Quote:
Originally Posted by DeluxeMoustache
And those models don’t capture the things that matter, and make a difference between the jobs that goalie A and goalie B are asked to do
In terms of analysis, they mask the real problems the team has
My opinion is based in the actual events, not a quantification of a subset of measurable elements of the event
I lol because
- you even agree that the models have limitations
- I have a hypothesis about how the limitations and general observations about the Flames’ play
- I lay out in detail where the gaps are and what the model misses.
- I support it with detail on specific events
It makes me laugh because you on one hand, you talk about my gut or opinion. Then you turn around and say the goalies at the other end are getting the job done and ours aren’t . Sounds like your gut to me. Because I’m telling you how the shots are different.
They are simply not being asked to do the same job.
Although you acknowledge the models have limitations, you steadfastly refuse to accept the general notion that the deviation of the Flames W-L from the model is actually related to the model’s shortcomings.
Just gotta get a save, amirite?
|
Sigh ... we do this every damn game.
1. The models aren't perfect and they will improve.
2. The models are applied to every team and every goalie
3. The models take into account danger by region on the ice and entry point for the shot (tip, rebound, pass into home plate)
4. So a team that doesn't shoot dangerously will have less counts on the scoring chance and high danger level and with that accrue less impactful shots towards expected goals.
They're not perfect. They can certainly be wrong.
But over the course of a season it's certainly doubtful that a team or a goalie has found a way to generate for or against in these models in a consistently different way than other teams or goalies.
That's the rub.
The Flames with the same coach and system had top 5 metrics in most defensive categories last year and had the 4th best goaltending. This year same metrics and the 29th ranked goaltending.
But you want everyone to believe they've found a way to give up higher danger chances than the opposition in the same system with the same coach getting the same metrics which I'm just not willing to concede (nor should I as there isn't a lick of data to support that)