Friday, February 25, 2011

What's In A Name...Or a Conference?

In yesterday’s online version of The Hockey News, Adam Proteau penned an interesting essay arguing that the NHL’s division structure should be scrapped in favor of straight-forward conference rankings. His is essentially a three-pronged argument. First, it completes the modernization job the league started back in 1994 (when the named divisions – Patrick, Norris, Adams, and Smythe – were retired in favor of geographic names). Second, it makes understanding the standings easier for the casual fan, who no longer has to ponder the meaning of a division leader getting an automatic ticket to a top-three conference seed. Last, Proteau offers an equity argument – it is fairer to teams that are more successful than a weak division winner that gets a free ticket to a top-three seeding.

Proteau makes a logical argument. What he argues would streamline the organization of teams (in fact, would go in the direction opposite to that which Major League Baseball has pursued over the last 40-plus years by adding divisions to a “league” structure) and would provide a level of competitive transparency (only one structure in which the teams compete).  I just don’t happen to agree with the prescription, and here is why. Rivalries.

Football markets the logo on the jersey, basketball markets players, and baseball markets its history. What does the NHL market? They have tried players, but that is pretty much an Alex Ovechkin/Sidney Crosby phenomenon, at least in trying to expand to the casual fan. The logo? Can’t compete with the NFL in that regard. History? Going to be hard doing that when you didn’t grow up in the game in the same fashion baseball fans generally do and develop a decades-long attachment to the sport from childhood.

I think Proteau is on to something in arguing against the competing divison/conference influences on seeding and the unbalanced schedule. The underlying argument – that divisions don’t matter, except to the team finishing first – is one I agree with. But I think there is a way to fix the problems in that arrangement and increase the volume of rivalries without abandoning divisions.

Make divisional play matter. Go back to four divisions. I would go all the way back to the Patrick – Adams – Norris – Smythe days, but those are details. No, go back to divisions and make them matter by establishing that the top four teams in each division make the playoffs (a nod to the past). Not only that, but you have to make it out of your divisional round (rounds one and two) to play for a conference title.

For the regular season, it makes divisional games matter more in that you are fighting for a top-four seed in that group, and that group alone (not for a top-eight spot against teams of other divisions as is the case now). Add to that the narrower geographic focus of a division-based organization and division-based means for making the playoffs, and rivalries could intensify, built around the natural ones that existed before the modernization began (some of which are preserved in the current alignment, like Detroit/Chicago or Boston/Montreal).  I think proximity matters here, and competition within smaller geographic areas creates more heat between fan bases than competition between conference rivals greater distances apart.

A 30-team league presents problems in balancing regular season schedules, and there really is no good way to solve this problem if you employ four divisions. In August 2009 we proposed a two-conference/four-division format with 32 teams (taking some liberties in moving/adding teams), but the point is creating a structure that intensifies rivalries in a context that means something to all of the teams involved. You are one of eight teams (or seven, for two divisions in a non-expansion scenario) fighting for four spots, period. No complicating factors such as whether a team finishes first and gets a guaranteed extra-divisional seed or relying on a weaker division to capture a playoff spot as a five-seed in the division (but with more points than the four seed in another weaker division). This is the pool in which you swim – top-four and you’re in, bottom-four and you’re out. Period.

Will that mean that a weak team wins a division, or worse, that a very good five-seed in a division will miss the playoffs while a weak four seed in another makes it? Yes. But the NFL endured sub-.500 team making the playoffs, and the world of sports was not reduced to rubble (they even won a playoff game against the defending league champion). Moreover, the nature of salary caps and free agency could very well mean that to the extent there are weak teams from time to time within a division making the playoffs at the expense of better teams in other divisions, this will be spread around the divisions over time. It is harder to keep teams together, and if there are five or six strong teams in, say, a reconstituted Patrick Division this year, chances are that in four or five years it might be the Smythe that has those strong teams.

