The Broken Section Seeding Process

18 Feb

At the end of the regular season in high school hockey, a factor that has little to do with on-ice results weighs on teams’ fates. This week, the coaches in each section meet in smoke-filled rooms to set the seeding for their respective section tournaments. (In most cases, this is now done electronically; in 7AA they literally do meet in a smoke-filled room at Tobies restaurant in Hinckley, a tradition sadly interrupted this season, when Duluth Marshall had the nerve to schedule a game for itself on the sacred Wednesday night.) The coaches cast preliminary seeding votes, view the results, debate and make cases for themselves if they think the first vote hasn’t been fair, and then vote a final time to decide who gets which seed. To avoid sabotage, coaches do not rank their own teams, and the highest and lowest vote for each team is thrown out. Even so, the process is messy, controversial, and is responsible for an average of 37 aneurysms per year among users on the USHSHO Forums.

The leading controversy this season, as it so often does, comes out of 7AA, where Cloquet leapfrogged Grand Rapids to claim the 3-seed. The storyline in 7AA all season long was that one of its three big contenders (Elk River, Duluth East, Grand Rapids) would claim the top seed, and thereby avoid having to play two games against top-flight competition to make the State Tournament. Cloquet, however, threw a wrench in things. I was at the Rapids-Cloquet game this past Tuesday, and it was as stunning a high school game as I’ve seen in a while, as the Lumberjacks jumped all over Rapids and forechecked them to death in a convincing 4-1 win. It was a coming-out party for a young Cloquet team, and an exclamation point on a sudden, ugly late-season turn that has an incredibly talented Grand Rapids team struggling to find answers.

Grand Rapids, interestingly enough, doesn’t lose much of anything with this arrangement: there’s not much distinction between being the 3-seed or the 4-seed in 7AA this season. With losses to Elk River and Duluth East, they knew they were going to have to go through the Hounds and Elks regardless. The order is just different there, and they play Andover instead of Duluth Marshall in the first round, which I’m not convinced is a drawback, either. Likewise, even if Cloquet beats Marshall a third time, they’ve earned themselves a date with Duluth East, who’s beaten them 5-0 and 5-1, respectively, this season. For Rapids and Cloquet, the actual difference is minimal.

Backers of top-seeded Elk River, on the other hand, are crying foul. Their squad’s reward for the top seed could well involve a semifinal meeting with a team that spent a fair amount of time in the top 5 in the state this season, and they’re understandably leery that the Rapids sleeping giant will awake at the wrong time. (Mixed in here, one suspects, is a fair amount of frustration over the location of the 7AA semis and final, an entirely separate issue that has also not treated Elk River well.) No one out there honestly thinks Cloquet had a better season than Rapids, but two of the factors that have long swung section seeding meetings—record in the section, and performance in the second of two meetings between teams—tipped the scales. The logic the coaches used shouldn’t have surprised anyone who’s paid attention to section seeding over the years, though that doesn’t mean it’s the right way to do things.

Over in 8AA there was controversy as well, but the reasons were entirely different. Bemidji had stumbled through the first half of their schedule, with four losses to section foes, before turning it on down the stretch. The Lumberjacks beat St. Michael-Albertville, who had been the presumptive 2-seed, in their only meeting, and also avenged an earlier loss to Roseau. In a section that was a total seeding mess, one might have thought that Bemidji’s late-season run and avoidance of truly bad results would boost them up to the 3-seed, perhaps even the 2-seed; that would certainly be true if we applied the logic the 7AA coaches seemed to use. Instead, the Lumberjacks were consigned to the 5-seed. That result that does have its own internal logic, as they lost twice to 4-seed Brainerd, but seems awfully harsh for a team that is playing well right now, is 3-3 against the teams immediately above it, and that most observers would agree is one of the two or three most talented in the section.

