Everton have conceded 13 goals from 20 shots on target. That really takes some doing. Here’s how it looks visually (I resisted the urge to scribble wildly on a bit of paper). Yellow dots are shots that ended up in the net, blue dots are shots that Howard saved:
Using my expected goal model I can assign each shot a probability of it actually going in. Put ’em all together in a set, simulate the set thousands of times over and here’s how the distribution of goals looks for these shots:
As you can see from the graph there’s just over a 1% chance of Everton concededing exactly 13 goals from the shots they’ve faced. There’s less than 2% chance Everton could have been worse. Keepers on even the worst sides save shots at twice the rate that Howard currently is.
Everton’s upcoming opponents will have to earn their goals by having a lot more shots on target than the previous ones have – no matter which defenders play in front of Howard.