Thursday, January 15, 2009

 

Flames Game Day

**Without wading through my archives to confirm, I'm pretty sure that this is the biggest NW Division lead that any team has had at any time since the lockout.

It's a nice cushion to have. The Flames have played 4 extra home games so far, and 13 of their final 20 will be on the road, so the schedule definitely gets tougher from here. (Although it's funny how your perception of the schedule difficulty varies with the quality of your own team; as I browse through it, I'm really only looking for long road trips and Sharks or Wings games. It's "that seems doable" for the rest of it.)

**Wonder of Wonders Dept.: Yesterday morning I listened to one of the most entertaining sports radio interviews in memory, and it was with Adam Proteau of The Hockey News. Why was it so entertaining? Because instead of talking about hockey, they talked about Proteau's past experience as an extra late in the Degrassi Junior High/Degrassi High run.

"One week I'm doing regular high school stuff, and the next week I'm behind the scenes at Degrassi with all the major characters smoking and playing cards for money. It was like Krusty the Clown." He also talked about being at a house party where Caitlin (wooo!!) and Neve Campbell were dirty dancing *with each other* -- "an image that is burned in my mind to this day". Good stuff.

**One of these days I need to write an opus defending plus/minus. It seems like the only time anyone (including me) can write about it is to expose its flaws, but the reality is that over a sufficiently long period of time (e.g. a few years, or a career), a lot of those flaws melt away. Again, the object of hockey (How To Win on the board game-style instruction sheet) is to score more than your opponent, and alternative efforts to measure contributions to this objective leave something to be desired.

I admire Staples, for example, for his ongoing experiment of tallying points & "unofficial assists" and setting them against "errors". There may well be something of value in there when it's all said and done, but some adjustments will be required if the totals are to make sense on their face. The Oilers are +10 on the season at "true" EV, but their forwards (of whom there is generally no more than 3 on the ice at a time) are +81. Likewise, Dwayne Roloson has an above-average SV% at EV this season, but by Staples' numbers he's done nothing but cost them goals. So some rejigging (or determination of proper baselines) is required.

I suspect pretty strongly that if you
  1. took the career +/- of everybody drafted last century (or say 1979-1999),
  2. deleted the 3 types of events I identified in my Brind'Amour post, and
  3. applied a season-by-season adjustment for team quality,
you would have one hell of a good list of the relative (EV) worth of those players.

**As much as I enjoy the discussions everywhere of Vincent Lecavalier's relative worth among NHL players (personally, I think he's closer to Mike Cammalleri than he is to Joe Thornton), it's really putting the cart before the horse. The most important issue by far is his value for his contract, not what Team X would have to give up in trade or by how much would Vinny help Team X right now. A perfectly wonderful hockey player on a contract that is too rich or too long (or both) is a liability in the grand scheme of things, plain and simple.

[Sidebar: Scott Taylor was on the radio this morning, and as an aside during discussions of the Vinny-to-Habs rumours, proclaimed that the Kostitsyn brothers are "much better than the Sedins". Sports pundits say a lot of dumb things, but Taylor has the early lead for Most Ludicrous Take of 2009.]

**There's a lot of good quasi-academic work being done in the hockeysphere, mostly in the area of statistical analysis. What I think would be cool is if someone was really into Literature Review-type stuff. At the top of my mind right now is the systematic disconnects (where they exist) between (A) the beliefs & expectations of the fans/media, and (B) the beliefs and actions of coaches (and also management).

I swear: the next time I read a handful of articles/blog posts/etc. about a slumping hockey team, and no one criticizes the coach's propensity for mixing lines ("How the hell are these guys supposed to develop any chemistry!"), it'll be the first.

**Tonight the Flames are in San Jose to face the Sharks (830PM MT, RSN West). The borg says the Flames are going down, but that's not how I roll. I predict that Dion Phaneuf goes off offensively, Rob Blake goes at least -2, Patrick Marleau continues to make me regret every bad thing I ever said about him, and both goalies make a lot of big saves in a 5-4 Calgary victory. Go Flames.

Comments:

Ahem. BIG saves. Do it right or don't do it at all.

And, with regards to Lecavalier, I just can't get my head around people talking about trading for him like it's a good thing. The contract is enemy number one and would put me 30/70 at taking him, even if he was free. But he's not free. Tto get him you, apparently, need to give up every good player in your system. Awesome.

