Mason Mount: A Statistical Analysis of Who Will Fit in Thomas Tuchel’s Chelsea System

Thomas Tuchel’s arrival to Chelsea Football Club has created a myriad of emotions within the fanbase. Tuchel is presented with a difficult challenge to improve results, replace the biggest Chelsea legend in former central midfield player Frank Lampard, and to win over the fanbase in such a fragile time. When he benched Mason Mount in his first game against Wolves, questions were asked.

Tuchel is regarded as a tactical chameleon and mastermind, and this has already been evidenced by the success of re-integrating the three at the back formation in what has seemed to be a 3-4-2-1 formation so far. A striking difference with Tuchel has been how he describes and sets up his midfield. Tuchel has described that he wants to play with “double 6’s” in the midfield who act as holding midfielders, yet the Germans requires these dual 6’s to quickly move the ball around the pitch and to retain possession. The defensive requirements of the CMs are lessened if the squad is chronically in possession of the ball and forcing the opposition to chase the game.

With that said, the profile of a player that Tuchel is looking for in the midfield is likely one that has a high pass completion percentage, can beat a press, applies pressures and interceptions when off the ball, and plays more progressive passes to open up low block sides that are pinned back by Tuchel’s tactics. To assess and assert who may be the best fit to play in the midfield at Chelsea, I have analyzed each individual match data per each center midfielder in the squad, except Billy Gilmour, simply because he lacked enough minutes to justify an analysis that would lead to reliable conclusions.

This analysis will analyze each CM individually and report the results, and then have a comparative synopsis after presentation of all individual results. Individual players have been analyzed for non-penalty goals, non-penalty expected goals, assists, expected assists, minutes played, shots on target, pass completion percentage, touches, pressures, tackles, interceptions, blocks, completed dribbles, shot creating actions (SCA), goal creating actions (GCA), progressive passes, and progressive carries. Data analyzed was from, and only data with Lampard as manager was used. EFL Cup matches were excluded from analysis. UCL matches were included for analysis.

Mason Mount: Individual results and analysis

Mason Mount

Basis Descriptive Statistics

Mason Mount played a total of 1792 minutes and received 4 yellow cards in the Chelsea midfield. During this time on the field, he registered 2 goals and 4 assists. Mount is generally praised for his work rate and off the ball hustling, and this notion is supported by the numbers. Mount had 416 pressures, 51 tackles, 24 interceptions, and 30 blocks in the 2020/2021 season under Lampard’s management. Overall, Mount was heavily involved in the Chelsea midfield in terms of pressing and defensive work. Mount has also exceeded his expected assists (3.8) by registering 4 assists, but he is underperforming his xG, as he has 2 goals yet was expected to have 2.6.

Mason Mount completed 82.93% of his 994 attempted passes under Lampard, and 106 of those passes were progressive passes, which are defined as a pass that travels 10+ yards towards the opponent’s goal. The midfielder recorded 38 shots under Lampard this season with 12 of the shots being on target, for a shot target percentage of 31.5%. His conversion rate for shots is 5.2%. Mount carried the ball 970 times and 128 of those were progressive carries towards the opponent’s goal. Finally, Mount completed 27 dribbles prior to Lampard’s dismissal.

Plots and Correlation Analyses

As previously mentioned above, Mount has underperformed his expected goals, and the same trend is true for non-penalty goals to expected non-penalty goals. As evidenced by the below plot, Mount has exceeded his nPKxGs twice under Lampard across all the minutes he played and underperformed the expected goals in all other appearances in which the xG value was not 0.0.

The stats behind Mason Mount and his non-penalty expected goals
A graph showing Mount’s non-penalty expected goals and his non-penalty goals

Mason Mount played 106 progressive passes and produced and 82 shot creating actions (SCA). However, is there a correlation between these two metrics? Specifically, does Mount create more shots for teammates the more he plays progressive passes? The answer is below in the scatterplot.

This season's progressive passing statistics
A graph to show the statistics behind Mason Mount’s progressive passing and shot creating actions

I conducted a correlation analysis for the above data in the scatterplot, and found that there is a statistically significant, medium-strength, positive correlation between Mount’s SCAs and Progressive Passes (t = 2.15, df = 21, p = 0.04, r = 0.425). This means that the more Mount plays progressive passes the more likely he is to create shots for the team, and this cannot be attributed to random chance, as the percentage of the correlated data that is explained by random chance is only 4.25%.

