Tuesday, December 16, 2008

Gimme Shelter

In this corner of the internet there are two terrific posts discussing Sam Gagner. In the first link, Tyler points out that Gagner has not been playing difficult opposition, and that he starts his shifts with offensive zone faceoffs far more often than in his own end. Both of which are true and relevant, in my opinion.

Prior to the season, during the Oilers' visit to Jasper, MacTavish discussed the kid line. He said that opponents can match up against one line, but it's very difficult to do that with two lines. And he thought that the Horcoff/Hemsky line would draw the attention of the other team's coach, leaving the kid line to face softer opposition. He actually used the words "softer opposition", which is probably why this stuck with me. Coaches talk about hard minutes, tough ice time, players who play a lot against the other team's best players, etc. Obviously the opposite exists, it is obvious to the eye, plus simple logic says it must. It's rare to hear any coach besides Randy Carlyle address it so frankly, though.

Anyhow, it has been pretty obvious in a lot of Oiler road games that the opposing coach has been jonesing to get his best forwards out against the kids. MacTavish has done a pretty good job avoiding it, but on the road there is only so much you can do. Often he just has had to accept the matchup, and try to give the kids a head start by beginning the shift with and offensive zone draw, or with the Oilers in solid possession of the puck.

Now the upside is that if Arnott is being run at the kids in Nashville, well that chunk of ice time vs his line won't be played against Horcoff's line, so they should do a bit better. And a guy like Gagner should do much better on home ice, where he can be sheltered. And though all players should be expected to do better on home ice, players who can log tough ice time and do well with it ... well the home numbers aren't going to be so much better for them. They probably play better at home, but see tougher ice time there as well.

So for Horcoff, he has a corsi +/- of +54 on the road, and +15 at home. That makes sense, he plays tough minutes, and he's good enough at it that other coaches generally don't like the match-up with their own best players.

On a per-game basis, that's +4.7/game on the road, and +1.5/game at home. For a difference of -3.2. That is shown in the chart below, as are the other Oilers.

We should see the best players near the top, complete players, the guys who can play power vs power and have success. And we should see the weaker players near the bottom, guys with narrower skill sets who need shelter from good opposition to have success.

I think that for the most part we see that here. I expected Horcoff and Visnovsky to be near the top, and Cole to outshine Penner by a wide margin by this measure, so it passes the sniff test. Of course there is surely a lot of randomness in there, and some practical reasons as well (relative health and linemates at home and away, etc.).

Still, there is a pretty strong indication of who is getting sheltered when MacTavish has last change. The surprises for me are Pouliot on the positive, and Brodziak and Moreau on the negative. And whenever something seems irrational on a MacTavish team, we follow the sound of the cowbell. Brodziak has a much better faceoff percentage at even strength than Pouliot (all players should be expected to have worse SH faceoff percentages of course, with fewer players on the wall to win the scrum if the puck goes there). Hopefully we see Pouliot taking the Carbonneau path soon, he's just a better hockey player than Brodziak I think. I'm not going to bother wasting my hope energy on a Moreau demotion to the fourth line, though, because it just won't happen. But you can do that if you want.

9 Comments:

Blogger slipper said...

I was having a look at Penner's faceoff differential before being pressboxed and since returning against the Blue Jackets.

-26 Before vs -2 After

Yet he's producing more, his linemates are producing more, and all of their underlying numbers have improved signifigantly (less than dramatically but more than some what).

Poor Brodziak, though. You have to feel for the guy: 103 D-zone draws versus 43 in the O-zone in a road heavy schedule. He's been the Oilers' best percentage option on face-offs while probably being their 4th best player at center. It's not really surprising to see this disparity between Brodz's home and away numbers with the opposing coach getting last change against him in the attack zone so often.

A couple more things that are worth mentioning regarding #51. First, I'm pretty sure when he's taing a draw on a mismatch that most often than not he's yanked as soon as the Oil clear then zone. So really he's put in a position where he's basically guaranteed either shots against, goals against, or a line change. Second, although he's started his shifts in the Oilers' end 103 times, he's only ended them there 75 times. This will require more in depth research, but on a cursory glance it appears that an average player is pretty linear with regard to those two numbers. Lastly, Brodziak has been benched at home twice this season so he's seen 18 games on the road versus only 8 on home ice where, if MacTavish has any sense, he has seen far fewer own zone match-up versus the Joe Thortons.

