All of that is true, it factors in none of that, but you are still giving it too much credit.
I don’t remember which resident stat expert told me this, but the publicly available expected goal models can’t even distinguish whether a preceding pass contributed to the quality of a chance.
Think about what that means…
It means a routine shot from the has the same difficulty as a guy receiving and finishing a pass on a two on one break.
It means a cross crease tap in is the same difficulty as a guy coming in with two defenders on his back and jamming the puck into the pads of a goalie already down in butterfly sealing the near post. Or the same difficulty as the third and fourth whacks at a puck already lodged up against the goalies pads… those all count as high danger saves.
In short: completely f’n useless in differentiating what would be a hard save or an easy save.
We need to wait a decade before we take any of those stats half as seriously as some around here do.
I appreciate your input and I couldnt agree more.
I am not a math guy BUT even I can see where some of these analytics fall short.
The HDSC is one area where I dont take the numbers too seriously. The examples you pointed out are exactly the kind of examples I have used on here before to show why that stat is flawed.
All you have to do is pay close attention to the details in a game and then the obvious conclusion is that the stat isnt granular enough to provide anything particularily useful.
This became obvious to me a few years ago when I watched a goalie seal the ice and the post with his pad and then watched a player (from 6 inches away) try to slam the puck through the pad 3 times. That puck had little to no chance of going in yet it all counted as 3 HDSC's.
The equivalent of 3 uncontested shots from 10-15 feet in front of the goalie. A place where the shooter can pick where he wants to put the puck and he is close enough in that the goalie has very little time to react. Not even close to the same scenario in terms of acutal danger and yet they are considered equal.
Also...(as you also pointed out) the HDSC's also dont take into account really important elements like contested and uncontested shots. Uncontested shots (in a dangerous spot on the ice) is something IMO this team gives up too much.
So we completely agree...that stat for the most part is basically useless in providing an accurate assessment of how dangerous a scoring chance really is.
Yet that flawed stat is used all the time on here to 'prove' how bad Skinner is.
