Dustin Penner Drove The Bus Uphill

It’s the damndest thing, I read it in an article from another numbers-focused blog. Obviously, I didn’t bother to confirm it myself. I went back to the blog but what’s there doesn’t accord with my recollection. If others have been labouring under the same impression I’m not sure what to say. I generally don’t bother to double check the numbers on most of the blogs. Oh well.

I don’t like the way Gabe tracks zonestarts, by counting every faceoff a player was on the ice for. That may be a useful stat in itself, but it should have a different name, so that people don’t try to use it like Vic’s zonestarts. You can often get away with it, because they’re often close, but not always. That chart of defencemen is a case when they’re not close.

If you only look at faceoffs at the start of a player’s shift, you see almost all Oiler D in a pretty tight band around 40% OZone start%. Staios is the big exception, way up at 49%. And Johnson is the other exception, way down at 32%, but I think that’s mostly a sample size issue, because he started very few shifts with a faceoff (ie. changed on the fly a lot more than other D).

But the Oilers spent a lot of time trailing last season, and you gotta score goals to catch up, so it makes sense that they would start their good defencemen more in the offensive zone in those situations. If you only look at “close” games (tied in 3rd, within one otherwise), the numbers start to widen. I’ve got Staios, Strudwick, Smid, and Grebeshkov being (relatively) sheltered with 50%, 46%, 46%, and 45% respectively. And I’ve got Souray (especially), Visnovsky, and Gilbert doing the heavy lifting with 34%, 38%, and 39% respectively. Of course, this is only zonestarts from when these players were Oilers, not before or after trades.

Which zone a player starts in most often is going to have an impact on his Corsi plus-minus, but it will only affect his point totals a few points here or there.

Not many goals are scored at even strength in the 10-15 seconds after a faceoff loss. And a number of goals are scored on line rushes, which balances out the disadvantage of starting in your own zone.

Now, on the powerplay, losing that faceoff is a different and much more costly matter, but I think the over-emphasis on even strength face-off losses comes because those costly powerplay faceoff losses are deeply imprinted on fans.

Another interesting way of seeing who the money guys are is to look at who is being sent out there, that is whose shift’s beginning coincide with a faceoff and in what zone. Of course, this means you have faith in the general correctess of the shift charts but from what I could tell, running those numbers on the habs made sense.

Basically, you have a guy, Plekanec, who attended 389 DZone FO and another guy, Gomez, who attended 288 (they both had similar icetime overall). Gabriel noted a few months ago that some players are on the ice for many icing and Pleks is one of them. Plek’s breakdown was thus:

79 FO after an icing
111 coming along somewhere between the beginning and the end of a shift
199 actually occuring at the moment a shift of his began.

Gomez’s breakdown is 46/101/141.

Another side of this is some guys are on the ice for a lot of their opponent’s icings. Pleks was on for 59 of those, Gomez 75. The OZone breakdowns are:

Plek’s 303 OZFO: 146 faceoffs occured along a shift, and Martin sent him on the ice for 157.

Gomez’s 352 OZFO: 198 occured along a shift, 154 times Martin sent him out.

If you look at the last 10 minutes of the 3rd period, you see that Pleks was sent on DZone FO 50 times, Gomez 20 and Gomez was sent of 37 OZFO, Pleks 23.

One last thing: I’m not sure I ran the numbers correctly, but it seems Martin made at least one personnel change on 79% of DZFO, 62% of NZoneFO and 51% of OZone FO.

At some point I’ll run the numbers league-wide, if somebody doesn’t beat me to it (please, do so, it’s a pain in the ass :)).

Those of you who read french can see a few tables here and here.