tinkering with the lineup and can be expected to eventually settle upon a workable formula. James Rodríguez Shines Everton’s purchase of James Rodríguez from Real Madrid was one of the more enticing off-season arrivals to the Premier League, and he has hit the ground running with three goals and three assists in five appearances. That is well ahead of his expected output, so he shouldn’t necessarily be expected to continue producing direct goal contributions at that rate, but even his underlying output has so far represented a clear improvement on the creative options Everton had last season. The sample size is very small, and so some variation should be expected over the course of the campaign, but on a per 90 basis, Rodríguez currently leads Everton in key passes (both from open play and set pieces), xG assisted and passes into the penalty area, in addition to also being the player who has most often moved the ball forward into the final third. All that while also getting off nearly 2.5 shots per 90. Reunited with Carlo Ancelotti, he has immediately become the team's attacking hub. Last season, no Everton player provided over two key passes per 90; Rodríguez has so far produced three. We certainly know a guy who’d relish the opportunity to work with a player capable of such quality deliveries from set pieces, particularly with a target like Yerry Mina on the other end of them.
Everton have so far combined good results with very solid metrics, and that despite playing both Spurs and Liverpool across their opening five fixtures. As we've already said, we need a larger sample of matches before we can really start to draw any sort of meaningful conclusions on team quality, but Everton’s underlying numbers were better than results last season, and with Rodríguez on board, top eight should be their minimum aim.
Will this be the last season we see Pep Guardiola in English football? If it is, the rest of the title-aspirants will not mourn his departure. Arguably it’s become harder than ever to win titles during his tenure. Of course, City only have two titles from his four seasons as Antonio Conte’s Chelsea took the top prize in 2016-17 and Liverpool won it last time round. Within this era, five of the six highest points totals in the Premier League era exist: two each for City (100 in 2017-18 and 98 in 2018-19) and Liverpool (97 in 2018-19 and 99 in 2019-20) and Chelsea’s 93 in 2016-17. Is this the new normal? Or will we see a decline if Guardiola departs? What is certain is that the combination of it possibly being his last season in Manchester plus a comparatively mediocre 81 points last season should be adequate motivation for Guardiola to set about reclaiming the title. To the consternation of some, once more, the metrics were extremely good but the outcomes were decidedly split. Two Manchester Cities operated in 2019-20. One version laid waste to defences and racked up simple, dominant wins without needing to get out of second gear. Watford gave up twelve goals in two defeats, Villa, Brighton and Burnley all gave up nine. Elsewhere, they couldn’t best Manchester United or Wolves across two games, and contrived to lose to Norwich. The sheer weirdness of annihilating Tottenham twice yet getting just one point can be filed under “football has elements of randomness” too. For the record in those two games, they recorded 48 shots to Tottenham’s 6, and a huge 6.3 to 0.5 in expected goals. Even in defeat, frequently, the elements of good process were right there. Some of the most eyecatching defeats involved red cards. The 3-2 defeat to Wolves at Christmas showed all their quality--they went 2-0 up despite losing Ederson in the 11th minute before giving up the lead late on. Not good was Chelsea’s convincing 2-1 June victory was helped by Fernandinho’s 73rd minute dismissal from where the Blues turned the screw and City didn’t muster a shot. March’s 2-0 defeat to United was pretty tepid too. As ever they dominated the ball but unusually created very little (just seven shots). The bottom line is that City’s record against teams in the eventual top seven (usual big six plus Leicester) was insufficient to see them challenge long and late into any season, let along one in which Liverpool barely dropped points. With a record of 4-1-7 with five out of six away defeats here, City just weren’t able to consistently dominate and beat the better teams in this league. This is where they were so strong in the previous two seasons (9-1-2 in 2017-18 and 9-2-1 in 2018-19) and where they faltered in Guardiola’s first season (2-5-5). In the big picture, City's metrics were great; but then again they always are. But when we drill down to game by game (as the chart above shows) we can see that City's problem was as much aligned to the variation in their performance levels in games as anything else. Look at 2017-18: nearly no games with over one expected goals allowed and only one game in which their total was exceeded. 2018-19 was pretty good too but with a smattering of games in which the defence looked more vulnerable. 