The coaches’ vote is a meat-making process that will, inevitably, lead to butthurt. I’m not terribly sympathetic to those who are upset, a position I will maintain even if my alma mater someday seems to draw the short stick in one of these controversies. (The fact that this hasn’t happened in recent memory has led to some laughable efforts to endow Mike Randolph with magical powers of ballot box stuffing or persuasion.) Teams just need to find ways to win, period. But I recognize it does affect things, and I can think of two ways that could remove some of the intrigue:

Use an algorithm to do the seeds. Plenty of people have designed algorithms that can spit out dispassionate rankings of high school teams. The Minnesota State High School league has even offered up one, QRF, as a valid criterion for making seeding decisions. I enjoy looking at computerized rankings, but this one is probably the worst of the bunch, to the point where I look at other algorithm-based rankings in creating my subjective ones, but completely ignore this one. (I’m sure an effort to look at its past ability to predict playoff results would back this up.) And while I’d love it if the MSHSL were to contract out Lee Pagenkopf’s PageStat or MyHockeyRankings (or *my* hockey rankings), these aren’t really realistic options at this point in time. (Though seriously, MSHSL, you know where to find me if you want me.) Given the finicky nature of some of these algorithms, and the different and ultimately subjective weights they can give toward things like strength of schedule or margin of victory while ignoring raw head-to-head results, I lean away from this option, and instead toward another:

Mandate that all teams within a section play each other once in a game that contributes to section standings. Right now, schedules are built around conferences, which are, for the most part, useless relics left for us by football, which doesn’t even use a conference system anymore. This is dumb on many levels, and leaves us with some ugly controversies when teams don’t play all of their section opponents, or play them a different number of times. Silly flaps between programs like Eden Prairie and Holy Family, which refuse to play for political reasons, lead to a lot of guesswork, use of competing forms of logic, and grievances. There is no quality control in scheduling, and the seeding process inevitably suffers.

Under this model, teams would play a sort of league schedule where they play their section opponents once. Home venues would rotate by year. Teams could schedule additional meetings with their section rivals, if they so choose; it only makes sense in the case of longstanding rivalries, and especially for the Greater Minnesota teams that face long travel times. Those additional meetings, however, simply wouldn’t count toward section standings. Section standings could then rely on a consistent point system just like a college conference, with tiebreakers such as head-to-head record or goal differential on hand to break any ties. (The goal differential one would need to have limits, though; top teams running up scores on the bottom of the section shouldn’t determine seeds.) We’d have a simple, clear order at the end, and a playoff structure to resolve any disputes.

I don’t expect change anytime soon, but sooner or later, the clamor for transparency should become too loud to ignore. With the complete death of conference relevance, hockey has a chance to adopt an added dose of sanity. So, much as I enjoy munching on my popcorn as 7AA undergoes its annual explosion, I’ll gladly forego that for the sake of a system that makes more sense.


One Response to “The Broken Section Seeding Process”

  1. Derek Felska February 18, 2017 at 8:54 am #

    I am a High School Coach in Wisconsin. When we seed, it works like this and the coaches don’t really submit lists and they just work through a progression. 1. Overall record, 2. Head-to-Head (most recent result is given more weight), 3. Common Opponents. This usually works pretty well. If there is a tie, then we coaches vote on it. We seed as many teams as we feel is necessary; otherwise we simply draw them out. Conferences are not a relic. In most cases they usually make sense based on geography. Its a way to limit travel budgets so they’re sensible. While its nice having metro and outstate teams go and play one another its ridiculous to eradicate conferences. Duluth East may be independent, but it doesn’t have many competitive peers close by and if it was to only have schools of similar size the area it would have to cover wouldn’t save on those travel costs. The MSHSL may have tried to get rid of conferences, but its at a detriment to schools. The Elk River conundrum is very understandable. They get nothing to being the top seed. Ever consider Grand Rapids threw the game against Cloquet because they realize the 4th seed is actually better than being the #3 seed?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: