The 2022 PFA Team of the Year has now been announced and it features some of the Premier League’s biggest names, but does the data justify their selection in the XI? In order to determine who’s been this seasons top performers, football analysts at BetVictor have taken matters into their own hands.
Using data scouting techniques, the BetVictor supercomputer has crunched the numbers assessing every Premier League player by their statistical performance over the 2021-22 season.
Players have been given a “Performance Rating” which is a measure (on a scale of 1-100) of how well they have performed compared to other players in a similar position. A score of 50 would mean a player is performing averagely.
The algorithm is designed to find the starting XI with the most optimised output per 90-minutes. Put simply, it is eleven players who do a lot of good things the most frequently. More details on how this was calculated can be found below, but in the meantime, here’s the Team of the Year according to the data.
BetVictor's Stats Based Premier League Team of the Year
Goalkeeper - Alisson Becker
Goalkeepers were ranked on shot-stopping performance, distribution, command of box, sweeper ability, and 1v1 ability. The best goalkeeper in the Premier League according to the stats is Liverpool’s Alisson. His 75.3% save percentage is the 3rd best rate in the league, he also ranked 2nd for pass completion with 54.2%, just behind his Brazilian teammate Ederson.
Alisson’s post shot expected goals conceded minus actual goals conceded is +4.1. This means he has conceded 4.1 goals less than expected, indicating his superior shot stopping ability. He is the joint 4th best ranked goalkeeper for this statistic. Furthermore, according to 1v1 goalkeeping models built by John Harrison (@Jhdharrison1), Alisson is also the best in the league in these situations, with the 29-year-old making +8.21 more saves than expected.
Right-Back: Trent Alexander-Arnold (Stats Based Premier League Player of the Year)
According to the data Trent Alexander-Arnold is not only the best rated right-back in the league, but his overperformance compared to other players in his position mean he is our Stats Based Premier League Player of the Year.
Trent ranks 2nd in the league for expected assists per 90 with 0.35, whilst also being the top-rated English and defensive player for this statistic. He’s also the top ranked player in the league for completed passes into the 18-yard box with 2.65 per 90, whilst his 8.42 progressive passes per 90 is second only to Liverpool teammate Thiago Alcántara. Trent is the 6th top performing player in the league for possession-adjusted interceptions with 3.23 and makes the 5th most shot creating actions per 90 with 4.73.
Centre-Back: Joel Matip
Joel Matip has had an excellent season and according to our algorithm is the best performing centre-back in the league, outperforming his counter-part Virgil van Dijk in several key areas.
Matip has the 3rd best rate of any centre-back for progressive passes with 4.94 per 90. He ranks 6th among centre-backs for passes into the final third with 6.13 per 90. He’s outperformed Virgil van Dijk for blocks per 90, blocked shots on target per 90, interceptions per 90, along with making three times more possession winning tackles of the ball than the Dutchman.
Centre-Back: Rúben Dias
Manchester City’s Rúben Dias was the second-best ranked centre-back, and for all his outstanding defensive attributes, Dias is in fact the best ranked centre-back in the Premier League for expected goals plus expected assists (xG+xA) with a rate of 0.17 per 90.
Additionally, Dias has the 2nd best pass completion percentage in the league with 94.3%. He also has the 2nd best successful pressures percentage of any centre-back with 44.4%, indicating his excellent judgement and composure.
Left-Back: João Cancelo
The versatile João Cancelo, who is as equally effective on both sides of the pitch, was rated the Premier League’s top left-back. The Portugal international is an exceptional ball progressor and makes the most carries into the final third of the pitch of any player in the league with 3.84 per 90, as well as making the 4th most progressive carries with 10.1 per 90.
In terms of his defensive output, his 2.38 possession adjusted ball winning tackles are the best achieved by any full-back in the topflight.
Defensive Midfield: Bruno Guimarães
It was very much a surprise to see Bruno Guimarães rank ahead of the likes of Rodri and Declan Rice in this position – for the record Rodri ranked in 2nd place. However, after taking a deep dive into his performance data it appears as though Newcastle United have a bought a player in which their takeover ambitions can be built around.
The 24-year-old is the top-rated defensive midfielder for non-penalty expected goals with 0.27 per 90. His 0.96 shots on target per 90 is a better rate than that achieved by teammates Callum Wilson (0.84) and Allan Saint-Maximin (0.83) – meaning the Brazilian brings a deadly goalscoring threat to his game. Something which has been spotted at international level, with him becoming a regular in Tite’s squad ahead of the 2022 World Cup.