If you're thinking about Lecavalier, just acquire Daniel Briere, he does the same damn thing, for the same money, but I bet Philly would give him away for a bag of pucks and Rob Schremp's thong.
 


Wow, Matt, lots of great stuff there. +/-, Staples' work, Literature Review, agreed on all counts.

Whatever happened to that little project of yours explaining every advanced statistic? I understand that it's probably insanely time consuming, but since your time is most valuable to me when it's spent on this blog, could you be less selfish and just write more ;)
 


If you're thinking about Lecavalier, just acquire Daniel Briere, he does the same damn thing, for the same money, but I bet Philly would give him away for a bag of pucks and Rob Schremp's thong.

Last season, Daniel Briere ranked 9th in QualComp among Flyers forwards. He managed 1.84 PTS/60, -10.0 Corsi/60 and -19.

At the same time, Lecavalier led his team in QualComp. He managed 2.42 PTS/60, -1.1 Corsi/60 and -15.

The comparison really isn't there; Lecavalier has been facing tough players for the past two seasons, and still doing better than Briere.
 


Whatever happened to that little project of yours explaining every advanced statistic? I understand that it's probably insanely time consuming, but since your time is most valuable to me when it's spent on this blog, could you be less selfish and just write more ;)

I had an idea regarding that, but figured that if it came from me it wouldn't exactly be met with open arms. Oh well. Let's see.

I think the idea generally is that someone writes something and someone nitpicks one aspect of it and the assumption is that it should all be thrown out. The benches clear and it gets nasty.

I have access to a server, and am able to put whateverthehell I want on there. I propose I just put up a wiki already, and then people can come together to fill in the blanks. If people have nitpicks for wording, they can change the wording without starting a firefight. I know certain people seem to be very protective of their writing, but really I think a lot of people would get a lot of use out of the information batted around. Hell, even just having people posts scripts with regards to how to pull data from NHL.com would go a long way to people making their own versions of Vic's TOI, to get the answers they want without having to wait for him to make the webpage/form for them. Everybody wins. Neutral, public territory. Discussion of ideas without trolls. Crazy enough to work?

Anyway, I don't want to go through the effort of putting up the Wiki if it's just going to be me, but if you guys can drum up support for it, I can set up your infrastructure.
 


I will forever maintain that the QualComp for Tampa in 2007 was completely screwed. I don't doubt that Lecav got stuck playing against the toughs, but the rating is ridiculous. I guess that's what happens when every player on the team sucks otherwise.

Briere is a pretty big step down from Lecav, I don't disagree there, but he's going to be free. If the option is Briere and then whatever we can turn Gagner/Cogliano/Eberle/etc. into or Lecavalier and an empty farm, I'd take Briere.

And, jumping back to QualComp, yeah, Lecavalier played the worst, but he had the best. Are we going to be able to provide him St. Louis and Prospal in Edmonton, after paying him? Is MacT going to suddenly implement the counterattacking system that Tampa had?

There's just a lot of assumptions involved in saying that Vinny is going to drop 110 points on a checking line while the rest of our money fends off the Iginlas.
 


I suspect pretty strongly that if you

1. took the career +/- of everybody drafted last century (or say 1979-1999),
2. deleted the 3 types of events I identified in my Brind'Amour post, and
3. applied a season-by-season adjustment for team quality,

you would have one hell of a good list of the relative (EV) worth of those players.


Matt, if you're interested in seeing a list of this type, I've calculated these numbers for all players since 1968. Some of the numbers are posted here.

It's not exactly as you describe, due to limitations of available data. The SH goals are removed, but exact numbers aren't available so they're estimated based on team SH GF and GA and the player's role on the PP and PK. Also, EN goals and 6-on-5 goals are not removed. It does compare a players on-ice EV record to his off-ice EV record.

The top 4 are Bourque, Orr, Jagr, and Gretzky, so it must be doing something right. 5 and 6 are more surprising - Mark Howe and Eric Lindros. It also rates great players on terrible teams like Marcel Dionne and Borje Salming highly.
 


Ender: I like the idea, but cautiously.

I had kind of a dull, dead-end job doing up various maps one year, and I spent a good portion of that year as an administrator on Wikipedia. There are some pretty nasty firefights and political stuff that goes on over there, so I wouldn't ocunt on a statistical wiki getting the job done.

