Hey Former Avs: How You Doin'?

I think he’s had a couple of glaringly awful turnovers and plays, and they were truly abysmal, that people noticed. Those bad plays have people more carefully deconstructing his game, and there’s a large bit of confirmation bias present with it.

I think, as I say in my blog, he’s been uneven. But it’s also… been 13 games, and b, he’s playing the best competition.

I guess I don’t think he puck moving has been that horrendous. It’s clearly regressed but to the point we should be benching him, or looking at trading him? I don’t think so. (also if his puck moving was so atrocious, he’d have a a lot higher SA/60 than he does, especially since he’d be giving the puck away to better players. ).

(Also I want to note that it’s very difficult to argue against you and separate your points from other points made in these here parts. I obviously disagree with the harshness of your evaluation of his play but I don’t think it’s outside the realm of reason. However some of the other evaluations out here seem downright preposterous to me.)

That’s kind of what I mean though right. We see one bad play and it becomes an image in our head. Then we see everything through the lens of that one bad play, and then it becomes confirmation bias.

seriously, yes, they do suck. especially wrt defenders.

I’ll go into more detail later (at home, not at work) but I think they are very good, albeit no where near perfect, measurements

Yeah, I do believe in the stats I cite. (If I didn’t I’d be kind of a prick, just citing stats to be contrarian). Last season, when the Avs Corsi was awful yet the Avs were winning games and the Corsiatti were almost reveling in calling the Avs lucky, I read up on quite a few of the advanced stats out there, including Irreverent Oilers Blog (Vic Ferrari & company), the Contrarian Goalie, and Behind the Net.. They have had some really good stuff on Shot quality, shots, and how it relates to a players play and chances of future success. I don’t always agree with their conclusions (For example they say that there’s no difference in shot quality after a big enough sample size. After looking at the data, and how they interpreted it, I disagree that there’s an element of Shot quality in there. However, it’s relatively small).

For defenders I think it takes 5 stats to really tell the story. SF/60, SA/60, Qual of Team. Qual of Comp. & Zone starts.

The reason I like Shot differential/60 (SF/60 and SA/60) is because I feel they get rid of that shot quality component., to an extent anyways Corsi counts missed shots and blocked shots, which seems like double edged swords. They are a measurement of possession, but it seems like a team could excel in playing a bend, don’t break defense in which they get in shooting lanes and cause shots to hit players/go wide, and attack on counter attacks. Also I’m not so sure official scorers can get a good measurement of “missed” or “blocked” shots, but shots are easy. If it hits the goalie, it’s a shot. And, these days, shots are recorded pretty accurately by NHL scorers (although there’s are arena bias’ ).

Qual of Team and Qual of Comp are good (yet imperfect) indicators of linemates and opponents faced and Zone start tells you whether a player is being protected in the defensive zone. On ice sv %, is a pretty good one too, but it’s pretty specialized.

I think those 5 (6) stats paint a solid picture defensively. FWIW, as Gabe at Behind the net likes to say, Corsi and Fenwick (a specialized version of Corsi) are used in pretty much every single NHL front office as an evaluation tool. These stats aren’t something that just magically came from some math nerds head. They are something NHL front offices, the guys with their jobs, paychecks and reputations on the line, use to evaluate players.

I trust them far more than someone’s opinion. Even someone’s opinion whom I respect, like Doc, Dario, or Beachie. The reason is because we could get 20 smart, knowledgeable hockey people together (without sharing thoughts), watch a random player for any number of games, and unless he played lights out or utterly abhorrent there would be about 17 differing opinions on how well he played.

That said, hockey stats aren’t very accessible. That’s another reason I like SF/60 and SA/60 more than Corsi is because the numbers actually mean soemthing to me. For example Quincey: 35.7 SF/60 and 29.6 SA/60. That means if Quincey played an entire game (at even strength) the Avs would outshoot their opposition 36-30 (on average). That’s something I think fans understand, because we know how many shots are in a game for both teams usually. Corsi is much more abstract (I’m not sure how to actually calculate it). QoC and QoT are abstract as well, but are based on opponents and teammates Corsi.

I think that the people who promote the advanced hockey stats just haven’t done a good enough job explaining them yet. Maybe that’s an unfair criticism, and I know it’s a complicated sport to break down into stats, but, like QB rating, hockey stats sort of suck.

I think that’s a very valid criticism. Like I said, I did a lot of research into exactly what Corsi is, how it correlates to team winning etc just so I could argue properly with some of the those guys over on those other sites. And I also have a pretty heavy Math background