2019-20 has a weird mix of high xG for/low xG against games and a bunch of matches in which the team is either outperformed, gives up a lot in defence or it's pretty close. Bottom line, there was less consistency and that's partly how they contrived to lose nine games having only lost twelve combined in the previous three seasons. The dichotomy between both shot quantity and shot quality at both ends of the pitch was fairly remarkable too. As ever City took a lot of shots. Their 19.5 per game was higher than any of their previous seasons and evoked Carlo Ancelotti's 2009-10 Chelsea team, which was the last Premier League team to get anywhere near this kind of volume. They were good shots too, with the xG per shot value of close to 0.12 only a rizla-width behind that of Arsenal's own 0.12. Loads of shots of very good quality equals loads of goals, 102 in fact, which tied their 2013-14 total and was only four goals behind that of 2017-18. As ever, all their attackers had decent contributory seasons, and eight different players played more than 900 minutes and contributed between 0.49 and 0.98 goals (or assists) per 90 minutes. It's daft really. The varying outcomes and lack of Champions League success will mean that it may take time to really reflect adequately on what Guardiola created in this team, but the attacking unit remains utterly stellar. But the defence. The defence. Seven shots against was up from six in the previous two seasons. Such volume remains are ludicrously good and again speaks of good process. However, the value of these shots was league high at 0.12, significantly up from all of Guardiola's three previous seasons and essentially the same value as the shots they were taking. So Manchester City, one of the best attacking teams in world football, gave up the same shot quality as they take themselves. That's a problem. Throughball vulnerability was a part of this. Across 2017-18 and 2018-19 City gave up 21 shots from throughballs and three goals. In 2019-20 alone they allowed the same volume of shots against (21) but this time gave up seven goals. This kind of vulnerability is a frequent weak spot for pressing teams, and it is to City's credit that they managed to resolve a weakness here after Guardiola's first season.. 2019-20 was not successful here though, and Kyle Walker and Ederson apart, the whole backline has felt erratically assembled for a while. The obvious switch at the back was that (then) 34 year old defensive midfielder Fernandinho was now a starting centre back, and with Aymeric Laporte out, the rotation grew more frequent. A stalwart back in 2017-18, Nicolas Otamendi has become a player who is in and out of the team while John Stones appeared to fall right down the pecking order with Eric Garcia ending the season with a bunch of starts. Stones' demise is curious but possibly not undeserved. An experimental metric I calculated to establish how frequently defenders were actively clearing their defensive zones versus the amount of successful plays their opponents were making saw him rank dead last of all centre backs in the league, and by some margin. As the new man in midfield, Rodri was eyed as a potential culprit towards this new found defensive instability, with ideas that he had yet to perfect the dark arts of Fernandinho, but this was pretty speculative. On top of this City once more conceded some very stupid goals, twice seeing Ederson bypassed from 40 yards or so, from Che Adams and Scott McTominay respectively. Mo Salah did the same a couple of years back and they're such horrible goals to give up, one would think that there were tweaks that could be added to the philosophy just to avert this. Personnel For a club that has spent widely, and acquired Raheem Sterling and Kevin De Bruyne in one summer window back in 2015 there's a nagging feeling that in recent seasons City haven't quite nailed their transfers. Admittedly, when a team generally performs as well as this one has across Guardiola's reign, it's tricky to purchase players that intrinsically improve the starting eleven. Case in point being Riyad Mahrez who was the one main signing in the summer of 2018. He has been a top tier contributor to goals when on the pitch in both his seasons at the club but is no guaranteed starter. He would likely start every week for any other club in the league (Liverpool could squeeze him and Mo Salah in somehow) and in 2019-20 was extremely effective and slightly unsung as De Bruyne in particular rightly took the plaudits:
So far this summer in the transfer market, there has been a move to shore up defence with a novel idea: "buy a relegated centre back". Arriving for around £40m, Nathan Aké has long appeared the type of player that could move to a higher level than Bournemouth, not least because he is so comfortable on the ball. As a left footed centre back on the cusp of his peak years, he was a relatively scarce commodity and does actually scope out as a fairly capable defending defender on top of passing reliability. It's hard to see him as a direct upgrade who will slot instantly into City's first eleven, but no doubt he will get minutes, and ease the pressure Laporte had to shoulder as the one natural left sided centre back. There is talk of Kalidou Koulibaly arriving, but at this stage gossip around him is as reliable to the transfer window as Santa Claus is to Christmas, so it's hard to be confident here. If the past was represented by the sad sight of Leroy Sané returning to his homeland, the future is now here in the form of 20 year old Ferran Torres, who arrived from Valencia. His former team struggled extensively in 2019-20, and it had an impact on the numerical impact of much of their squad. As such we have a tricky chicken vs egg problem when evaluating him just from numbers; as we well know, the video is required too:
The vast majority of Torres' career has been spent playing off the right side, and how he fits into this team is hard to answer. Raheem Sterling plays off the left, so not there, while the lefties Mahrez and Bernardo Silva tend to operate off the right and Phil Foden is the breakthrough talent already in the squad and covering both these positions. Foden's development really should not be stymied and in particular he shows up extremely well for finding space in the final third. There is likely a positive skew that was assisted by playing often in the post restart fixtures, which were played at a noticeably kinder pace than what came before, but Foden's final third ball receipts were considered "under pressure" on only 6% of his receipts, the lowest in the league by a good margin:
What that chart highlights too is the real hole that has opened up in this squad this summer: the departure of club legend David Silva. Again his presence in midfield last season at 34 years old poses questions about how readily opponents could get through City, but undoubtedly, his on the ball influence was far from dimmed. A rock solid contributor to Guardiola's possession game, and an utterly reliable secondary creator for the team behind De Bruyne, he was still good for 22 league starts last season. Who replaces him? The squad is deep enough to take the hit, but it's not as if a clear answer is wildly evident. Early season we may see some more Guardiola flexibility in selection, this time in midfield rather than just defence. Projection 78, 100, 98, 81 divided by 4 equals 89.25. That's Guardiola's four Manchester City seasons. Sporting Index opened up at 88.5-90 which the eagle eyed amongst you will notice has a midpoint of 89.25. The metrics are still there for this team. They won 26 games last season, and for various reasons ended up towards the bottom end of any realistic point expectation. That doesn't necessarily mean they are primed to overhaul Liverpool who have managed to be consistent as well as dominant in the last two seasons, much as City did before and alongside them in 2017-18 and 2018-19, but they do have the tools. City need to reclaim consistency and perhaps with a crazy schedule for all, Guardiola may be wise to flip the off switch more often when his team are coasting and games are won. Enough of the metrics that undermined City in 2019-20 land in the high variance category, and the functional control this team can still exert over the vast majority of their opponents will surely see them go very deep into the season and contend. Would I bet on them to win it? I don't think so. That's not to say they can't, they are betting favourites with good reason but City have been the metric darlings of nearly every season in the last decade, yet they've won just four titles. You need to be both very good and have things also go your way to win this league. If Guardiola is to leave though, he'll no doubt want to go out on a high.
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course Follow us on twitter in English and Spanish and also on LinkedIn
Wolves, or to give them their full name "Little Portugal", have very quickly become a Premier League club that can match up against any team in the league. Across two seasons, they have proven to be consistent, resilient and well worth their consecutive seventh placed finishes. Of course, like many teams outside the television big six, Wolves harbour ambitions of more than that, and may be slightly frustrated that they couldn't exploit the down years for both Tottenham and Arsenal and do what Leicester did in 2019-20 and finish higher up. At least in part this inability to broach the gap upwards can be attributed to a slightly slow start. Wolves had a 2-7-2 record after a 1-1 draw at Arsenal on 2nd November, and while there were good results in there, including a 2-0 win against Manchester City and draws against Manchester United and Leicester City, it's likely that this period was that which impacted the final table most. Was there a problem? Not really? Metrics weren't at their strongest during this period as we can see from Wolves' two season expected goal trends, but there was good reason: Belfast, Yerevan, Turin, Istanbul and Bratislava formed a Europa League tour for the club during a period in which they played as many non-league fixtures as league (eleven each). With a relatively small squad and a fairly consistent first team, it's actually to the team's credit that they got through that opening period of the Premier League without further decline. The big story in the chart above is the defence. In the back half of 2019-20 Wolves gave up under one non-penalty expected goal in a game fourteen times. In eight of those games they allowed under 0.5. Wolves post-lockdown defence was structurally the best in the league. Wolves have achieved this defensive stability by being one of the most clear-eyed tactical teams in the division, and a world away from the hard pressing espoused by other managers. That Nuno Espirito Santo is both Portuguese and a former goalkeeper for Jose Mourinho is highly indicative of Wolves style of play:
