Depuis l’antiquité, l’homme a joué à des jeux où un objet était frappé avec un bâton incurvé.

It is, however, likely only a matter of time before the league adds this moot measure to its nightly game sheets, one more “baseball” stat to appease those who fail to comprehend that the best explanation for the game of hockey lies in a slight variation of the phrase “scat happens.”

I thought \"#### happens" was something stat-heads are accused of saying too often (ie. the whole "luck" thing).

Hockey is of course an entirely different beast than baseball. The way the game is played and the interdependencies of the players makes it way harder (impossible?) to determine an individual player's value the way you can in baseball. But I'd be very itnerested to hear what smarter people than I have to say about improving our knowledge of hockey.

Faceoffs seem to be a silly thing to ignore, but there is plenty of hogwash (+/- for instance). I think the only reason it's still used is because no better defensive measure has come up. Blocked Shots as a statistic seems to be increasing in usage the last couple seasons.

But in the end I just don't think it will be possible for me to read a box score of a hockey game I missed and be able to piece together what happened the way I can with baseball. I guess we're spoiled in that way.

Depuis l’antiquité, l’homme a joué à des jeux où un objet était frappé avec un bâton incurvé.

Oh boy, time to bust out my grade 9 French.

Since time immemorial man has played a game where an object was hit with a (I'm guessing) curved stick.

Huzzah! Only had to guess for one word.

I'd go as a 'stat-zombie' for Halloween, but no one would get it.

I'd go as a 'stat-zombie' for Halloween, but no one would get it.

Clearly there needs to be a SABR Halloween party.

#2 Disagree with your take on +/-. +/- is essentially all that matters (at least in terms of evaluating a player at even strength), though there are a lot of adjustments that need to be made. The information is there to give you what you actually want from +/-. (which is conceptually goals scored and allowed while on ice above a replacement level player. Yes, you do have to adjust for who you're playing against and with. Including an adjustment for goaltending)

And to evaluate how effective a player was on special teams.

Hits, blocked shots, faceoffs,etc. Dunno. They're not precisely trivia in that they paint a picture of how a guy plays. But if they're valuable they must manifest themselves on the scoreboard.

I picked through the details of this years ago and was pretty satisfied with the results. No longer interested enough to do the work. I'd be surprised if it's not being done by somebody.

What the #### is a 'slide rule' and why do anti-statheads always mention it (in very close to 100% of these types of articles?)

I'd go as a 'stat-zombie' for Halloween, but no one would get it.

SABR ought to throw Halloween parties. I think I'd go as Carl Pavano's mustache.

I would stuff two pillows under my shirt, wear a skull cap to appear bald, plaster all the obviously inaccurate/flat out wrong statements to my person and go as BPro

Since time immemorial man has played a game where an object was hit with a (I'm guessing) curved stick.

Huzzah! Only had to guess for one word.

I had forgotten "depuis" and "frappe", but shouldn't "a des jeux" be translated as "games", not "a game"?

I would stuff two pillows under my shirt, wear a skull cap to appear bald, plaster all the obviously inaccurate/flat out wrong statements to my person and go as BPro

I'd put on a black robe, skeleton mask and a scythe and go as Bud Selig.

I had forgotten "depuis" and "frappe", but shouldn't "a des jeux" be translated as "games", not "a game"?

Quite right, I spent about half an hour trying to muddle through joue and jeux (at one point I was thinking a young girl was involved somehow) then when it dawned on me it was "play" and "game" I was so proud of myself I totally forgot about the words inbetween and its plural form.

You were probably thinking of "jeune," which means "young." Young girl: Une jeune fille.

Depuis l’antiquité...

So, what's this from? If you google hit, you get a number of hits. It looks like it's from the French wikipedia entry on ice hockey--is that it?

#2 Disagree with your take on +/-. +/- is essentially all that matters (at least in terms of evaluating a player at even strength), though there are a lot of adjustments that need to be made.

This is more or less when I meant. As is it's not terribly useful.
I remember once fiddling with +/- to get an "expected +/-" of what a normal player on team Y would likely to have as a +/- in X amount of minutes. Then comparing that to what the actual player got. Not nearly taking into account all the variables you mention, but it brought up some interesting results.

