There was an interesting note in Elliotte Friedman’s 30 Thoughts this week about the Ottawa Senators.
8. MacLean said last week the Senators gave up 11 chances per game last season, and are up to 18 a night in 2013-14. That’s tough on your goalies. The organization is concerned with repeated mistakes and with the lack of progress by Jared Cowen and Patrick Wiercioch. You can see the stress on MacLean, going through his first major adversity as a head coach in an intense market. But, his job being in question makes little sense to me.
I am very skeptical that this is true. One of the things that people digging into numbers have found out is that Corsi% (or Fenwick% or Shot For%) is highly correlated with scoring chances. Eric Tulsky sums it up here:
There is a strong correlation between shot differential and scoring chance differential, so the simple shot differential tells us most of what we need to know. And since others have shown this for a team here or there, this isn’t news; the analytical community has long known that shot differential is very important and shot quality effects are minor.
I assume that MacLean is talking about scoring chances in all game states. An increase from 11 per game to 18 is astonishing. The Sens have given up 57.6 shot attempts per 60 minutes this year. Last year, they gave up 54.1 shot attempts per 60 minutes. If we look just at shots on goal, the Sens have gone from 30.8 SA/60 in 2012-13 to 34.2 SA/60 this year. So a 6.5% in shot attempts against, an 11% increase in shots against (the Sens don’t block shots) and, if MacLean is to be believed, a 64% increase in the scoring chance against rate.
I don’t believe him.
There are a lot of people scattered about the internet counting scoring chances. I was so skeptical of MacLean’s comments that I dropped a line on my Twitter feed to see if anyone was counting the scoring chances. Sure enough, a fellow by the name of @Wham_City was counting chances.
— scott (@Wham_City) April 8, 2014
The columns on the left are scoring chances in all game states; the columns on the right are scoring chances at ES. Let’s focus on the scoring chances against. He’s counted 3.5 chances more per game than the number than MacLean gave for 2012-13 and about 1.5 fewer per game than the number that MacLean gave for this year. It’s worth noting that the Senators were getting absolutely speed bagged during the period that he counted this year – they gave up about 61.4 shot attempts per 60. In other words, the time when he was counting was during the worst period of the year and he’s still not showing anything near the increase suggested by MacLean.
Let’s deal first with the overall numbers. According to @Wham_City, the Sens gave up 0.2634 scoring chances per shot attempt against in 2012-13. I took that to four digits to distinguish it from the rate at which he found they were giving up scoring chances in the first 21 games this year: 0.2631 scoring chances per shot attempt against. If you look at scoring chances per shot against, you get 0.463 in 2012-13 and 0.451 in 2013-14. It’s bang on, which is consistent with what Tulsky found about the relationship between scoring chances and shot attempts.
If we look just at ES, we see the same thing. In 2012-13, he found 0.274 scoring chances per shot attempt. In 2013-14, he had 0.251 scoring chances per shot attempt. In 2012-13, he found 0.488 scoring chances per shot; this year, he found 0.427 scoring chances per shot. Again, it’s very similar and, if anything, the scoring chances per shot attempt or shot are down this year.
Robert P., of Jewels From The Crown chimed in with another data point about the even strength shots allowed by the Sens:
— Robert P. (@RobertJFTC) April 8, 2014
So we’ve got three pieces of data here that suggest that the Senators’ scoring chances allowed did not explode: Tulsky’s study, @Wham_City’s counts and that from Robert P. Against that, we have MacLean’s stats. I’ve got no idea what’s going on here, although I have three guesses. First, Ottawa has switched scoring chance counters this year. I don’t worry too much about who’s counting the chances – in my experience, most people who consume a lot of hockey end up with pretty similar ratios of chances for/against, particularly over an extended period of time. Second, the fact that so many goals have gone in on Ottawa’s goalies this year may have caused the counter to treat shots differently.
The third possibility, and the one I’d bet on if I had to guess, is that this is some politicking. If you’re trying to convince management (or ownership) that you need new defencemen, one way to do it is to claim that the scoring chances against have exploded. You just have to hope that management and ownership don’t know enough about the relationship between scoring chances and shot attempts to get suspicious.Email Tyler Dellow at email@example.com