There is more than one way to be progressive on the ball, and that is evidenced by the number of progressive carries a player has. To determine how often Mount is progressive in his carries, I created and analyzed the plot below.

The midfielder's progressive carries and touches
Mason Mount and his progressive carries and touches

It is quite apparent from a visual scan that the above plot has a positive correlation, but the key point is whether the positive correlation is statistically significant, otherwise random chance better explains the results if the correlation is non-significant. The above plot has a highly statistically significant, medium-strength, positive correlation (t = 3.05, df = 21, p = 0.006, r = 0.555). This means that the chance of the correlation being caused by random chance is six-one-thousandths of a percentage. Essentially, there is very little in the above plot that is due to random chance. Mount is creating progressive movements and carries the more he touches the ball.

Although Mount created more shot creating actions with his progressive passing, does the same trend bare truth for goal creating actions (GCA)? Below is the corresponding plot and analysis. For those wondering, GCAs are most akin to a “hockey assist” or secondary assist. The caveat is that a player can register more than one GCA per each goal: this would mean the player could have dribbled past an opponent, passed to a teammate, and that teammate plays an assist to a different teammate.

Mason Mount's progressive passes and goal creating actions
A graph detailing Mason Mount’s progressive passes and goal creating actions

The correlation here is difficult to discern due to the nature of the data of GCAs having a relatively lower and less frequent values than progressive passes. However, it can be noted that Mount plays many progressive passes that do result in GCAs downstream in the subsequent attacking play/phase. To analyze the assertion I just made in the previous sentence, I ran another correlation analysis. Progressive Passes and GCAs by Mount resulted in a slight negative correlation, but it was not statistically significant (t = -1.08, df = 21, p = 0.291, r = -0.230). This means that 23% of the correlation is explained by random chance, and therefore it is best to posit that progressive passes are not a good predictor variable for GCAs for Mason Mount.

Finally, a simple boxplot can go a long way in data visualization and analysis. I created a boxplot of Mount’s pass percentage.

Mason Mount and his pass percentages
A graph showing Mason Mount and his Mason Mount and his pass percentages

The above plot shows several key pieces of data: the median, 25th and 75th interquartile ranges (IQRs), the lower and upper bounds of the probability distribution function (PDF), and outliers. The IQRs are displayed by the upper and lower (75th and 25th, respectively) bounds of the grey box, and these values correspond to one standard deviation above and below the median value on the PDF, respectively (this is but one way to display and evaluate the variance around a central tendency, which in this case is the median). The “whiskers” of the plot are the upper and lower bounds of the PDF, meaning that values outside of the “whiskers” are outliers, and those are displayed by the unshaded circles.

Overall, Mount has a median pass percentage of 84.1%, and the deviation around the median is approximately 3.5%, meaning that when Mount performs one deviation above the median, his pass percentage is predicted to be around 87.5%, and when he performs one deviation below the median, the pass percentage is around 80.5%.

Mason Mount playing in the Chelsea midfield against Sheffield United
Mason Mount fighting for the ball against Sheffield United


I think Mount’s off the ball work will be greatly appreciated in the Chelsea midfield, and that Mount will be integral to the pressing system that Tuchel will undoubtedly continue to refine with the squad.

Mount will likely want to upgrade his progressive passing in terms of creating goal and shot actions if he is played further up the pitch in Tuchel’s system, as opposed to play in the “dual 6’s” role.” Mount will likely also need to increase his non-set piece assists and focus on becoming more of an advanced playmaker in attack, and a tireless pressing center-mid in transitions and counter presses.

Additionally, he will likely have to also find a way to continue his passing accuracy while attempting more progressive pass attempts that lead to GCAs under Tuchel’s system.

My verdict is that Mount should thrive with Tuchel’s tutelage and tactics. He should be a staple in Tuchel’s Chelsea starting XI most weeks in the midfield. He seems like a player that is primed to succeed in this system.

Written by Travis Flock @crossroads_cfc

Edited by Tom Coley @tomcoley49

Follow us on:

Leave a Reply