This Week in Stats: Is De Gea the Premier League’s best goalkeeper?

David de Gea is considered to be the Premier League’s best goalkeeper, but do the stats back that up? Is the Manchester United man really the top dog?

Goalkeepers get a rough deal, as they’re paid less than players in other positions. Yet it’s often said teams can’t win big prizes without an excellent shot stopper. Manchester United may not be winning top honors at present, but in David de Gea the Red Devils possess one of the best goalkeepers in the world.

He has won United’s player of the year award four times in the last five years, so he must be doing something right. That fact suggests their defense is not doing much right, but that’s for another day. But is de Gea the best in the Premier League? And can numbers help us determine who is?

Goalkeepers aren’t well served by statistics. The most widely available data tends to focus on attacking players; goals, assists, shots, chances created and so on.

The Premier League present a Golden Glove award to the goalkeeper who keeps the most clean sheets. De Gea won it in 2017-18, perhaps surprisingly, for the first time. But a clean sheet is a team award, as a keeper may not have a save to make thanks to the efforts of the players in front of him.

You can look at save percentage, and de Gea topped that chart too. However, it assumes all shots are equal, when we know that is not so. As is now commonly the case, expected goals can help us here.

By looking at the total on target xG which a goalkeeper faced, we can get a more accurate picture than save percentage alone can provide. Understat has the data, and here is what it shows us for the 2017-18 Premier League.

Bear in mind penalties and own goals are excluded, as the former wildly distort the figures due to their high xG values, and it seems harsh to punish goalkeepers for the latter. The figures here are listed by club, so let’s drill down into some player stats.

De Gea is not the master

That’s not a typo in the headline. There was a goalkeeper who saved a higher proportion of the expected goals he faced than de Gea managed to. And that man is Burnley’s Nick Pope.

Jose Mourinho rested de Gea in the final league match so he could be fresh for the FA Cup final. Sergio Romero went on to make three saves, including one clear-cut chance, against Watford as United won 1-0. Subtracting those figures from United’s total gives de Gea an expected goal save percentage of 56.5.

Nick Pope began 2017-18 on the bench, and came on in Burnley’s fourth match when Tom Heaton sustained an injury. England’s newest international goalkeeper saved 58.2 percent of the xG he faced, putting him top of the Premier League tree.

Pope wasn’t the best in Europe, as Atletico Madrid’s Jan Oblak prevented a remarkable 60.8 percent of the on target expected goals he faced from finding the net. But considering he’d never played in the top division before, the Burnley stopper performed superbly. Pope made the plane for Russia, but he arguably should be England’s first choice.

Poor performers and differentials

Had it not been for Heaton’s dislocated shoulder, Pope may not have played at all. However, some teams used more than one goalkeeper thanks to poor performance from their initial number one choice.

It can’t be coincidence that the bottom four teams in the above table split their goalkeeping duties this season. Of those sides, Watford’s Heurelho Gomes saw the most action as he made 24 appearances, but nobody made more than 20 for the other three clubs. Adrian and Joe Hart shared the gloves for West Ham, Loris Karius and Simon Mignolet did likewise for Liverpool, and Fraser Forster was bombed out by Southampton in December, to be replaced by Alex McCarthy.

It’s also at the bottom of the table where there are the biggest discrepancies between regular save percentage and it’s xG cousin. Huddersfield and Leicester were ranked 18th and 19th respectively for their save percentage, but 12th and 14th when the fancy stats are brought into play.

Southampton and West Ham saw the opposite occur, as they were mid-table for save percentage, but bottom on expected goals. How much that tells us is debatable, but it illustrates how important chance quality is when trying to assess goalkeepers.

Next: The best player on every team at the World Cup

We need to talk about Karius

After a rocky debut campaign at Anfield, Loris Karius appeared to have made the Liverpool goalkeeper’s slot his own in 2017-18. His errors in the Champions league final undid most of his good work, though. Even if Karius is mentally tough enough to recover, the majority of Kopites will not be forgiving.

The Reds have long been linked with Alisson of Roma, and may have purchased him this summer irrespective of what happened to Karius in Kiev. The Brazilian’s xG save percentage of 47.3 in Serie A would put him fifth in the Premier League table.

As Liverpool only prevented 36.5 percent of the expected goals they faced from going in, it looks a no brainer. But Jurgen Klopp used two goalkeepers as we know, and their performances were wildly different.

Simon Mignolet had a dreadful season, saving just 27.6 of the on target xG he came up against. It was down to Karius to drag Liverpool’s figure off the bottom of the table, and he did so handsomely.

The German’s xG save percentage was 54.4 percent, which would put him third in the Premier League table. His sample was small, but if we include last season Karius still registers a very decent 52.1 percent.

Of course, there is far more involved in assessing a goalkeeper than the stats presented here. Karius may be a good shot stopper but the errors he makes spectacularly undermine that. There’s a very capable keeper in there somewhere, but whether he makes the grade at Anfield remains to be seen.