It does sound exactly like the sort of thing I'd enjoy reading, I don't suppose you saved any of your tinkering?

Of course at the end of the day, even with corrections it's still a fairly rough aggregate of events. Everyone's going to get their share of minuses because they happened to jump over the boards at a poorly chosen time. You just hope that all evens out. Though I suppose it's true of baseball too. Some guys are just going to get more bloop singles than others...luck of the draw.

You were probably thinking of "jeune," which means "young." Young girl: Une jeune fille.

Most likely, I'm always thinkin' of the Jun Fills.

You're always thinking of the son of June?

No, that would be poika kesäkuussa...duh!

A run through of "young girls" in google translate has the following candidates for most apt description

g?nc q?zlar
unge piger
tüdrukud
fiatal lányok
and
cailíní óga

I think if you try to pick up a young lady with any of those you'll be sitting pretty.

#15 No chance. The work went away two computers ago.

What the #### is a 'slide rule' and why do anti-statheads always mention it (in very close to 100% of these types of articles?)

A "slide rule" is a ruler, with moving parts that allow it to be used as a primitive calculator. It's basically one step up from an abacus. In ye olden days, you would use them to work out maths problems e.g. to calculate logarithms. Anti-statheads mention them because it is their era's way of saying "computer nerd," and because they are ludicrously out of touch.

You're quite feisty for a guy who wears a pocket protector.

Clearly there needs to be a SABR Halloween party.

Costumes for Sexy Stat Geek and Sexy Phillie Phanatic are still available.

Ahh, the slide rule.

I come from the dawn of history -- i.e. the period when the geek world was transitioning from slide rules to calculators.

Our physics teacher -- a Jesuit who was approximately 300 years old -- insisted we use slide rules. We objected since we had these superior and "expensive" calculators we'd had to buy for chemistry the year before. He insisted we use slide rules. Pressed further, he asked us what we were going to do if our batteries ran out during a test. We responded we'd plug our chargers into the wall. I believe this response damned our eternal souls and he still insisted we use slide rules on the tests.

It's a delightful combination of futuristic and obsolete.

Kinda like this?

I used to know how to use a slide rule.

Been there, used one... but i still watch the hockey and baseball games too, just for the hell of it.

Speaking of zombies, this thread might be dead but I missed it over the weekend...

Hits, blocked shots, faceoffs,etc. Dunno. They're not precisely trivia in that they paint a picture of how a guy plays. But if they're valuable they must manifest themselves on the scoreboard.

Yes, but not necessarily in a short sample. The extended hockey numbers sound like component stats for pitching, like hits and walks allowed. In the short term, even up to a season, luck might not even out and a guy (like oh say Matsuzaka) might allow tons of walks that don't show up on the scoreboard or ERA. But knowing his walk rate gives you a better idea of how he is likely to perform in the future. So in hockey, maybe a guy wins a lot of faceoffs, but that doesn't show up in +/- since he isn't blessed with teammates that can convert faceoff wins to scoring chances, or maybe even he just played in a division where the opposing goals are manned by Brodeur and Lundqvist and a healthy DiPietro. But knowing the faceoff stats tells you that this player might contribute a lot to the scoreboard if brought to a team where he can feed shooters like Crosby or Ovechkin. Hockey could very well be a game of component stats like baseball. The core dynamic of shooter vs goalie is not so unlike that of pitcher vs batter, even down to the small sampling rate of ~30 trials per game.

Adjusted plus/minus is essentially the currency of hockey sabermetrics. The adjustments in the case are situation (even strength), quality of teammates, quality of competition, quality of start location (much harder to score if you only play when your team has defensive zone draws, etc). There are also versus replacement level versions, I believe.

Corsi is essentially plus/minus based on shots for and against instead of goals. Basically, the FIP to plus/minuses ERA.

Alan Ryder's 2010 review just came out in the last few weeks. I get the (possibly mistaken) sense that he deviates from the consensus a little more than others, but it gives a pretty good overview of modern hockey stats: http://hockeyanalytics.com/2010/10/2010-nhl-review/