It's been apparent for weeks now that Northwestern isn't going to the NCAA tournament, but like any die-hard college basketball fan, I'm still interested in seeing how this year's NCAA field shakes out. There are battles between the nation's elite for #1 seeds, late charges by schools outside of the bubble, collapses by schools that appeared to be NCAA locks, do or die conference championship games in the one-bid leagues, it's great theater.
When the dust settles on Selection Sunday and the bracket is announced, there is always a bit of controversy. After all, someone has to be the last team out, and that team's fan base will complain. But the complaints won't stop there; we'll inevitably hear about certain teams being over-seeded, others being under-seeded, the committee being biased towards a conference, the committee being biased against a conference, and on and on. We've all seen it before, and we'll see it again. The NCAA tournament is a wildly popular sporting event; therefore everyone and their mother has an opinion on it. Nothing wrong with some good debate.
But this year, the complaining has started early. We're a month away from Selection Sunday, and John Gasaway of Basketball Prospectus is already taking up the lance against the NCAA selection committee, in a multi-part series entitled "Just give me the damn field". His conclusions and my response after the jump:
In the first part of the series, Gasaway argues that the committee currently does a good job in selecting which teams get at-large bids. What he doesn't like is how those teams are seeded:
But once the field of 68 teams has been selected, an entirely different process needs to take place, one based on fairness. Currently teams are seeded according to the same criteria used for selection: how many games did Team X win, against whom, and where. We’ve known for some time now that this method is not the best way to predict how well a team will do in the future. What we owe to every team is a bracket based on the best knowledge we have. Right now that’s not happening.
Let me say that I agree with some of what Gasaway says here. Taking margin of victory into account is certainly a better method for predicting a team's future performance than simply looking at wins and losses. Ken Pomeroy's rankings and Jeff Sagarin's predictor rankings (both are which are trying to predict the results of future games) are heavily based on average margin of victory.
That being said, I have several problems with Gasaway's argument. First of all, I find it distinctly unfair to use one method for determining who gets into the field, then turn around and use a completely different method for seeding the teams. Think about it: in order to determine who gets the 37 at-large bids, the committee needs to rank the potential at-large teams from 1 through 37. But once that process is done, the committee should blow up those results and use a new method to determine the seeds? What kind of sense does that make? Whatever method the committee chooses to use, they should stick with that same method throughout the entire process.
But more importantly, I don't think there is anything wrong with how teams are currently seeded. I don't find it the slightest bit "unfair" that the committee might seed the bracket such that, say, a #6 seed could end being a favorite in Las Vegas over a #3 seed in a potential second round game. The current process rewards teams for quality wins and punishes them for bad losses, and if it happens that a school gets a worse seed than another school it would be favored over, too bad, that school should have either won more games or scheduled tougher opposition. The problem with seeding based solely on average margin of victory is that such a method doesn't put nearly enough emphasis on actually winning games.
To see a huge flaw in Gasaway's argument, here's a thought experiment: let's assume two teams (Team A and Team B), by a remarkable coincidence, played the exact same 30 game schedule during the season, and play each other twice. Team A wins each game against Team B by 3 points, and wins its other 28 games by an average of 5 points per game, to finish the season 30-0. Team B goes 25-3 in its other 28 games, but posts an average margin of victory of 20 points, even with 3 losses, and finishes the season 25-5.
Under Gasaway's method, despite the difference in won-loss record favoring Team A, he would propose giving a better seed to Team B. After all, using any margin of victory based system, Team B would be favored over Team A, and according to Gasaway, "fairness" requires Team B getting the better seed, despite the fact that Team A won more games against the exact same schedule and beat Team A twice. Does that sound fair to you? Me neither.