He also makes the 10th most successful pressures in the league with 7.22 per 90, along with the most possession winning tackles of any player with 3.39 per 90, meaning he offers exceptional intensity and aggression, making him the perfect profile for the Premier League.
Central Midfield: Kevin De Bruyne
What can be said about Kevin De Bruyne that has not already been said? Having already claimed Premier League Player of the Season along with winning the PFA Players’ Player of the Year award for the past two seasons, there would not be too many surprised to see him do it again.
KDB was by far the best performing midfield player, achieving a performance rating 11.93 points higher than anyone other player in this position. His rate of 5.54 shot creating actions per 90 is the best in the league, he makes the most key passes per 90 at a rate of 3.39 and is the top-rated player for expected assists with 0.39 per 90. He also outperforms his non-penalty expected goals by +0.37 per 90 (non-penalty goals minus non-penalty xG) which is the best of any player.
Central Midfield: Thiago Alcántara
Liverpool’s Thiago has been a creative spark in midfield, helping The Reds to create space, control the tempo, and break opposition lines with his excellent movement and range of passing. He is the top ranked player in the Premier League for progressive passes with 10.3 per 90.
The Spaniard’s 75.6 completed passes per 90 is the seventh best rate in the league, the best of any Liverpool player and the second best of any non-Manchester City player. He also makes the eighth most progressive carries in the league with 9.59 per 90.
Forward: Mohamed Salah
In homage to the free-flowing attacks of Manchester City and Liverpool, we opted to rank forward players in a single data pool rather than split them out by centralised strikers and wide players.
Mohamed Salah was of course the best rated forward player; his 0.34 expected assists is the 4th best rate in the league and his 0.55 no-penalty expected goals is the 2nd best in the division behind teammate Diogo Jota (0.61). Salah also had the most shots on target of any player in English football with 1.60 per 90.
The Egyptian was also recently voted as PFA Fans’ Player of the Year, capping off yet another spectacular season for the forward.
Forward: Son Heung-min
The shared Premier League Golden Boot winner Son Heung-min was nailed-on to make the Stats Based Team of the Year. He ranks 6th in the Premier League for non-penalty goals minus non-penalty xG with +0.19 per 90, indicating his above average finishing ability.
Of his total shots, 54.7% were on target, the 2nd best rate achieved by any forward player again revealing just how clinical the South Korean is. And there is of course the small matter of his 23 league goals.
Forward: Riyad Mahrez
Mahrez is both a creator and goal scorer for Manchester City, he has a very well rounded profile which sees him rank well across several statistics. His 2.35 completed passes into the 18-yard box per 90 is the best rate of any forward player and sees him rank 6th in the league for this statistic.
He ranks 5th in the league for goals per 90 with 0.66, and 17th for non-penalty goals with 0.42, as well as ranking 7th in the league for key passes per 90 with 2.47. He is also joint 11th in the league for expected assists with 0.22 per 90.
To calculate the form rating we began by creating a statistical profile made up of the stats that we want to judge a player on. This is done for all the typical positions in football and tailored accordingly. Each players performance data is then loaded into the relevant profile. Next, using standardisation and a weighted scoring model, players are compared by their statistical output and assigned a total Z-score. A Z-score is a way of measuring something in relation to the average.
For instance, we can rate how one centre-back compares to every other centre-back and see if their statistical performance is above or below average. The Z-scores are converted to a 1-100 scale to produce the form rating with 50 being the average. The Player of the Season was determined by finding the player with the biggest gap in performance compared to the average of players in their position (the highest scaled Z-score) – which in this case was Trent Alexander-Arnold.
Full data is available upon request.
- Performance data was collected from FBref following the end of the season.
- FBref data is subject to marginal changes and updates overtime as data accuracy is verified. As such, we can only guarantee insights are accurate based on the data set as of 31/05/2022.
- Players with less than 900 league minutes played were omitted from the study to improve the data sample.
- Insights are only true amongst players with a minimum of 900 Premier League minutes played.
- Players were rated across a range of stats designed to measure goal scoring, chance creation, possession & build up play, and duels & defence (which includes aerial ability). Each of the categories mentioned were weighted differently depending on the demands of the position in question.
- Defensive statistics were possession adjusted to improve the quality of insights.