12/16/2008 2:29 pm  
Blogger Scott said...

Vic, first off thanks for the work here. Are these absolute numbers or are they organized in terms of a rate? I assume the per game thing means that they're absolute numbers. I think that makes a significant difference to some of these figures, especially for guys like Stortini, Reddox and Strudwick who are probably being flattered by being put in the middle of the pack.

I'm actually quite surprised by Penner's numbers relative to Cole's since the two have been put into similar situations both with Horcoff/Hemsky and also on some version of a Moreau checking line.

12/17/2008 1:16 am  
Blogger Vic Ferrari said...

On Penner, yeah, he's played well since he got back from the benching.

Thing is, he looked pretty good before when he started in the right end of the rink. He's one of those kinds of guys that would probably look great playing for Quenneville but would get ground down to nothing playing for Hitchcock.

He just isn't very good in the defensive zone or neutral ice. But he has good hands and the play doesn't die with him much in the offensive zone. So he's an obvious fit to play with good linemates in offensive situations.

A guy like MacTavish though, who seems to lean heavily toward good ice time (either PP or EV linemates) as a reward ... must be a piss off that Penner is the best fit for those gigs, and largely because he's not particularly effective elsewhere.

And somewhat ironic that MacTavish's powerplay, which seems like the league's least imaginative most years, and relies very heavily on point shots ... that's a natural fit for Penner as the big man in the crease.

12/17/2008 6:56 am  
Blogger Vic Ferrari said...

Yeah, Brodziak doesn't have a hope in hell of putting up good EV+/- numbers given the gig he's had. As you say, he's been very good on the faceoff dot at evens (granted, his linemates deserve some credit) so he's getting thrown out there for a bunch of them in his own end. And he's just not close to being the best choice for the guy to be on the ice after the faceoff is won or lost. He's not had great linemates, either.

I haven't noticed him subbing off the ice after a draw very much, granted quite a few games I've just caught on the radio this year.

This post is more about the swing from road to home though. I suspect that's mostly just because Horcoff's line is taking more of the burden on home ice. Teams have fewer own zone draws on home ice as well, that helps a guy like Brodziak. Maybe just a lot of it is coincidence as well, I dunno.

12/17/2008 7:05 am  
Blogger HBomb said...

Vic, one question about your tool for generating the Corsi numbers: where you've got Goals for and Goals against listed, is that EV 5 on 5 only, or ALL even strength situaions (i.e. 4 on 4 included)?

Great work as always. The more games I watch, the more I am convinced by my eyes that the guys at the top end of the Oilers batting order (specifically, 27-10-83, 71-44-77), they're not the ones that are the problem with this team right now.

Staios and Moreau, however, two veterans rewarded for years of service with four year deals that are currently only in year two - they've both been pretty shitty from what I've seen, and last night's debacle against Chicago was a microcosm of that.

12/17/2008 8:27 am  
Blogger Vic Ferrari said...

Scott

Yup, good point.

12/17/2008 8:32 am  
Blogger Vic Ferrari said...

In fairness to Staios, he was saddled with Strudwick for much of the early going. I think the other thing that has made Steve look bad is the PK work. He's never been any hell as a PKer imo, and seems to be getting quite a bit of first unit tie now as well. That was never going to end well.

12/17/2008 8:34 am  
Blogger Matt said...

And whenever something seems irrational on a MacTavish team, we follow the sound of the cowbell.

That is just a fantastic line. I intend to steal it (paraphrase it, anyway) in the near future.

12/17/2008 11:29 am  
Blogger HBomb said...

Vic: I'd agree with that about Staios and the PK. I have problems blaming one guy for something systemic. And yes, Strudwick wasn't helping matters.

But the fact is, Staios has lost a step and is a VERY marginal #4 defenseman at best right now.

12/17/2008 1:12 pm  

Post a Comment

<< Home