Has Tyler Roberts Turned A Corner In Midfield?
Jacob Standbridge casts a look over Tyler Roberts’ recent performances in midfield for Leeds…
Tyler Roberts will have been proud to win man-of-the-match in Leeds’ recent 0–0 draw with Chelsea. Written off by many as a player only or not even fit for the Championship, performances like this show he is capable of performing at Premier League level.
Roberts’ 2020/21 season has been stop-start. Unable to register significant game time in his preferred position up front, he’s largely played as the most advanced of Leeds’ midfielders. Supplanted by new signing Rodrigo after a few early starts, he played less than a game’s worth of minutes between early October and mid-January. For a month over Christmas, due to form and illness, he failed to appear at all.
In recent months things have been different. A strong substitute showing vs Brighton prompted a run in the team, making ten appearances in the following twelve games, including 5+ consecutive league starts for only the fourth time in his Leeds career.
Precisely how much impact Roberts has had can be debated, but to the eye he seems to be more involved in games in 2021, and he’s registered two assists, along with two goals and two further assists disallowed for offside.
In a recent press conference, Marcelo Bielsa said Roberts needed to demonstrate that he should play for Leeds “not because anyone is missing but because he deserves to keep his place.” With Leeds reportedly in the market for five new signings and his contract up in 2022, Roberts’ position in the squad could come under threat.
Key to succeeding in a Bielsa squad, especially if you cover a player unlikely to be dropped like Patrick Bamford, is multi-functionality. Looking at Roberts’ midfield performances in 2021, we can try and assess if he merits a place in the 2021/22 squad.
What has changed?
Using StatsBomb statistics via FBref, we can see how Roberts performed in midfield across a variety of metrics this season. To assess this, I split his season into two tranches (2020 & 2021), removed his appearances up front,* and calculated per 90-minute rates (p90) for all metrics.
*The datasets include the time Roberts spent upfront in Leeds’ two games against Fulham and the 0–0 vs. Chelsea. This is due to FBref’s limitations. These were included to give as large a sample size as possible, but do distort the data slightly.
This table shows how Roberts’ performances have changed in some passing and creative metrics:
Although the numbers are not significantly higher, they do back up the idea that Roberts has got more involved in Leeds’ play.
Though he’s mostly completing more short passes, Roberts’ passing now moves the ball upfield more effectively, completing an extra 2.08 passes into the final third p90. His involvement has also increased closer to goal, with more Key Passes (passes that lead directly to a shot) and Shot-Creating Actions (the two offensive actions leading directly to a shot), and increasing his xG p90 from 0.04 to 0.15.
This table features Roberts’ performance across key possession metrics:
Standing out is Roberts’ significantly increased progressive carry distance. In 2020 this was even under goalkeeper Ilan Meslier’s 48.5 yards p90. Roberts is now passing and carrying the ball forward much more.
Where he gets onto the ball impacts how much he gets involved in Leeds’ play. As shown, Roberts touches the ball more often, and he now registers 8.5 more touches p90 in the middle 3rd of the pitch. From this and the larger number of received passes, you would expect to see Roberts moving into more effective deeper and/or wider positions to be more active in the construction of Leeds’ attacks (as Leeds build-up down the flanks).
I should caveat that increased volume does not imply increased quality. Indeed, despite putting up more game involvements, Roberts’ xA p90 is down in 2021, which could indicate issues with his final third decision making. This said, to have increased his own performance mid-season is to be commended.
This article won’t look at Roberts’ defensive numbers, which remain largely similar throughout the season. Having been at Leeds for Bielsa’s entire tenure, he is very comfortable with the man-marking system.
How does this look in-game?
From looking at Roberts’ numbers we should expect to see him dropping deeper and/or wider more often in 2021. To compare, I reviewed Roberts’ performances against Manchester City and Southampton.
These games are similar: in both the opponents utilised a high press, whilst Roberts played as Leeds’ advanced midfielder, was substituted in the second half and received a similar percentage of the passes targeted at him — 71% vs City and 68% vs Southampton.
However, there’s a clear difference in how much Roberts was involved in the games, demonstrated by how often Roberts was the target of passes. A rate of 27.8 p90 vs City is significantly bettered by his 41.4 p90 vs Southampton.
A major contributing factor to this will be the positions Roberts occupies on the pitch. Against City, only 3 of the 12 passes he received came to him in wide zones of the pitch, with 4 in zone 14 (the central space outside the opposition penalty area) and the remaining 5 in deeper central zones. Against Southampton, the picture’s very different, with Roberts receiving over half of his passes (13 of 23) in wide areas, 7 in deeper central zones, 3 in the box, and none in zone 14.
This is not to say that Roberts should not look to receive the ball in zone 14 — to the contrary, it’s a very valuable area to get possession in. However, given that Leeds build up in wide areas and vacate the centre, you wouldn’t expect as large a proportion of his touches to be in that zone as they were against City.
As mentioned, Roberts carries the ball more often in 2021, and this ability has proven a useful tool in Leeds’ armoury. Here against Southampton, he progresses the ball from the halfway line to almost level with the box, holds his own against two markers, and draws a foul in a dangerous area:
This season Leeds have struggled against teams that press high to disrupt their build-up. In 2021, Roberts’ carrying has helped relieve that pressure, and by moving the ball up the pitch he helps join up Leeds’ attack and defence. Here, he holds onto the ball to slow down play and draw in his marker, before turning and driving into the space in behind:
Roberts also uses this ability to draw fouls. Against Southampton, he drew two from Jannik Vestergaard. Here, Vestergaard is forced to foul Roberts, who could otherwise drive into the space he vacated or look to play in Raphinha or Bamford:
Fouls won can become dangerous opportunities. In the free kick Roberts wins above, Leeds narrowly missed from a well-worked routine.
