Quote:
Originally Posted by Bingo
Vladar didn't have a great night. He gave up five goals, one very suspect, one somewhat suspect.
Are you debating that?
The average NHL team gives up 12.5 high danger chances a game, the Flames 11 last night.
|
No, that wasn’t my point at all
Let me put it this way. I do not believe that any individual event has an xGA of over ~.33.
The probability of a shot being a goal has as much to do with defensive positioning, and where shooters place shots, which are not quantified, as much as where the shot is from, and what type of shot it is, things a model may factor in
So if the only 3 shots taken are the ones resulting in goals 3 and 4, and the one that Stone deflected, the xGA is probably no more than 1.
Very saveable shots contribute a small but non zero xGA.
My point simply was what it was. The guy asked what was awful. I said the xGA model, because it has limitations. Low total shots and high number of actual high danger (and again not ‘high danger’ as models are able to measure) chances are anomalous, and the model which is a simple weighted average comes up lean