Monday, April 14, 2008

Odd stat on the PP

So I'm reading through a HF Post regarding the Grebeshkov signing when Dawgbone mentioned (in reply #151) that by the numbers Grebeshkov wasn't a great option on the PP this season. I wasn't sure this matched my eye (at least, when thinking about Grebeshkov at the end of the season) so I went to the behind the net site to look at the Oilers PP data.

Here were two data sets, regarding Oilers D, that caught my eye, and I'm wondering if anyone can explain the results to me:

In terms of points/60 min, he was 5th amongst Oilers D, behind, in order:

Staios 4.25 (very small sample size, 1 PPpoint in 14 PP minutes on the season)
Gilbert 3.01
Pitkanen 2.11
Souray 1.72
Grebeshkov 1.26

In terms of +/- per 60 min, Grebeshkov was 1st amongst Oilers D:

Grebeshkov 6.29
Souray 5.15
Staios 4.25
Gilbert 3.68
Pitkanen 3.28

I'm not sure what exactly causes this discrepency, if Grebeshkov contributed to it because of something he did, if he happened to be on the ice with the Oilers best PP players for a greater percentage of his PP time than other Oilers D, or if he was simply the benefactor of being on the ice while others made nifty plays to score goals. Any thoughts?

*EDIT* I now see thatJonathan at coppernblue has beat me to the punch. Wow, that's kind of weird, to have wondered the same thing like that.


Blogger PDO said...

Two answers.

#1) Dumb luck.
#2) With Grebeshkov on the ice on the PP, we were more likely to actually work the puck down low instead of living and dying by the point shot, which thus produced more offense.

4/15/2008 6:11 pm  
Blogger Vic Ferrari said...

MC has been talking a lot about PP shots this year, speeds. At first blush that seems wrong to me, but you'd have a helluva a time arguing against it honestly with facts.

I mean any hot streak, for a player or team, is driven by a hot PP shooting%.

The problem is that there is absolutely no repeatability or predictive value in shooting%, from the first half of this season to the last half. Not at either the player or team level.

At a glance, points and shots-on-net-while-you-were-on-the-ice are the keys here, especially relative to teammates. And the SH stuff is so small that it's mostly just noise.

So that bodes well for DET, CAR, MTL, NYR and L.A next season (with the same key PP guys in place).

And it bodes well for Hemsky (yowza!), Horcoff, Souray, Penner, Stoll, with Gagner creeping into that group by the end. And rising at speed, he'll probably be the best of the bunch within a couple of years. Pitkanen and Pisani fairly close to that group as well.

Again, it's just a rough and dirty quick look, but this doesn't bode well at all for Cogliano, Nilsson, Gilbert, Grebeshkov. Granted their all still young and should improve, but I don't think we can bank on continuation of some crazy high (20%+) PPshooting%'s while they were on the ice, like happened in the last stretch. Especially when these guys weren't getting the points to match.

PPshots+ rate, relative to teammates esp, seems to increase expected shooting% and reduce SH shots against as well, all at once.

Again, PPs are pretty much straight offence for the team with the extra man, Corsi, Fenwick and + and - fade in importance a bunch. On ice shots+ rate and plain old points-per-hour soar, at least by the looks of it.

I'll get back to it this summer if nobody else on the blogs digs up the info and analyzes it.

Back to point: Grebeshkov's EV+/- was built on terrific SHsave% behind him and a good PPshooting% in front of him. The numbers with real sustain (check Lidstrom, Jagr and Crosby, damn!), they just aren't there for Grebeshkov, in fact they are downright poor.

4/16/2008 2:24 pm  

Post a Comment

<< Home