Ultimately this is what xG is. A measure of chance quality rather than just assuming all shots are equal. When it comes to shot stopping we have to go beyond the xG because we only need to consider the shots that end up on target and need a goalkeeper to save them. High quality, high xG chances can be hit off target, can be blocked or the actual shot can be a tame one. This is where the Post-Shot xG measure comes in. Also known as the Expected Goals on Target.
This measure only considers the quality of the shot once it's known to be on target, and known to require a save. It measures the quality of the shot based on how close to the corner of the goal it is, how far out, it's speed and angle. Ultimately it tells us how often a particular shot is likely to result in a goal as a percentage. So a PSxG value of 0.50 suggests that shot will result in a goal 50% of the time. The flip side of that (which we are interested in here) is that it also suggests 50% of the time that shot should be saved by the goalkeeper.
Using similar charts we can plot the total Post Shot Expected Goals for each team horizontally and then plot the performance against those expected goals to see how well each GK performed. Again Southampton are on the far right of the chart having faced a total of 85 goals worth of shots. At the other end are Arsenal again who have only faced 35 goals worth of shots. Arsenal's goal keeper having faced half the job of the Southampton keeper and much less of a job the Matz Sels had to earn his Golden Glove award.
At the top of the chart are Bournemouth's goalkeepers who have saved 9 goals more than expected. They are the standout side this time instead of Brentford. The Brentford performance being actually below average which goes to show that while Flekken has made plenty of saves, the average difficult of those saves has been lower than other teams. This is the information that Save% alone can't give us. The other standout team is Tottenham, followed by Everton.
Forest's Matz Sels has performed above the average by 1-2 goals, but again, it's not an outstanding position on the chart per se. It is however, much more impressive than David Raya who sits below the trend line, suggesting he actually should have performed better given the quality of shots he's faced over the season. The worst performers are Wolves & Brighton's keepers, also Ipswich and Fulham. Wolves and Brighton have underperformed on both the Save% chart and the PSxG chart so that should be a cause for concern for them going into next season.
The second chart shows the same measures for individual goalkeepers as opposed to the team performance. Here you can see Verbruggen of Brighton is the worst-performing keeper on the chart. Ederson & Pickford are the best performing. Sels sits in the top 5-6 performances. Mark Travers played limited games for Bournemouth before moving to Middlesbrough, but his number suggests he's worth considering as our serious No. 2 keeper option (I say this not having watching him play / distribution of the ball).