But the key here is the playoffs. That is where the rivalries get amped up a few notches. And that is why it is important as part of this exercise that: a) four teams in each division make the playoffs, and b) that they play one another in the first two rounds. If familiarity breeds contempt (if not outright hatred), two playoff rounds against teams you’ve already had a long and presumably intense regular season series with will make players’ blood boil (not to mention their fans). That kind of intensity is something that could be of considerable interest to the fan that has not followed hockey especially closely.

I think the NHL lost something when it discarded a piece of its history in abandoning named divisions for points on a compass. I think it lost more when it abandoned divisions as a basis for competition in favor of conference seeding. They tried to get something from both worlds -- divisions and conferences -- and succeeded in creating the mush that Proteau (to his credit) would like to see remedied.  To me it has had the effect of sanitizing the sport – the rivalries don’t seem quite as intense. Six games with a division rival don’t carry with it quite the same intensity when you can make the playoffs at the expense of teams in other divisions (as is the case in conference-based seeding) than is the case when you are fighting with that team for one of four spots in the division in which you play. And, you have the possibility of having to face that team again (and one of two others) in the first two rounds of the playoffs just to get to a conference final.

Conference-based play has the charm of symmetry – 15 teams in each conference, and you can build a schedule around that which can accommodate the need to balance the number of intra-conference games you play, promoting a home-road balance in extra-division games, and still arriving at a total of 82. It would be harder to do this with the asymmetry of two divisions of eight and two of seven teams. But that problem is offset in large part by focusing on the fact that it is the intra-divisional results that matter for seeding. With a 30-team, four division arrangement you could start with the premise that each team plays a fixed number (six?, eight?) of intra-divisional games and build a schedule out from that to accommodate concerns over whether you wanted a team to face all 29 other teams in a season or if you wanted to designate another division for home-and-home games, etc.  In that respect the schedule might not look much different than what we have now, but the consequences of it change since it is the intra-divisional results that take on more meaning.

I was not much of a fan of the actions the league took in the 1990’s to abolish one connection with their history and to construct a divisional structure that mattered only in the context of the conference in which they competed. Completing that journey – abolishing divisions altogether – just does not seem to me to be the direction in which the NHL should go. I think it needs to establish an identity built around the restoration of the sort of rivalries that gave the league much of its personalities in the days of the Patrick, Adams, Norris, and Smythe divisions. I think it’s time to make divisions matter again for the players and for fans.


Dave Nichols said...

Your proposal is solid, but in fairness it should include the addition of two more teams, for 32 total, so there are equal number of teams in each division.

part of the big imbalance problem in baseball is that there's one divison with six teams and one with just four, and obviously it's unfair to the division with six that they have to compete with two more teams than in another division.

but yeah, four eight-team divisions where you had to play out of your division to make the conference championship would be a return to awesome.

The Peerless said...

The scheme I sent up in August 2009 did have 32 teams in it, but at the moment the likelihood of adding two franchises seems almost too small to measure

Goat said...

I'm sorry I didn't read this earlier today. I absolutely love this notion.

Divisional games would be as huge as they are in football. A byproduct of which would be a vastly-heightened sense of interest in the majority of said games, as opposed to the current recurring bouts of Tuesdaynightagainstthepanthersitis that seems to infect the region.

Diane said...

As Dave says, the addition of 2 teams for 32 teams total would help in the balance. 8 team divisions would help in that regard. But can hockey afford to expand at the present time? But, if the answer's ever yes to that question, I'd say, 2 divisions, 2 conferences and do the playoffs from there.

Anonymous said...

I'm totally with your idea of going back to the divisional play aspect of creating rivalries and criteria for qualification for the playoffs. However, I balk at the idea of a 32 team NHL. The league is already over-expanded with corresponding dilution of talent. If anything, contraction is in order to a more realistic 28 team league with 7 teams in each of the 4 divisions. Let the purge begin.