Nuno's set-ups are well established at this stage. Sixteen teams used three centre backs at some point last season [t's funny how these fashions pervade] yet only two of them set up with three centre backs in every game, Wolves and Sheffield United. The core of this team is well established too but will undergo some kind of remedial work going forward thanks to the absence of two key parts: the full backs. Stalwart Matt Doherty has joined Tottenham while Jonny has sustained a long term ACL injury and is likely to be unsighted for many months. In 2018-19, Wolves fielded 14 players for over 900 minutes in the league and in 2019-20 it was 13. They have operated with a very small consistent squad for two full seasons and are now obliged to adapt especially on the flanks. On the left Ruben Vinagre may have felt he was stepping out of the reserve slot he's held for two seasons but the signing of 31 year old Lyon rotational full back Marçal suggests his role may well persist for now. In the interests of sample size, this visualisation covers three seasons; he's experienced and was playing in the Champions League three weeks ago, but he will need to show robustness and reliability as being a Wolves full back is not a position for the underinvolved.
Never knowingly under-involved is Adama Traoré and the early word is that he may end up covering Doherty's role, at least to start the season. He's well capable of playing this position and anywhere else on the right flank and persists in being a unique and curious player. Last season saw a further rise to prominence outside niche stats fans fawning over his dribble volumes. Firstly, his four league goals were spread across three league games, Manchester City home and away and Tottenham. No better way to attract attention than scoring key goals in big games. However, more encouragingly, we saw further end production in his creative numbers. Nine assists was big-boy output and not wildly ahead of expectation. All in all for players with a decent volume of minutes (say 1500+) he was top ten in the league for open play key passes (1.6 per 90) and resulting xG Assisted (0.21 per 90). The reliability of how he made those goals was hugely positive. When you have a thing and you can keep doing your thing and nobody can stop you doing your thing, then you've really got something! But if it doesn't result in goals, your thing might not be that useful. Traoré's thing used to be facing up a defender and then killing them for strength and pace to make space for a further pass. However, the further pass wasn't necessarily a thing too. Now he's showing in the Premier League that goal creation is his thing. Beat a man on the flank and find an attacker in the box. For the defender it's nightmarish: they know what is coming but not if they can stop it. He's probably stronger than you and even if he isn't he sure looks like he is. Check out his very specific chance creation here, and how all his assists have a common thread:
Part of the Wolves oeuvre is ball carrying at pace. It's not just Traoré either. Some experimental numbers I generated for speed with directness of ball carrying saw Diogo Jota rank fastest in the league, with Raúl Jiménez ranking fairly high too. Ironically, Traoré ranked a way back for this specific measure, which was likely a function of his tendency to stand up to the defender and essentially stalling before moving, but he ranked top in the league for average distance per carry into or inside the final third, with Jota third. Wolves as a whole attempt more actual dribbles (~take-ons) than any other team in the league, and carry the ball for longer distances everywhere high up the pitch right up into the box. This is a team that relies strongly on attackers that can transition through zones with the ball at their feet, as we can see when we look at longer carries for those three players:
Elsewhere it's intriguing to know whether the club has secured the future of the Portuguese national team in Vitinha and Fábio Silva or attached around £60m to pure potential that will not pay off for some seasons. Each significantly lacks experience of good quality men's level football and thus eludes anything but the most scant review in data. Vitinha at 20 years old has played over 1000 minutes in the Portuguese second tier and grabbed fleeting substitute minutes in the second half of Porto's season, while Silva at just 18 years old had a similar profile in Porto's first team with time on the pitch minimal and mainly via the bench. The impact each can make is difficult to know, and it seems hard to envisage that either will quickly shift the established first teamers out, but they do bring much needed depth to the squad in general and it will be hoped that they can settle and contribute quickly. Projection In a compressed season with very little room for recovery, Wolves will probably reflect that a lack of European football is a real boost to their Premier League chances. Pre-season points projections concur with their seventh placed finishes and expect another season in the nominal "best of the rest" slot. Nuno's style has kept Wolves hard to beat, but there were almost as many draws as wins last season and a shade more attacking power might be the recipe for really pointing upwards and shaking up the top six mix. With a smaller squad and specific and key contributors, injury luck will always have an impact. With that, Wolves are as reliable as they come, and the range of outcomes for this team is likely a lot smaller than that of certain others eyeing similar positions in the table such as Leicester, Tottenham, Arsenal or Everton. It's a good time to be a Wolves fan, and should continue to be so.