Also, I'm not sure if this is what you're going for, but check out HAG on Yahoo! I don't know who all frequents that spot, but there are some pretty bright people and decent resources over there.
 


Briere is a pretty big step down from Lecav, I don't disagree there, but he's going to be free.

Only if the Flyers will take Moreau and Staios in exchange. Even then, I probably wouldn't do it.
 


Jonathan:

It's not so much for me, really. I've more or less washed my hands of the oil blogs mainly because I seem to drum up drama without intending to. That's not to say it's bad, or a waste that people are doing what they're doing, just that me commenting tends to be a bad idea.

And I don't really care about the stats either. Some are good, some are bad, but they don't really change the way I personally look at the game.

I mention the wiki idea mostly as a "put your money where your mouth is" sort of thing. I'm all for the communication and writing stuff down, I am. The issue is that anyone who does want to learn about the stuff you guys are talking about need to dig through a bunch of back-posts to try to figure out what's going on. When you add to that that Vic (nothing against him, but he's not particularly friendly to n00bs), controls most of the stats because he seems to be the only one with the script to pull data (although maybe Tyler does as well) you run into an issue where people who would like to do something new can't without fighting every step of the way.

Now that's my opinion, so take it with a grain of salt, but I just wonder - with how much everyone has talked about the new stats and why people should respect them, why nobody has even tried a wiki. It seems like all talk, no action, and I'm willing to put up a wiki to see if that's the case. Maybe to try to remove some drama, who knows.
 


quain said:

I will forever maintain that the QualComp for Tampa in 2007 was completely screwed. I don't doubt that Lecav got stuck playing against the toughs, but the rating is ridiculous. I guess that's what happens when every player on the team sucks otherwise.


I feel the same for some teams. In the case of Tampa the correlation with jonathon's metric (timeonice/jonathon.html, except I didn't bother to delete defensemen from the array) is .90. That's just a quick copy/paste/sort/delete anyone with less than 30 games and check correlation.

The general order is the same, Gratton moves up several notches, but at a glance it's similar. Desjardins painted his picture with a million dots, jonathon pushed some paint around with a spatula, yet the images are usually strikingly similar, team by team.


Of course playing tough minutes in the SE is a lot different than playing tough minutes in the west.
 


overpass: incredible stuff.

Both your methodology and your explanations/discussion are terrific if not impeccable.

I may devote another post to this, but if I don't, let me say that the *best* thing about a list like that is the spotlight it shines on guys who were on rotten teams for most of their careers and never made a good run at the Cup.

I think it's easier to maintain a "yeah, but what did he ever win" prejudice looking at boxcar numbers than at a list like this. Really, what more is someone like Mark Howe or Mats Sundin supposed to have done?
 


Of course playing tough minutes in the SE is a lot different than playing tough minutes in the west.

Very, very true. Or even from doing it in Philadelphia, for that matter, although the decreased focus on divisional play should lessen that somewhat.
 


Really, what more is someone like Mark Howe or Mats Sundin supposed to have done?

Pfff. Obviously they just didn't have enough "jam".
 


Pfff. Obviously they just didn't have enough "jam".

I always thought they lacked a sense of clutch, a fifth gear, and a proper air filter.
 


Jass, did you click on that Caitlin link? She has *still got it*. :)

(Sacamano and I "met" her when we were in high school.)
 


I think the wiki idea's fine in principle. Yeah, it's fraught with the same difficulties that any wiki is, in terms of vandalization, dipshittery, etc., but as long as no one takes it as Serious Business (I don't think you see the same kind of fighting on TV Tropes as you do on Wikipedia), it should work in principle, and I think it would certainly help accomplish a lot of the things Matt's tried to with his articles, and Vic with his (largely disused) NHL Code blog. I know I'd like to do a better job with the faceoff stuff I discussed with Bruce at MC's last week, but lack the tools.
 


Matt said:

Jass, did you click on that Caitlin link? She has *still got it*. :)

(Sacamano and I "met" her when we were in high school.)


Basement threesome?
 


Ha - Vic beat me to it.
 


Did I ever. If such a thing is possible, I think she's even hotter now.
 


I suspect pretty strongly that if you

1. took the career +/- of everybody drafted last century (or say 1979-1999),
2. deleted the 3 types of events I identified in my Brind'Amour post, and
3. applied a season-by-season adjustment for team quality,

you would have one hell of a good list of the relative (EV) worth of those players.


Got my vote.
 

Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?