I understand that many people find it a bit frustrating that the NCAA uses, in part, the outdated and overly simplistic RPI computer ranking system to seed the field. After all, if you're trying to determine what's a quality win and what's a bad loss, KenPom or Sagarin will tell you a lot more than RPI. But the committee's primary emphasis is, and should be, wins and losses, not margin of victory style points. Some people currently criticize the college basketball regular season for being irrelevant compared to the tournament, but can you imagine how ridiculous it would be if winning games became secondary to average margin of victory? You'd see teams blatantly running up the score and leaving their starters in the entire game, long past the point of the game being decided. After all, cutting the deficit from 35 to 20 in the final few minutes of a game could be the difference in what seed you get come March. No thank you.
Now while I disagree with what Gasaway wrote in Part One of his series, he doesn't have anything totally ridiculous in there; we just have different opinions on how the committee should put the bracket together. That's fine. But in parts two and three, he becomes arrogant and downright off the wall.
Part two, entitled "Chamber of Secrets", discusses a Twitter debate between The Daily's Dan Wolken and NCAA vice president Greg Shaheen over the selection process. Gasaway follows up that with this gem:
Dan and I share a dream: to be in the room with the selection committee while they’re doing what they do. Dan wants to be there so he can report on the process and make it transparent. I want to be there so that once the committee has selected 68 teams and is taking a coffee break I can seed the field correctly while they’re all out in the hallway.
Gasaway, calm down bro. Ok, so you've got some strong opinions about how the seeding process should change. Good for you. That being said, you still haven't written anything remotely resembling a strict set of criteria for how this process should go. It's not enough to simply call for a general change to seeding based on projected future performance, you have to actually detail how it's going to go. Condescendingly implying that you could come up with a better bracket during a brief coffee break is, quite frankly, ludicrous. Tone it down a few notches please, you're coming off like an angry, delusional sports radio caller who thinks he can do a better job than his favorite team's head coach.
Maybe there are future articles in this series detailing how you want the process to change, I don't know. I just hope such articles use better logic than Part 3, in which Gasaway argues the NCAA merits of George Mason, citing data that the committee will mostly (and justifiably) ignore. He says George Mason should get in because:
1. They are 22-5 and haven't lost since January 8th
2. They have a better efficiency margin in conference games than Butler did in conference games last year (Butler plays in the Horizon League, GMU is in the Colonial Athletic Association)
3. CAA teams have won 6 games in the past 5 NCAA tournaments, despite never getting better than a #9 seed.
Point 1 certainly works in their favor, and although it fails to account for strength of schedule, at least they've been winning. Point 2, while impressive, is irrelevant to the committee, especially since it's a comparison between teams in two different conferences and doesn't mention the relative strength of said conferences. And as for Point 3, Gasaway should be embarrassed to use such a blatantly cherry picked stat, since four of those six wins were by George Mason five years ago (the very definition of an outlier). If he thinks what happened in the NCAA tournament five years ago should be the slightest bit relevant towards who gets into the NCAA tournament this year, then he's certifiably nuts. But remember, he could seed the whole field while the committee takes a coffee break. Please.
The article goes on to make a bizarre comparison to the 2008 collapse of banks in Ireland, the general point being that if people saw how truly incompetent the committee was, fans would be outraged and lose all faith in the system. I could be wrong, but I don't think there's a massive group of angry fans out there ready to attack the selection committee; after all, bracketologists like Andy Glockner and Jerry Palm have gotten so good at predicting the field that fans already pretty much know how the process works. I do know that if people found out the committee was considering how well a conference performed in the NCAA tournament 5 years ago, there would be outrage.
So, Mr. Gasaway, if you somehow end up reading this, I'd ask you to elaborate on exactly how you'd seed teams should the committee give you the damn field, and urge you to be a lot less condescending towards the NCAA. I enjoy your writing and, like you, I want to see an increase in the use of advanced basketball stats by both the mainstream media and the NCAA selection committee. But if you don't use a little diplomacy, and instead keep trying to ram your opinions down people's throats, you're not going to get anywhere. Explain in detail why things need to change, without the implication that everyone else is an idiot, and maybe change will follow.