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course Follow us on twitter in English and Spanish and also on LinkedIn
In two years since his return from Sweden, Graham Potter's career has progressed extremely well. A stylistically eye-catching year at Swansea was enough to attract Brighton’s attention and the transition from Allsvenskan to Premier League was complete. A (young-ish) English manager had trodden an unfamiliar road before being (kinda) fast tracked to the top? What will the papers say? Potter’s predecessor, Chris Hughton had brought Brighton to the Premier League and kept them up for two seasons. The remit to stay in the division had been fulfilled, at least, for all that the 2018-19 season cut it far too fine, with a run of no wins in the last nine games ceding control to those around them. You only need three worse teams, and there were three worse teams-- Cardiff, Fulham and Huddersfield--but effectively chance was the dominant factor that kept them in the league. Owner Tony Bloom, a man with a solid history in probability, will have noted bottom five metrics including a bottom two ranked attack. With some justification, one can imagine those involved in the decision making process envisaged future seasons of a similar ilk. The defensive onus brought by Hughton meant that scrapping in the bottom half was likely to persist, despite quite extensive recruitment. So could the appointment of Potter change that dynamic? It represented a clear gamble regardless. One thing was certain: Potter would get this team playing a different style of football to Hughton. Far more orientated towards possession and far less reactive. One thing was uncertain: would this change of style keep Brighton away from the trapdoor into the Championship? The topline conclusion is yes. A haul of 41 points was five points clear of 2018-19’s total and one ahead of Hughton’s first season, while a fifteenth place finish matched 2017-18. There were runs where the team struggled to win as six draws and zero wins in nine games pre-lockdown in 2020 testified but they never quite got hauled into the relegation mix. Nine wins overall was… the same as Hughton managed in both his seasons. So has Potter been a success? The metrics have improved, but we do have to recognise that with five games to play, they were nine points clear of relegation, so all but safe, and soon after walked into Manchester City on a good day and shipped over five expected goals, a total only exceeded all season when City thumped Watford 8-0. The backend of the season saw metrics decline, but largely in fundamentally meaningless fixtures: The broader picture is better. In raw terms Brighton’s per game xG differential worsened from -0.43 in 2017-18 to -0.50 in 2018-19 and at the point of safety in 2019-20 was improved at -0.24. There’s enough in that to suggest that Potter’s vstyle has had some effect, and we can see how the line between expected values for and against closed as the season progressed. It’s still far from stellar but it’s enough to feel Potter’s second season is deserved. He has made a small trade off in defence to give the team something to work with in attack, and when we consider that Hughton’s attack was close to league worst, we can understand the purpose here. Hughton's focus on defending was certainly reactive; witness the position of his sides in this table of blocked shots from teams in recent seasons:
...and see how Brighton's passes per defensive action has evolved away from a complete bunker under Hughton towards something more balanced under Potter:
For personnel, the transition from the first season in the league to the second was stark. Lots of relatively low priced players arrived from a wide variety of leagues in year one--and few made the team with any regularity. It was easy to wonder if Hughton, detached from recruitment, was interested in playing the players he rated rather than new signings, but Potter has followed a similar thread, and few of that first tranche, Dan Burn, Yves Bissoma and Martin Montoya apart featured heavily in 2019-20. Potter’s first summer saw more targeted signings and less variety with all but Leandro Trossard arriving from within the English game. Adam Webster, Neal Maupay and Trossard all featured heavily and validated their purchases while Aaron Mooy eventually converted from a loan and Tariq Lamptey was a neat January pickup who quickly