Last time out we looked at what we think our goalkeeping model actually measures and the ability of keepers to continually make saves from Zones 1 and 2. We found that the range in ability in Zone 1 was pretty narrow, but as we got further out centrally into Zone 2, the apparent ‘talent gap’ grew much wider.
As a reminder. Here’s how the Zones have been split:
Only De Gea, Vorm, Reina, Hart, and Mignolet were above the Premier League average in both Zones 1 and 2. We can now move on to the wide areas in Zone 3. Here’s the plot:
The thick black line through the middle represents the Premier League average. Again, there is no ‘statistically’ significant keeper outside the curved lines. However, as with Zone 2 the spread is quite wide – the difference between the ‘best’ and the ‘worst’ keepers here is around 14%. We have previously looked at over-aggressive positioning when facing shots in Zone 3 using Lloris as an example. Lloris hasn’t been included in these plots due to lack of games/sample size, but we can confirm he’s below the thick black line here.
De Gea, Hart and Mignolet are the only keepers to be above average in all 3 Zones we’ve looked at so far. Can anyone keep it up when facing shots from Zone 4? Let’s take a look:
There’s less than a 10% spread between all the keepers in the sample here, meaning there’s an even smaller gap in apparent ability than we saw in Zone 1. The difference here of course is that the average save % is much higher as you’d expect due to the greater distance from goal. It was from shots from distance that we first introduced the idea of the Keepers’ Comfort Zone.
If we compare the numbers in that piece (which used a much smaller sample size) to the ones in the last graphic, we see a fair amount of movement over time. This is in keeping (ahem) with what James Grayson was saying in his last look at keepers using our numbers. Goalkeepers seem erratic. This is down to a fair slice of luck as James suggests, but we also think a chunk of it is down to poor decision making by keepers. All of them are prone to it, it’s just that some are prone to it more than others.
As we’ve previously stated, we think this is due to them being drawn from the Comfort Zone – the back half of the 6 yard box. A lot of keepers aren’t even drawn there – they seem to prefer to position themselves aggressively quite naturally. Some of the feedback we’ve been getting seems to assume that we’re advocating that goalkeepers should never move from their line. We’re not saying this at all. Circumstances dictate that a keeper needs to come out to narrow angles at times. What we are saying is that the decision making could be much better in regards to when to come and how far.
So after looking at Zones 1 to 4, we’re left with just two keepers – De Gea and Hart – who are above average in each Zone. If we chuck all the data into one pot we can rank all the keepers over the last 3.5 seasons to see who comes out on top in terms of shot-stopping based on shot location:
We can see how many goals above the average each have been stopped in this time and also how many have been stopped per 100 saves. This helps to normalise everything as the varying degrees of work they’ve had to do gives some keepers more opportunity than others to bump their numbers up.
Overall, this table seems intuitively right looking at the names in green and those in red. Cech is perhaps surprisingly low, but as we’ve explained before, he’d be a lot higher if Andre Villas-Boas’s defensive line had been a lot lower at Chelsea. In any case, Cech no longer looks as solid as he did when Mourinho first came to Stamford Bridge. Unfortunately the data isn’t available publicly to test whether this was actually the case.
De Gea is king of the keepers who love to stay home in the Comfort Zone. Hart and Mignolet, have been fans of staying home a lot too. However, Hart has made some high-profile howlers having had some shocking rushes of blood in recent times. Mignolet too is regularly being drawn out now he’s moved to Liverpool. Mignolet isn’t being caught for doing so yet, although Soldado’s disallowed goal last weekend was a warning sign against future bad decisions. Mignolet has, however, been getting caught from distance more as his starting position has become more aggressive. Begovic is doing the same. All three keepers have lost ground on De Gea in our table.
Moving forward we hope to develop the model to see if there’s a pay-off here. Does a more aggressive keeper force more shots off target? Can we find optimal keeper position depending on where a shot is struck from?
Follow us here on Twitter.