The Finer Things
by Adam Wodon/Managing Editor
In watching the Super Bowl, something struck me. And it wasn't a Terrible Towel.
Nowadays in the NFL, it seems the referees have a much harder job. Was the elbow down before the ball came out? Did the ground cause the fumble? Did the nose of the football touch the white line of the end zone? Does one knee equal two feet?
It's not that these calls didn't have to be made 40 years ago. It's that referees, and the public, didn't wring their hands so much over them. Now, with every game on TV, televised with 12 cameras and varying angles of instant replay, every bit of minutia hashed and rehased over and over again — it almost forces us to find flaws in places where we once didn't even think flaws possibly existed.
Such it is with the NCAA's ice hockey tournament selection process.
There was a time when "The Pairwise" was a strange animal, even to the large majority of hockey coaches. If it was known at all, it was only vaguely, and without a complete understanding. (article: 1997.) The online media went a long way towards educating the public, and coaches, on the intracacies of the system used by the Men's Ice Hockey Committee to select and seed the NCAA Tournament. (article: 2000.)
But has too much knowledge become a bad thing?
Ultimately, no. It's always better to know, than not to know. But the knowledge of the system has helped to illustrate its flaws, and, as a result, it's also led to changes — well-intentioned changes meant to fix things, but that, by and large, only muddle the issue. In other words, it has forced us to concentrate on tiny issues — to find flaws in places where we didn't even know to look 10 years ago.
And while it's important to get it right, sometimes you tinker so much that your nice, fine, but somewhat flawed friend morphs into Michael Jackson. Perhaps the original was good enough, warts and all.
Don't get me wrong. I'm as guilty as anyone for writing about the process and nitpicking the thing to death. But just to be fair to myself (?), I have also written plenty in recent years that the "fixes" are worse than the ailments. Or at least not worth the bother. (article: 2003.)
Recently, in a USCHO article, Boston University coach Jack Parker lashed out at the Pairwise somewhat. He noted that we have moved away from the original usage of the Pairwise criteria: something that would only be applied to make decisions when the RPI of teams were too close to call. In fact, this interpretation more or less holds up until at least 2000. But somewhere along the way, the committee started just taking the Pairwise chart as is. Parker says that the different Pairwise criteria, besides RPI, are based on too few games to be adhered to so religiously.
This may sound strange coming from someone who, at one time, specifically lobbied for the committee to do just what Parker laments. But, as it stands, I actually agree with Parker. (That makes two: I'm also on board with him on the belief that nets should be larger. He, however, has more championship rings than I do.)
But I only agree because I know what he's really saying — whether he knows it or not (wink, wink). He means, we need to use KRACH. That's right. He said it. See how I did that?
KRACH is far superior to RPI. So if you're going to baseline the selections on one formula, KRACH should be it, not RPI. And what have we been saying for the last few years now? That not only should RPI be replaced by KRACH, but that the committee should JUST use KRACH and forget the other criteria, since they are pretty dubious anyway. Or, if you really find value in criteria such as record against top teams and record "down the stretch", then apply them with some more finesse than is currently done. (See our KRACH Primer and Open Proposal.)
I've always been a major proponent of using a completely transparent, objective system. But I want to make sure that the right system is used, and that regardless of the system, the committee avails itself of some common-sense wiggle room in order to avoid grave injustices it would otherwise box itself into. (article: 2003 and article: 2005.)
Since then, there has been recognition of the flaws in RPI and the system, but these were solved not with KRACH, but with some dubious "tweaks." All of this has only heightened the urgency for KRACH. (Its advocacy can be found in a 2003 article, a 2004 proposal, and a follow up in 2005.)
All of this said, as the time approaches where we'll be following the daily minutia of the Pairwise, KRACH and RPI hot and heavily, it's always helpful to remind ourselves that we have a great sport. And once the teams are seeded and the pucks are dropped, the NCAA Tournament never ceases to amaze.
So let's not lose sight of any of that.
But it's also worth trying to get the best system possible with the least angst. And — face it all you math geeks out there, full-time or wannabes — it can also be fun, too.
In the mean time, when all is said and done this season, I can guarantee one thing: This year's NCAA Tournament will be a lot better than this year's Super Bowl.