Fresh questions surrounding the Leeds United goalkeeper position have been put to bed for now.
It feels like Leeds United are never too far away from goalkeeper uncertainty but for now, there is a sense of stability.
Leeds were 1-0 down and still in the game when Noni Madueke whipped a corner right in towards Karl Darlow on 38 minutes, his penalty area resembling a wrestling ring on Royal Rumble night. Even still the Welshman should have claimed, instead he fumbled the ball into his own net and that felt like game over.
"Yes, overall, of course he is disappointed with the situation,” manager Daniel Farke admitted of Darlow after the error. In reality, it had little bearing on the end result - Arsenal would likely have won anyway - but what it did was bring the goalkeeper debate back to the fore.
Darlow replaced Lucas Perri last month after the Brazilian was openly criticised by Farke for his role in goals conceded against Manchester United and Newcastle. The Leeds boss was swift in his decision to change things and so it was natural for some to think he could be equally decisive following the Arsenal mistake.
But Farke has stuck with Darlow and while performances since haven’t been spectacular, they have been steady enough to bring stability. Leeds don’t currently need a world-class goalkeeper, they simply need someone who can largely stop the shots they’re expected to and keep things calm with the ball at their feet.
Measuring Karl Darlow’s Leeds United performances
The most effective way to measure a goalkeeper’s shot-stopping ability is by using a metric called Expected Goals on Target (xGOT) - also known as post-shot xG. While xG quantifies the quality of a chance, xGOT measures the quality of a shot after it has been taken, offering a way to evaluate how likely a goalkeeper would be expected to save it.
Simply put, if the xGOT number is lower than the number of goals conceded in a game, the goalkeeper in question has let in efforts they should have saved. If the xGOT number is higher, it means the goalkeeper has performed well and made some unexpected saves. Like with xG, xGOT is measured between 0.01 and 0.99, with the higher value equalling an almost certain goal.
For example, it can be seen that Darlow performed below-par during last month’s defeat at home to Arsenal, conceding four goals from eight shots which had a total xGOT value of 1.51 according to FotMob. In isolation, Madueke’s corner, once taken, had an xGOT value of 0.11.
Leeds needed Darlow to bounce back quickly, with a massive game at home to Nottingham Forest up next. Those ahead of the Welshman did much of the work with a dominant performance but from 1.48 xGOT, he conceded just one goal through Lorenzo Lucca’s close-range header.
Against Chelsea, Darlow conceded twice from an xGOT of 1.46, suggesting a slight under-performance - it’s important to note xGOT only records shots on target, meaning Cole Palmer’s late miss was not included. But his best performance came at Aston Villa, conceding just once from four shots with an xGOT of 1.84.
Improvement clear albeit unspectacular
That such a number came from just four shots suggests Darlow had to face a higher concentration of good strikes, and there was little he could do about Tammy Abraham’s leveller which recorded an xGOT of 0.96. His first-half save of Amadou Onana’s effort was a highlight and the speed of mind to scoop the ball away from Ollie Watkins was equally impressive.
Read More
In that three-game spell, Darlow has essentially conceded four goals from an xGOT of 4.78, which simply means he has prevented close to one goal. And while that might not seem spectacular, it’s important to note only seven Premier League goalkeepers have prevented more than one goal all season.
At the same time, Darlow’s superiority over Perri with the ball at his feet has rarely been in doubt. The Welshman is actually in the Premier League’s top 10 per cent of goalkeepers for accurate long balls per 90 minutes (7.73) while also out-performing his main competitor for successful passes (21.55 vs 18.69), successful pass percentage (55.9% vs 53.6%) and accurate long ball percentage (31.7% vs 26.5%) - all per 90 minutes.
It’s clear Darlow is not a world-class goalkeeper and nor is he the long-term answer to Leeds’ goalkeeping question, but after the mistake at Arsenal it was imperative he steadied the ship quickly. For now at least, there is stability.
Continue Reading