featured. Brighton have moved quickly again this summer adding distinct quality--with a sideline of injury concern and post-peak aging--in Adam Lallana. With Mooy heading for Shanghai, it appears reasonable to assume that Lallana will effectively take his place in the squad and overall this summer feels more like squad evolution than revolution. Joël Veltman arrived from Ajax for a tiny fee and the bones of Wigan were examined to take young midfielder Jensen Weir. Veltman’s arrival, specifically, gave Brighton a wildly overstocked centre back department. Adam Webster and Lewis Dunk were the main starters last season with Shane Duffy as an alternative. Ben White spent the season starring on loan at Leeds, while Matt Clarke similarly had a great time on loan at Derby. Dan Burn converted into a left back to get game time, but negotiating this corps ahead of the new season is not easy, with all mentioned likely worthy of playing regular football at a level near to that which Brighton require. White finally signed a new contract recently while Clarke returned to Derby on loan and Duffy headed across the border to Celtic, but there's still a surplus here. Either another move is imminent or someone's going to miss out on substantial minutes. Overall though it feels like there's a degree of continuity being sought. Investments in players of the right age to progress in the long term have been made over time. A good example is Neal Maupay who followed up his 25-goal Brentford season with 10 in the Premier League. He scopes out at around a one in three striker here, which is closer to average than you may think, and adds further value from the volume of defensive work he does leading from the front. He's an interesting reference point for other strikers making the step up in leagues:
Projection Improved metrics under Potter have led to a projection of safety in the Sporting Index opening lines. They landed at around 42-43 points here which would represent a perfectly adequate return. However, one has to wonder what the long term plan is? It's still early days for Brighton in the Premier League, but projecting to low 40s still puts a team in the realm of risk for relegation. As such, it feels like from a performance evaluation perspective Potter may need to eke out further progress, if not this season then soon. His first season was entirely fine, and avoiding any relegation battle should be the target again, but Brighton aren't quite good enough yet to ease off the gas and naturally land mid-table security. In recent seasons, Bournemouth laid a useful blueprint for a smaller club playing a style that backed their on ball ability, but also their inability to adapt that over time eventually saw them vulnerable in a down year, and they paid the price. Brighton and Potter would do well to heed the lessons dealt out there.
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course Follow us on twitter in English and Spanish and also on LinkedIn
Leeds are back in the top flight after more than a decade and a half away and the consensus seems to be that they are here to stay. Both the mainstream bookmakers and the spread betting markets have them in line for a mid-table finish, and they’ve certainly spent like a team determined to reestablish themselves as a genuine Premier League outfit. James has already written an excellent piece on the club’s promotion campaign, so we won’t go into too much detail about that here, but here are the salient points. Leeds had the best metrics in the Championship for two seasons in a row. After failing to convert that into promotion in 2018-19 due to a bad run of results down the final stretch, they did so in 2019-20, going up as champions. The club’s defensive record was particularly strong. They conceded just 35 times in 46 matches and also had the league’s best metrics. They gave up fewer shots (8.78 per match) than any other side, and the average quality of those shots was also one of the lowest in the division.
That was partly down to the efficiency of Marcelo Bielsa’s patented all-pitch press. Leeds were relentless in harrying opponents over the length and breadth of the field.
In attack, they comfortably outshot everyone, and did so while still maintaining one of the highest average shot quality marks in the league. All that off-ball movement, the pretty patterns and interchanges, helped to create a high volume of shots from good locations.