Roberts, Klich and Rodrigo
This season Leeds have predominantly used three players as the advanced midfielder — Roberts, Rodrigo and Mateusz Klich. Calculating the same p90 metrics as before, it’s instructive to compare Klich and Rodrigo there this season to Roberts’ 2021 numbers. First, we’ll look at their passing and creativity:
Whilst Roberts’ numbers lag behind Rodrigo’s, Klich leads the way for volume. However, Rodrigo is much closer to Klich for Key Passes and Shot-Creating Actions (and would lead in the latter if dead balls weren’t included) which reflects the two players’ differences. Klich creates more through volume, whilst Rodrigo looks to create fewer chances of higher quality.
Given the profile of the players and their numbers, Roberts appears more of a ‘Rodrigo’ than a ‘Klich’, and it is telling that of all current Premier League midfielders, the player FBref rates as most similar to Roberts is Rodrigo.
Next, we’ll look at them in possession:
Mostly, the story is the same: Klich puts up the highest volume, though Roberts and Rodrigo both dribble more.
Whilst Klich carries the ball forward the most each game, Roberts’ individual carries progress the ball the most, at a rate of 3.22 yards per carry versus Klich’s 2.62 and Rodrigo’s 2.35. If Roberts can maintain this rate but scale up his volume of carries, his progressive carry distance p90 would quickly exceed Klich’s.
As touched on before, carrying is an area of strength for Roberts. This table includes additional p90 metrics that demonstrate his ability on the ball, including when put under pressure. Here, his performance is similar to or better than that of his colleagues:
Overall, Roberts has not yet demonstrated the output quality to make up for his comparatively lower volume of involvement. However, Klich and Rodrigo have been two of the most consistently productive creators in the league this season, so we shouldn’t conclude that he is performing poorly. He’s just not creating at an elite level yet.
Whilst he does need to improve both his volume and quality to challenge for a starting berth, Roberts’ numbers and familiarity with the system make him a perfectly serviceable backup option for Leeds’ advanced midfield position. His improvement across these metrics this season is a good start and at 22, he has more time to kick on further.
What does he do well? What can he do better?
As mentioned before, increased quantity doesn’t imply increased quality. Indeed, Roberts’ Expected Assists p90 are down in 2021. On average, he’s creating more but weaker chances than before.
Watching back his 2021 match involvements, several positive qualities leap out. His footwork is good and allows him to take on opponents in tight spaces, complementing his ball carrying. He demonstrates ability to make a variety of different passes, including a useful clipped ball down the touchline to bypass pressure from in front.
Following an aerial pass, Roberts receives a short pass from Costa and plays a quick line-breaking pass to Bamford, leading to a one-on-one:
Roberts wins and keeps the ball, and after looking forward plays a long ball to Bamford:
Roberts perfectly weights a through ball to Bamford to score:
More broadly, Roberts generally performs well when playing instinctively, a skill which would be hard to otherwise develop. When Leeds are in attacking transition or if he receives the ball under pressure, he usually can progress the ball upfield or recycle it to a teammate.
With more time on the ball, his decision-making becomes less consistent. Sometimes he holds onto it too long and the attacking momentum dissipates. Here, having already turned on the spot with the ball, Roberts takes too long, fails to find one of the open options, and plays a poor cross:
Along with a propensity to hit crossfield balls poorly (though as shown for the own goal in the 2–1 victory over Sheffield United, he can execute them well), Roberts’ final third decision-making is inconsistent.
Here, Roberts releases Bamford into the box, but Bamford is dragged quite wide onto his weaker foot and can only shoot into the defender. Instead, Roberts could have found Klich to release the unmarked Jack Harrison:
Roberts often takes speculative, low-percentage shots when other options are on. Here, instead of playing in Raphinha, he clears the bar with a shot:
Opportunities like this could considerably improve Roberts’ xA if he made better choices. Bar the two goals disallowed for offside, Roberts’ shots this season have come from low-xG opportunities, and he often shoots with little composure or into opposing players. This is very much something to work on.
Conclusion
Roberts has shown himself to be technically capable but even in this better run of games there have been games where very little he tried came off. Any increase in the volume of his play needs to be accompanied by a more consistent level of decision-making and quality in the final third. This is also something Roberts is aware of, recently citing himself as his own harshest critic and saying that consistency is the “main thing” for him right now.
That said, he already does some things very well and it is encouraging to see him progress during this season. It should be enough for him to earn a new contract and continue as part of Bielsa’s group of 18 ‘starters’, as both backup to Bamford and midfield cover.
At 22, time is very much on his side and if Roberts can continue on his upward trajectory rather than plateau, he can become a more important player for Leeds than he already is.
You can follow Jacob Standbridge on Twitter @PocketTrumpeter.
If you enjoy this content and want to help us create more and better pieces, why not consider signing up to our Patreon page to get bonus material?
If you have any interest in contributing to this blog, get in touch with us @AllStatsArentWe.
