Nice article Bingo.
Thanks Eldrick for posting this.
The use of the word 'advanced' to describe Corsi and Fenwick is very generous.
There has been the same fundamental problem with the really irritating overemphasis on sv%.
There have been articles where people have tried to assume shot location as a proxy for shot quality, and after that took it to where they had ridiculous implications. Because they couldn't make a significant correlation, they discard it and essentially argue that shot quality balances out over the course of a year and doesn't impact shooting percentage or save percentage. Then they go on to compare goalies on the basis of save percentage. (And don't get me started on looking at individual game sv%, what a waste of time)
I like Valiquette's approach to this, it is actually potentially useful. I look forward to the new NHL tracking capabilities giving the ability to qualify data by situation. Defining the groups of situations which lead to statistically significant results will be a part of the fun and I like where Valiquette is coming from
Quote:
Originally Posted by EldrickOnIce
|