Whichever way you look at it, Leeds were a dominant team, taking 65% of the shots, accumulating 68% of the xG and scoring 69% of the goals in their matches. A clearly defined play style, strong overall metrics and a solid defensive base are all generally predictive of a good first campaign in the top flight. To illustrate this point, over the last 10 years, only two teams have been promoted with a better average goal difference per match than Leeds’ figure of 0.91: Newcastle in 2016-17 and Wolverhampton Wanderers in 2017-18. Like Leeds, both conceded under a goal per match during their promotion campaigns. Both went on to only concede around 17% more goals in the Premier League, compared to an average for promoted teams of around 40%, and both finished in the top 10. Leeds have a good base to work from, and they've also been able to retain almost all of the important squad members from their promotion campaign. Hélder Costa and young goalkeeper Illan Meslier, who got a run of games as first choice after lockdown due to Kiko Casilla’s suspension for racist language, have both had their loans turned into permanent deals. Jack Harrison will stay on board for another season on loan from Manchester City. The one key departure is that of Brighton loanee Ben White. He started all 46 matches last season, and became a key figure in the centre of defence, impressing on both sides of the ball. Reports suggest that Leeds were keen to sign him permanently, but Brighton instead decided to bring him back into the fold there. It was, therefore, unsurprising that one of the club’s first off-season signings was Robin Koch, a 24-year-old central defender from Freiburg in the Bundesliga, capped three times by Germany. White was the more proactive defender in his regular partnership with Liam Cooper, stepping forward more often to contest possession and intercept, both on a outright basis and proportionally to all defensive actions when they started together in the centre of defence. Cooper cleaned up behind and was a dominant aerial force. Koch seems to profile more like Cooper than White, with a lesser proportion of proactive defensive actions than the man he replaces. But it is important to add some context here. Freiburg were one of the most passive and deep-defending teams in the Bundesliga last season, and Koch often played as the central of three central defenders. In such a system, it is entirely normal for the player in that role to be the least proactive of the three. As such, it is hard to get a handle on just how apt a White replacement he is from the data. Their passing profiles are more similar, and Koch does seem to have solid range there. Qualitative scouting has presumably left the recruitment team confident that he can slot into White’s role. Cooper was again the less proactive element in his partnership with Pontus Jansson in 2018-19, so there is little evidence he’d be capable of defending in a different manner. As well as Koch, Leeds have also signed a trio of younger players in Charlie Allen (16), Cody Drameh (18) and Joe Gelhardt (18), the latter of whom posted good shot volume in a very limited sample size of just over 500 minutes at Wigan in the Championship last season. But the big money has been spent on the club-record, £27 million signing of Rodrigo from Valencia. Readers with good memories may recall Rodrigo from his one-season loan at Bolton Wanderers back in 2010-11, but the player who arrives at Elland Road has developed considerably since then and is now a full Spain international. He, like the entire Valencia team, didn’t have a great time of it last season, but in 2018-19, he provided a goal or assist for every 180 minutes he was on the pitch in terms of both real and expected output.
Valencia were a counter-attacking team and Rodrigo was often a key part of their forward transitions, dropping off the front to receive, turn and combine with onward runners. Quick and intelligent in his movements, he has a skillset that should mesh well with Bielsa’s system, even if, as with Koch, he comes in from a team considerably less aggressive out of possession than this current Leeds side. The age profile isn’t great, but he should prove to be a direct improvement on Patrick Bamford at the centre of the attack. Leeds have also shown strong interest in Rodrigo de Paul, a creative midfielder who has been pretty consistently good at Udinese over the last couple of seasons. He doesn’t provide a great deal of defensive output, but the old Bielsa motto of it being easier to teach a talented player to run than make an energetic player talented may just ring true here. He would certainly represent an upgrade in one of the two more advanced midfield roles in the 4-1-4-1/4-3-3 that has served Leeds so well over the last couple of seasons.
Maintaining a stable squad while upgrading in key positions seems a sensible way of confronting the club’s first Premier League season in 16 years, even though there is still a degree of uncertainty in terms of the performance levels that can be expected from some of the existing squad members. Many of these players were, after all, involved in a mid-table Championship finish in the season before Bielsa’s arrival. Any individual deficiencies will be that much clearer in the top flight. With that said, the front-foot style of play of the collective should be sufficient to get Leeds enough points from some of the league’s blander teams to provide a solid platform for a mid-table finish. Theirs is an exciting approach and it will be intriguing to see how they fare in their attempts to go toe-to-toe with teams towards the top of the table. An opening day trip to Anfield and champions Liverpool will give us an early taste of what to expect.
If you're a club, media or gambling entity and want to know more about what StatsBomb can do for you, please contact us at Sales@StatsBomb.com We also provide education in this area, so if this taste of football analytics sparked interest, check out our Introduction to Football Analytics course Follow us on twitter in English and Spanish and also on LinkedIn