Sunday, October 30, 2005

They call them the Streak

Well, damn, that sure was a fine win there, wasn't it? It wasn't on TV, at least not that I could find, so I can only go on Rod Phillips' primal screams and game reports, but hey, 5-1 against a team that hadn't previously lost in regulation can't be a bad thing. By all accounts, all those people that we needed to step up did, and we controlled play most of the night (sacamano has most of the highlights). It's a very nice bounce-back from the heart-crusher that was Tuesday (as an aside, I may have overreacted in that game report, but man, was that crushing).

Anyway, coming off one extended streak and potentially sashaying into another (we should beat Columbus, if we remember that this is how we should play), I thought I'd look into something: how streaky are the Oil? I think the initial reaction of anyone who has followed them the last few years is "really goddamn streaky," but if Tom Benjamin teaches us anything, it's to actually look into things before spouting off random opinions.

So, going back to 2001/02 (the farthest ESPN's schedule thing goes back), I looked at the Oilers' tendency to streak. Defining a streak as any stretch of three games with consistent results (wins, losses, undefeated [wins with ties interspersed], or winless [losses with ties interspersed]), I surveyed the results. Unsurprisingly, we like our streaks: in the last three seasons, we've spent at least 46 games each year in the midst of a streak, and have always had at least 10 streaks a season.

This, in itself, isn't really good or bad news: we can spend 60 of our games in streaks for all most of us care, if they're all winning streaks. But here's the flip side of the coin: good lord are we ever inconsistent with our streaking. In 01/02, we started the season with three straight "good" (wins or undefeated, obviously) streaks, followed that up with four straight bad streaks, then mixed winless/losing streaks with undefeated streaks the rest of the year. In 02/03, a lot more of the same: winless, undefeated, win, loss, loss, undefeated, undefeated, winless, win, undefeated. 03/04 went loss, undefeated, win, winless, winless, loss, win, winless, undefeated, undefeated, win. Most streaks average about three or four games, with the occasional ridiculous seven or nine (especially in 02/03).

Okay, so fine, my point: the team, over the last few years, seems sort of rudderless. They can put together wins, or losses, but exactly how, why or when that's going to happen is more or less a mystery (oh, except that we almost always finish strong). We'll go undefeated in four only to lose another four straight, go winless in nine but slap people around for seven games just a week later, and so on. Momentum is about as meaningful to the Oilers as a solemn prayer to Zeus.

Now, theoretically, leadership should be being able to keep a team consistent: sometimes a team just doesn't have the talent to win, but at least they're out there fighting for it, or a team has the talent, and just has to stay focused. Obviously, the Oilers were never bad enough these last few seasons to be entirely without hope, but they can't really put anything solid together for more than a 15-day period, either. It seems as though they're pretty much entirely at the whim of players going cold or hot, and when they're hot, our leadership sails smoothly, but when they're cold, all we can do is throw up our hands and wonder what's going wrong (as MacT did several times this past losing streak). One hopes the addition of people like Pronger and Peca would smooth this out--they both have "leadership," after all--but that hasn't been borne out yet (it's a little early to tell, though).

Essentially, then, here's what we got: we might be good, we might be bad. The fans don't really know, the pundits don't really know, but most disturbingly, the management never really seems to know. As a result, we let our sunny optimism make us warm each fall, then spend the season getting batted around like a wounded mouse in front of a curious cat. Now, I'm still going to enjoy it while we're winning, and sit in a dark room with a bottle of cheap bourbon when we're losing, but come the end of the season, if we haven't at least been consistently something, I'm not giving management any more rope.


sacamano said...

I'm not sure I'd attribute the Oilers' streakyess to bad leadership over bad players.

One thing that you can always say about the Oilers (well, almost always), is that they generally don't take nights off in terms of effort. They always bust their hump out there.

If it was a case where some games they tried hard and some games they just floated around (e.g., Rangers of old), then I'd be more inclined to blame leadership.

