Quote:
Originally Posted by Enoch Root
If you are aware of the error bars, yes. But not only are the parameters not available, but most people looking at this would not know how to interpret them even if they were available.
You are implying here that there is NO way to evaluate goalies if this stat is useless, and that is a pretty ridiculous position to take. We can, you know, watch them. And people with an understanding of the position (i.e. goalie coaches) can watch them with a professional lens. Suggesting that we spend no more than the minimum cap space on them suggests that they have no value, and that there is no difference between them. Which is beyond silly.
It is widely regarded as the most important position in hockey, and one of the more important positions in team sports. And your bias against goalies is nothing short of bizarre. Suggesting that players in other positions never make great plays is also a completely ridiculous statement - of course they do. It happens all the time, and one need only watch highlight packages to see a perpetual stream of examples of this. More importantly, the fact that a lesser ranked goalie might make an incredible save, in no way refutes the fact that there are other, better goalies. And remember, even the 'horrible' ones, are among the best 60 or so in the world at their craft - of course they are going to make incredible saves, they are elite at their position. Again, that does not eliminate the possibility that other goalies may yet be even better.
In ANY position, art, or craft, there will be 'the best' and those that are progressively not quite as good. Anyone who understands stats will inherently understand this extremely simple fact (and I know that you do), which leaves your extremely outlying opinion of their value, and your weird view that there wouldn't be a rankable distribution of skill, as to be only explainable by some extreme personal bias.
|
You're right. I'm not claiming that goaltenders are unable to be ranked, but if people like you, me, DeluxeMoustache, and Jay Random (and I'm sure, many of our readers who don't post) can't agree on a method of doing it, then it means there is insufficient publicly available data, not that it isn't possible. This is where revenue rich teams like Toronto and New York have a huge advantage: they're all limited to the same on-ice salary cap, but there's no limit to the amount of money they can pay for analysis. We'd do it for free for the Flames because we (irrationally?) love them, but without access to proprietary data like a DB of shots, pressure, score of game, time of game, result of shot, etc., it's a foolish undertaking. And you'd need to pay people to watch every game and either subjectively or objectively make a dataset from it.
I'm not implying that it's impossible at all, but I am implying that given imperfect data, we can either dismiss it because of its imperfections or use it knowing its limitations.
Try to predict the top 20 goalies for next season using whatever metric you like. Then try the same exercise for 2025-26. It's pretty close to playing the lottery if you ask me. That's why I'd rather funnel money to contracts that are much easier to project (defense and forwards) and pay the minimum for things that are necessary but lottery tickets.
We were one goal away from Adin Hill and Stuart Skinner being back-to-back Stanley Cup winners, and I doubt any scout or analyst would have predicted it since there's far too much variation.