"This shows how stupid the Corsi thing is..."

How Patrick O’Sullivan keeps getting ice time is beyond me. . . . He is driving his negative traditional plus/minus (and pulled down Shawn Horcoff with him).

Of course, when the Oilers traded for him last year, one argument we heard was that he must be a good player, look at his great Corsi number in Los Angeles.

Corsi isn’t full understood only by its critics, but also by its advocates. It’s a new stat. I think it’s fair to say we’re all trying to figure out how much merit it has in rating individual players.

I’m not on the fence anymore, as it has too many false negatives and false positives for me to accept it, but I can see it has some uses in rating line play, certainly in rating team play, and likely in rating how players do with various linemates, as Derek has done recently.

Gabe, did you find a video link to that edition of Coach’s Corner. I can’t find it. The one on the CBC website for March 27 has no mention of Corsi, so I wonder if he did two version of his segment. I’m curious to see how Ron MacLean explained Corsi to Cherry.

That should read “Corsi isn’t misunderstood just by its critics, but also by its advocates.”

Of all the “advanced stats” Corsi is the most useless.

I don’t know if Gabe is old enough to remember the Soviet teams of the 70’s but they would have garnered the worst imaginable Corsi ratings since they refused to shoot the puck unless they had a prime scoring opportunity and would, in fact, withdraw from the offensive zone if they didn’t like the setup. They won more than a few games.

Corsi also assumes that all teams have the same propensity to shoot the puck when, in fact, they don’t. Weak teams like the Oilers will shoot from anywhere while strong cycle teams will get artificially penalized as they spend far more time in the offensive zone but take fewer shots.

If you want a meaningful stat, do the work and log zone possession. It might actually reveal something useful.

Gabe, you write, “Obviously (Corsi) assigns group credit for individual skills. But so do most other statistics that we don’t sneer at – play with Mario Lemieux and you’ll score a bunch of goals; it doesn’t invalidate goal-scoring.”

I think there’s some valid reasons why we tend not to sneer at goal-scoring (although Warren Young did get some heat, didn’t he?), but we do have some questions about the value of shots-at-net when it comes to rating individual players.

On a goal scored, we can accept that Lemieux does much to create the goal, but his linemate, say Young or Rob Brown, still has to tap it in, still has to get into position to score, fire it home. So we know there was some contribution from Rob Brown, maybe small, maybe big, We know, in some way, he earned that plus, and if you put this same guy with other decent forwards on a top line, he’d be able to make those same goals now and then, not as much as with Lemieux, but often enough.

On the other hand, when Lemieux and Brown were on the ice, we don’t know who was truly responsible for all those plus shot-at-net marks. If we watch the games, we know that surely Lemieux drove those plus Corsi marks, that territorial dominance, and Brown didn’t add as much. But we can’t be sure of that just by looking at their Corsi plus/minus numbers. We have to watch the game to have any real knowledge here.

This is even more true at the defensive end. All those shots against the Penguins when Brown and Lemieux were on the ice, were they Lemieux’s fault, Brown’s fault or someone else’s fault?

We don’t know, not unless we watch each shot against the Penguins’ net. If we did that, we would see, I strongly suspect, that Brown and Lemieux aren’t responsible for many of those shots against, maybe 40 per cent for Lemieux and 30 per cent for Brown, because he’s a winger with less defensive responsibility.

The rest of the shots at net that ended up as negative marks for them, but they had little or nothing to do with those shots against their own team.

Finally, when we think about why we put significance on a guy who scores goal, we know that if Brown scored a goal, and got a point, that the goal was a significant event, the only event that really matters in hockey.

But we don’t know really know the relationship between a shot at net and a goal being scored, do we? What I mean is, yes, we know there’s a strong enough co-relation between teams that have lots of shots at net and teams that score lots of goals, but perhaps there’s no real causation here.

Maybe, in fact, those goals don’t come so much from territorial dominance, but because the team is better at all aspects of the game, including: forcing two-on-ones through strong defensive checking; moving the puck out of its own zone; making quick and finishing rapid, lethal counter-attacks.

My guess is that 20-40 per cent of the goals come from territorial dominance — which Corsi will measure with all kinds of plus marks for each player on the ice — but the rest of the goals will come from rushes up the ice, which Corsi will give just one or two plus marks for (if there’s a shot and a rebound goal).

So a player like Lubo Visnovsky, formerly of the Oilers, might have a good Corsi number, because he moves the puck up ice well enough, but he also makes a high number of catastrophic errors that causes goals against. Not many negative Corsi marks comes out of those catastrophic mistakes, but a lot of goals against do.

In this way Corsi can mislead us, and I’d suggest that’s why Corsi told us Patrick O’Sullivan was a good two-way player with the Kings in 2008-09. It didn’t measure the nasty quality of the shots-at-net he caused, just the total of them. It didn’t measure how good players like Dustin Brown and Anze Kopitar drove positive Corsi, just that O’Sullivan was on the ice.

In closing, I accept that the top scoring teams will have an edge in Corsi, in territorial dominance, but I’d suggest perhaps that’s really not the key to that team’s out-scoring. Instead, it’s a sign of a strong team.

The way that team really manufactures goals isn’t through territorial dominance. It scores so much because of the way it creates and cashes in on high-quality scoring chances that spring out of battles won and great plays in the neutral and defensive zones.