When it is simply a case of consistent effort but inconsistent results, I tend to blame lack of talent.

mike w said...

>I tend to blame lack of talent.


How much of a difference was their between Colorado and Nashville this week? We're really talking about a razor-thin margin of error when it comes to a 3-2 win as opposed to a 3-2 loss. We've already a won and lost a few games based on a few shots from the point randomly going to the right spot. Teams like Ottawa or Detroit can get by on talent alone, whereas the Oilers cannot. Not that the Oilers are special -- I wouldn't be surprised if half of the league were the same when it came to getting wins and losses in streaks.

Also: Oilers, Wooo!!!

Colby Cosh said...

I wouldn't be surprised either. But apparently the Tom Benjamin science playbook doesn't come with a section on experimental controls. (What would be an average number of streaks? Or a representative number for some comparable team over the same time?)

Also, I'm baffled by the way this tendency is normally interpreted. This team goes through many periods where it is capable of beating anybody. I would think that's a clear sign that there is nothing too wrong with the talent per se. When a team that's terrific at its best suddenly starts losing, and there's no personnel change you can point to as being responsible, how can you blame the talent level?

Also: if you had never heard of the "Edmonton Oilers", but someone described to you a hockey team in an isolated city--a team forced by geography to take long, punishing road trips, and one that often welcomed visiting opponents 24 hours after they'd played in another city--you would probably expect that team to be pretty streaky. If the Oilers are especially so, the map of North America may provide a more than sufficient explanation.

Jasper109 said...

I think it all comes down to basic principles of statistics.
Take a 0.500 team (flipping a coin)and you will have straks of several wins and several losses in a row. If you flip a coin 80 times the same thing will hapen. You will have HHHTHHHTHTHHH or something similar a number of times.
Other than truly outstanding teams like the old Islanders, Oilers, Canadians the playoffs are pretty well like flipping coins. The Flames 2 years ago could have just as easily lost to the Canucks in the first round, but just had that extra bit of luck.
The difference in talent among NHL teams is so small that predicting what will happen on any given night is basically impossible.

sacamano said...


In part the problem is how "talent" is defined. For me, a huge part of what is commonly bundled as "talent" is consistency. We all know of players who are all-world in practice or on occassion (e.g. A. Kovalev), but can't seem to put it together all the time.

The difference between a good player (team) and a great player (team), then, is not that the great player necessarily has better skills, it is that he brings it every night. Good players (teams) are always capable of going through stretches where they are great - but for whatever reason they can't do it all the time.

To me that is simply one of the elements in a player's skill set - like a hard slapshot or speed.

You might disagree and say that "talent" should only include the physical ability to play the game while consistency is something external - equivalent to "leadership" on a team level, but I find that just a bit too simple.

Colby Cosh said...

I think your quarrel is with the English language, then.

Pleasure Motors said...

Yeah, again, the problem isn't so much that they're streaky, but that they're incredibly inconsistently streaky: it's not like they put together three straight winning (or losing) streaks punctuated by the odd loss (or win), it's that they go on incredily disparate streaks, often back to back.

I really think if it was only a question of talent, even if you include consistency in that definition, we'd at least be consistent in some respect. You can point to Detroit or Ottawa, but a team like Minnesota has been consistent over the years, too: they just haven't had the talent to be consistently good. Lemaire gets solid, even performances out of his players, even if the end result is poor because of their talent. Look at other middle-of-the-road teams the past few years, like St Louis, Nashville or Montréal: they'd go on their streaks, but they wouldn't win four, lose five, win three and lose another four--they put them together fairly consistently. My issue is with the fact that we can't be consistent, not that we're streaky (though the later could arguably be said to be a condition of the former).

sacamano said...

Until I see some numbers that take into consideration the schedule (i.e., road games vs. home games, quality of opponents, etc.), I'm skeptical that the Oilers are any more streaky than anyone else - especially with teams having similar records as us.