For some reason, nobody seems to publish home/road splits in terms of ES shooting – it’s not available on Gabe’s site or Vic’s site, as far as I’m aware of. I think it’s particularly interesting in the case of the Oilers this year, so I put the numbers together. Long time readers of the site will remember that, about three and a half years ago, I wrote a post in which I broke down the home/road shooting ratios between 2003 and 2008. At the time, I found that home teams got about 51.6% of the shots (and, obviously, road teams got around 48.4% of the shots).
Here are the splits for the Oilers’ forwards:
I’m open to hearing what people think the data means. You’ll note that the Oilers had a somewhat bigger split in shooting at home/road than the league average between 2003 and 2008. I suspect it’s notable that it was the young guys who had the biggest advantage that came from being on home ice. I’d suggest that this is because Renney can hide them at home but not on the road. Interestingly, there isn’t a sort of reverse effect on the road – nobody really does better on the road or anything.
I speculated in a post earlier this year that you’re going to have a hard time hiding guys who play a lot of minutes. Basically, if you intend to play them a lot, they’re going to run up against the other team’s best a lot. I wonder if that’s what happens with the guys like Horcoff and Smyth – there’s no effort being made to shelter them anywhere, indeed, it’s the opposite, so they just run up against the buzz saw wherever they go. I’ve made this point more than a few times but even if you don’t think much of Smyth and Horcoff, you have to acknowledge that they do get fed into the chipper shredder a bit. You can’t replace them with the kids without the kids taking on more difficult minutes and, as I’ve shown previously, guys like Horcoff/Smyth/Hemsky have held their own in chances with the kids when you look at them up against comparable competition (Hall aside, who’s a bit of a special case in that he seems ahead of Eberle/RNH/MPS). It’d be fascinating to drill down into the next level of this – and I suppose I could – and confirm that what changes for the young guys on the road is the mix of top six/bottom six that they play. It seems the most likely thing to me.
I’m still trying to wrap my head around how, exactly home ice advantage works. I think it’s a mix of a couple of things – the road time is more likely to be on the back end of a back to back, IIRC. I’ve always sort of wondered if line matching is a zero sum deal. Assume you have 48 minutes of ES TOI to split up. L1 plays 16, L2 plays 13, L3 plays 11, L4 plays 8. The thing is, there are only so many ways that you can organize it and, for example, if the kids play more minutes against the other team’s better players on the road, it should mean that the other players get to play against the other team’s lesser players. I guess I’ve implicitly assumed it’s a linear thing; maybe I’m wrong about that.
The thing is, we don’t see guys like Horcoff getting a bounce in their shot numbers on teh road – everyone does worse on the road, with the kids getting absolutely hammered on the road relative to home ice. Hold on a second though – the Oilers played one home B2B last year and 10 road B2B games. That’s a pretty substantial difference – you get a lousy night’s sleep traveling to the road game and you likely haven’t had the recovery time that your opponents had.
As it turns out, there’s at least some evidence that there’s something to this – the Oilers did slightly better in non-B2B games in terms of shots than they did in B2B games on the road. They got 44.6% of the shots in non-B2B games and 43.1% in B2B games. It’s tough though, with the vagaries of scheduling, to account for all of the factors that might bear on this though – it’s possible that they had a softer schedule in B2B games, for example. IN ANY EVENT, this is all a bit of a digression – the key point I’m driving at is that this should be more evidence that the guys like Horcoff and Smyth are serving some sort of a purpose by taking bullets at home. We can debate how effective they are at doing it – I don’t know that we’ve yet done a good job of defining what good possession numbers are for guys taking on the other team’s best – but they’re providing some sort of value in sheltering the kids and it shows up in a couple of different ways when you slice into the numbers.
Also of note: Renney mentioned today a mistake that he made, leaving his fourth line out against the Flames. Darcy Hordichuk’s positive share of the shots at home might be my favourite Oilers stat of the year.
Well, three things leap out: 1) Man did Ryan Whitney just get absolutely torched on home ice; b) wow are Peckham and Barker ever bad on the road (again, it would be interesting to break down what percentage of their time they spent matched against the other team’s best on the road – that’s a lot of shots for Barker to have been on the ice for and be that bad; and c) doesn’t Tom Gilbert have curious splits, in that his numbers are significantly worse on the road?
Whitney: guys can talk all they want about how he just needs a summer to train and get some strength – I don’t believe a word of it. I’m not a doctor but he’s just been breaking down in front of our eyes for two years now. There’s nothing that can be done about it but I’d plan on him being a bottom pairing/seventh defenceman next year. If the Oilers find someone who can play top two pairings and Whitne’s healthy…well, there are worse problems to have.
I assume Barker and Peckham aren’t being qualified. They’re two more guys in the RNH/Eberle pile, where it would be interesting to see if the other team targeted them on the road.
I talked about Ladislav Smid, Tom Gilbert and Jeff Petry a bit a few weeks back, in terms of their chance breakdowns:
In any event, Gilbert and Smid were a really solid pairing together, considering all the other problems this team has – 87 CF and 84 CA against the other team’s top six. Fantastic stuff. Smid and Petry have been almost as good – 111 CF and 114 CA. Without Smid, Gilbert was 45 CF and 51 CA. Smid, without Gilbert or Petry is just horrible though – 25 CF and 48 CA (34.2%).
If I’m right that there was a consensus in the Oilers’ front office that Smid was carrying Gilbert, you wonder if what they were really seeing was that they had three defencemen who can handle decent opposition and that one guy (Smid) got to play with one or the other of the other two most of the time. Smid would then always look sort of decent, while Petry and Gilbert would suffer in comparison when they weren’t on the ice with him. The fact that Gilbert easily trumped Smid’s chance numbers when playing with the dregs suggests to me that there might be something to this.
I don’t have an eye or a memory for pairings and matchups the way that some people do but were Smid/Gilbert playing together more at home, where the coaches could target them against the other team’s best more precisely and then Smid/Petry and Gilbert/cast of thousands on the road, where they had less control and wanted to have at least one decent guy out against the other team? There needs to be some sort of way to easily check this stuff.
What does it all mean? Well, there are a couple of questions that I think it’d be helpful if management could answer correctly. 1) Can you get better performance from players who are available than you get from Smyth/Horcoff/Belanger (god is it depressing how long those guys are signed for)? I don’t know the answer to the question but, man, the kids are gonna have to be good to make up for the difference. 2) Is there a real top four defenceman out there? They need another one pretty badly.
The first question is more interesting to me, from an analytic point of view. Anyone have ideas as to how it could be fleshed out? My thinking is establishing some sort of benchmarks for what guys on other teams do with similar ice time. The numbers seem pretty abysmal to me, even if I think there’s value in providing that shelter.