As the final minutes of Chelsea's trip to Liverpool slipped away at the back-end of last season, Andy Robertson's set piece delivery was greeted by a chorus of echoing voices screaming: 'Kepa!'
Dead ball situations were a prominent Achilles' heel for Frank Lampard's side last season and the anxiety among Chelsea's defenders as the ball sailed over their heads was palpable as they desperately urged their goalkeeper to come and collect the lofted pass.
Chelsea's 5-3 loss to the Premier League champions that night would prove to be Kepa Arrizabalaga's last appearance of the season as Lampard favoured 38-year-old veteran Willy Caballero for the final few fixtures.
Kepa was restored to the lineup for Chelsea's first competitive outing of 2020/21 against Brighton and proceeded to concede a long range effort which only asked more questions of his shot-stopping ability. Lampard's words after the match may have offered encouragement for Kepa, but the club are on the cusp of securing his potential replacement in the form of Édouard Mendy - something all but confirmed after the Spaniard produced another howler in Sunday's loss to the Reds.
The 28-year-old Rennes goalkeeper has now arrived at Stamford Bridge with a vastly superior reputation compared to the man he is replacing (although he could hardly have a worse standing than the much-maligned Kepa). But do his statistics back up this supposed superiority?
(Stats are for league games in the 2019/20 season compiled via FBref)
1. Save Percentage
A goalkeeper's job - primarily - is to prevent the ball from going in the back of their net. Revolutionary stuff. However, Kepa has horribly struggled with this basic concept.
Of the 99 shots Kepa faced in the Premier League last season, 47 of them flashed into the net, giving him a woeful save rate of 55% - the worst ratio among every goalkeeper in Europe's top five leagues with at least ten starts.
Mendy, on the other hand, saved more than three-quarters of the efforts he faced (for the second season in a row), the sixth-best record among the same contingent of goalkeepers.
2. Post-Shot Expected Goals
Kepa: -9.5 (-0.29 per 90)
Mendy: +1.7 (+0.07 per 90)
Save percentage can be misleading because it doesn't take into consideration the quality of the efforts goalkeepers are faced with. However, post-shot expected goals (PSxG) attempts to do just that.
To calculate PSxG, every shot on target is assigned a number between zero and one representing the probability of that effort beating a goalkeeper based on the strike's trajectory, speed and other characteristics. The sum of these values gives an estimate of the number of goals the model predicts a keeper facing these efforts to concede.
If the difference between a keeper's PSxG and their total goals conceded (the figures shown above) is positive, it is an estimate of the extra number of goals they prevented compared to the average keeper. If the figure is negative, it highlights how more goals they conceded than the model would have predicted.
So - deep breath - Kepa conceded between nine and ten goals more than he 'should' have last season. Only two goalkeepers in Europe's top five leagues differed by a larger, negative margin. The Spain international also underperformed in 2018/19 but to a less extreme extent.
Mendy actually saved between one and two more goals than expected. This is above average without being particularly elite.
3. Percentage of Crosses Stopped
Mendy may be three years older than Kepa, but he's at least four inches taller. The 6'6 Senegal international used this aerial advantage to successfully stop 10.2% of the crosses which came into Rennes' penalty box last season. This is an area of his game Mendy improved upon after moving from Reims to Rennes in the summer of 2019.
Despite standing at between 6'1 and 6'2, Kepa's 7.6% of stopped crosses last season was roughly average for Europe's keepers, but still considerably inferior to Mendy's return.
4. Actions Outside the Area
Kepa: 21 (0.64 per 90)
Mendy: 16 (0.67 per 90)
To play the type of high pressing game Lampard appears to aim for, Chelsea's goalkeeper - whoever that may be - must be prepared to snuff out an opposition counter attack be venturing beyond his 18-yard box.
Yet, neither Kepa nor Mendy are particularly active when it comes to defensive actions outside their area. (For comparison, Manuel Neuer - the golden standard - completed 1.64 per 90 last season.)
Mendy also tends to linger closer to his goal than Kepa. However, this difference could merely be a representation of the differing approaches of Rennes and Chelsea last season.