After months of constant speculation surrounding Chelsea’s goalkeeping crisis, it appears as though the Blues have finally decided to invest in a new keeper, Edouard Mendy from French outfit Stade Rennais FC. Below contains statistical analysis and comparison between the inbound Senegalese goalkeeper and Chelsea’s current no.1 Kepa Arrizabalaga.
The Spanish international has always seemed destined for intense scrutiny. Some of this started when Zinedine Zidane refused to sign Kepa during his first stint with Real Madrid. Chelsea then set a world record transfer fee for a ‘GK’ by spending 72 million British pounds for his services.
However, his most infamous moment, as everyone in the footballing community would recall, was his refusal to come off the pitch when ordered to do so by his then-manager, Maurizio Sarri. Since these unfortunate events, Kepa has suffered an abysmal run of form in the 2019/2020 season, extending through to the current season, including in the most recent clash against Liverpool. He has been grasping at thin air ever since in his failed attempts to re-discover the form and talent that made Chelsea outlay the excessive amount of money.
To further highlight the severe form decline Kepa has even been dropped from the starting XI multiple times for the veteran Willy Caballero, who to be fair has performed reasonably well but should not be considered as the regular starting keeper for Chelsea. When considering the difference in investment, age, and talent between Willy and Kepa, it is clear that Kepa has become a cause for concern, hence why Lampard has been planning to add additional reinforcements at GK before the transfer window slams shut.
Rumours have been swirling since before the new season that a new GK will be on his way to compete against Kepa. Names such as Andre Onana, Nick Pope, and Edouard Mendy have frequently been mentioned with potential moves to the club. More recently, it is the tall Frenchman Edouard Mendy that has been linked the most with a move to Stamford Bridge. Even Lampard himself has remained coy over a potentially new GK coming into the squad. When directly asked during the post-match interview at Brighton if a new GK may be coming, Lampard did little to quash the rumuors that Kepa has more competition coming into the Club and challenging for the No.1 jersey.
If Edouard Mendy is the new GK that arrives to compete with Kepa, a few questions will naturally be asked. How does he compare? How much of an upgrade is he over Kepa? Is Mendy another expensive, marginal gain for the squad? To compare the two goalkeepers, I have analysed multiple key statistics, including: matches played, no. of starts, total minutes, goals against, goals against per 90, shots on target against, saves, save percentage, own goals, penalty kicks against, post-shot expected goals, post-shot expected goals divided by shots on target, post-shot expected goals differential, crosses against, and, crosses stopped.
Base Statistical Comparison
Kepa has played a total of 2970 minutes in the Premier League for Chelsea, while Mendy played a total of 2160 minutes in Ligue 1 for Rennes. During these minutes, Kepa conceded 47 league goals, compared to Mendy only conceding 19 goals in Ligue 1. This equates to Kepa conceding 1.42 goals per 90 minutes, whereas Mendy only conceded 0.79 goals per 90 minutes.
Kepa faced 99 shots on target, saved 55, and finished last season with a 54.5% save percentage. Mendy showed significant improvement on Kepa, as he faced 74 shots on target and made 59 saves, for a save percentage of 78.4%. I think the key here is that Mendy faced less shots than Kepa, yet saved more shots (in less games and less minutes) than Kepa.
Mendy kept 9 clean sheets for a percentage of 37.5% whereas Kepa had 8 clean sheets, but a much lower clean sheet percentage of 24.2%. Mendy also faced 3 penalty kicks, which was one more than what Kepa faced last season.
Advanced Goalkeeper Metrics
Looking at more advanced metrics, Kepa had a post-shot expected goals (PSxG) against of 34.5, which is 13 goals less than what he conceded in the league. This means Kepa performed significantly worse than the statistical expectations of him. Meanwhile, Mendy had a PSxG against of 19.7 and only conceded 19 goals. This means Mendy slightly outperformed his expected numbers, unlike Kepa.
The next key metric is PSxG per Shot on Target (PSxG/SoT), which accounts for how difficult of a shot on target is, and then re-calculates the PSxG based upon the average difficulty of shots on target. Kepa had an PSxG/SoT of 0.35, and Mendy had a PSxG/SoT of 0.22. I interpret this to mean that Kepa faced more difficult shots, and that his PSxG is slightly inflated (compared to Mendy’s) because of the increased difficulty of the shots on target that Kepa faced. However, this theory can be tested by analysing the PSxG Differential, which calculates the expected number of goals based on how likely the keeper is to save the shot. So, a negative value means a GK saved less than expected shots on target, and vice versa. Kepa had a PSxG differential of -9.6, meaning he conceded nearly 10 more goals than would have been expected, even when factoring in the difficulty of the shot on target. However, Mendy performed much better, with a PSxG differential of 1.7, meaning he saved nearly 2 more goals than would have been expected relative to the difficulty of the shots he faced.
One key area of common fan criticisms of Kepa is his lack of ability to command his box and stop crosses. Therefore it would be expected that any new GK for Chelsea would be an improvement in these areas, so I analysed the number of crosses faced, crosses stopped, and a cross stop percentage between Kepa and Mendy.
Kepa faced 236 crosses into his box in the 2019/2020 league season, and he stopped 17 of those crosses for a cross-stop percentage of 7.2%. Meanwhile, Mendy faced 197 crosses into his box in Ligue 1 last season, and he stopped 20 of those crosses for a cross-stop percentage of 10.2%. Mendy is clearly an upgrade in this regard, as his percentage is greater than Kepa’s, yet he also faced less crosses and claimed more crosses than Kepa. There is reason to believe that Mendy will be an upgrade over Kepa when facing opposing crosses, but the improvement may not be overtly drastic.
The final metric I analysed is how each performs in the sweeper-keeper role, and I looked at the number of defensive actions outside of the penalty box. Kepa had 21 defensive actions outside of his own box, as compared to Mendy having only 16 such defensive actions. Luckily, this may be one of the only areas that Mendy does not represent an upgrade.
Even if Edouard Mendy does not upgrade the Chelsea squad with sweeper-keeper actions, as demonstrated above, he is still a considerable upgrade over Blues current GK Kepa based on almost every other statistic analysed here: goals against, GAp90, PSxG, PSxG Differential, crosses stopped, save percentage, clean sheet percentage, etc.
Considering these data points, Mendy is an objective upgrade over Kepa based on their respective performances throughout the 2019/2020 league season. Here’s hoping Chelsea’s new GK is announced soon and hits the ground running at Cobham sooner rather than later.
Written By: Travis Flock (@Crossroads_CFC)