Selecting NCAA tournament field a lot harder than it looks; much more
Inside the media mock NCAA tournament selection room
Picking a bracket and eliminating biases with definite set of rules
Rise of St. John's guard Dwight Hardy, Kalin Lucas' comeback, AP ballot
How important is the eye test? The answer is in the eye of the beholder.
This question came up last week during my two-day visit to Indianapolis, where the NCAA was holding its annual mock tournament selection seminar for members of the media. The main purpose was to educate us on the process the committee goes through to put together the bracket, but the more interesting moments came when we revealed our own philosophies and biases. For all the data built into these decisions, there's no way to get around the subjective nature of this whole exercise.
Thus did we digress at one point into a discussion on the so-called "eye test." If you're comparing teams, at what point should you throw all those RPI numbers aside and simply ask yourself, "I've seen both of these teams play. Who do I think is better?" ESPN's Joe Lunardi made an eloquent argument against it. "Let's say you look at a finely sculpted baseball player who's hitting .220," Lunardi said. "Then you see John Kruk, who doesn't look like he belongs anywhere on the field, and he's hitting .320. Who would you rather have on your team?" In other words, you can have your own opinions based on what you've seen -- and the committee members watch a ton of games -- but at the end of the day a team's results should dictate whether or not it gets a bid.
I basically agree with Lunardi's point. For example, in the weeks leading up to my visit to Indy, I had been ranking Texas first on my AP ballot. Yet as we were seeding the field, I argued that Pittsburgh should be ahead of the Longhorns. Why? Because Texas had one more loss (including a bad one at USC) and Pitt beat them on a neutral court in November. I ranked Texas ahead on my AP ballot because the Longhorns performed better on my eye test. But based on actual results, there was no way to justify seeding them in the tournament ahead of Pitt.
Consider also the role of strength of schedule. In last week's Mailbag, I listed the SOS rankings for the last 17 NCAA champions to demonstrate that a team's strength of schedule is not a good indicator of performance in the tournament. However, in the committee room you better believe it matters a lot -- as it should. A team's SOS ranking is a good tool to evaluate the type of season it has had. It is also important to me where a team's nonconference SOS ranks, especially if that team is in a power conference. I don't like power conference teams who duck games simply because they can. That's my own bias. At any rate, I think the committee's job is to assemble the bracket based on what has happened during the season, not on their predictions of what will happen during the tournament.
I'll take this philosophy one step further. After the first couple of rounds, there is always lots of discussion about whether the committee did a good job based on those results. This year's chairman, Ohio State athletic director Gene Smith, told us before we started that he felt good as a committee member in 2008 when all four No. 1 seeds reached the Final Four. I disagree with this line of thinking. Once that ball goes in the air on the first day, all this stuff goes out the window. Did Kansas' loss to Northern Iowa last year mean that the Jayhawks didn't deserve a No. 1 seed? Of course not. It only means Kansas was playing in the NCAA tournament, where upsets happen.
By the way, that also goes for the inevitable conference arguments. Last year a bunch of Big East teams lost during the first week, so people argued that meant the league was overrated. That's ridiculous. Conferences don't play conferences in the NCAA tournament. We don't throw away five months worth of games because some wacky things happen in the first two rounds.
Anyway, you can see the bracket we came up with here, but keep a few things in mind. First of all, we did this bracket in about a day-and-a-half of work. The committee does it over five days. Believe me, we did not burrow into these teams the way they will.
Second, a lot of what we decided was based on mock conference tournament results the NCAA provided us to simulate what the actual committee has to deal with. In this scenario, New Mexico State "beat" Utah State in the WAC final, so both teams are in our bracket. Virginia Tech "won" the ACC tournament, so they were a 9 seed. Texas was still undefeated in the Big 12 and "won" the conference tournament, so the Longhorns ended up as our No. 1 overall seed. And Michigan State "lost" its first game of the Big Ten tournament, failing to reach even the quarterfinals, so the Spartans didn't make it. In real life, the Spartans beat Illinois over the weekend and probably will not fall on their face like that, so I think they're in the field right now with room to spare.
Here are a few more nuggets from the seminar:
Because of expansion, the committee is going to have the added task this year of setting up an additional round, which has come to be known as the "First Four." When they start putting teams into the bracket (which doesn't happen until Sunday afternoon, by the way), the committee has the option of moving teams up or down one seed line to accommodate its principles. However, they will do everything they can to leave the last four at-large teams in those early games. The games will be played in doubleheaders on Tuesday and Wednesday, and the intent is to split those doubleheaders so each night there is one game between automatic qualifiers and another between at-large teams. The Tuesday games will feed into a Thursday first-round games, and the Wednesday games will feed into Friday.
One of the principles in bracketing asserts that two teams from the same conference cannot meet before a regional final. An exception can be made if there are more than eight teams from a conference, but that has never happened before. It is definitely going to happen this year. The only question regarding the Big East is whether it will get 10 or 11 teams into the field. We had 10, but Marquette was our first team out.
There will never be clarity over just how much emphasis the committee places on how a team performs down the stretch as opposed to early in the season. Two years ago, the committee removed the record in the last 12 games as a piece of criteria because they thought it was misleading. The NCAA wants the games in November to count just as much as the ones in February and March. Again, I disagree. Alabama is a great example. The Crimson Tide played poorly in November, but then Anthony Grant suspended his best player for a few games, and after that player returned the team took off. Now they've won 13 out of 15, including at home over Kentucky and on the road at Tennessee. To me, that's an NCAA tournament team, but I was obviously in the minority because the Tide were left out. At any rate, if the committee's mission is to select the "best" 37 at-large teams, then it only makes sense to give extra emphasis to the way the team is playing late in the season.
One team that did not look as good upon closer examination was Memphis. (And this was before the Tigers lost at Rice over the weekend.) When you look at the team's nitty-gritty sheet, you see they have four wins over teams ranked in the top 50 of the RPI. Look closer, and you see those four wins were comprised of sweeps over UAB and Southern Miss. When we further looked at the respective resumes of UAB and Southern Miss, we were even less impressed. We were then told that Memphis "lost" to UAB in the Conference USA tournament final. The fact that we put Memphis in anyway -- as a 9 seed, no less -- shows you just how weak the bubble is.
I was generally not in favor of expanding the tournament, so I was relieved last spring when the NCAA announced it was going to 68 teams instead of the 96 we had all been expecting. The NCAA's vice president of everything, Greg Shaheen, laughed as he recalled that everyone had assumed the 96-team bracket was a "done deal," and then when it didn't happened they assumed again it would go to that number within a couple of years. I didn't get the sense from Shaheen that he thinks that is going to happen -- and let's hope not. We had to really hold our noses as we put those last few teams into the field. I can't imagine the foul stench that would have enveloped the room if we tried to put in 28 more.
If they gave out postseason awards based on what happens February, St. John's guard Dwight Hardy would be an All-America. Over the last seven games Hardy has averaged 24.4 points and has twice cracked the 30-point mark. He sank the big game-winner over Pitt on Saturday. And yes, I saw that he stepped out of bounds on the play, but so what? Stuff happens.
Speaking of Pitt, that was pretty amazing that Ashton Gibbs could come back after a three-game absence and get his career high in scoring. Still, one Big East assistant coach told me he's skeptical of Pitt's chances to win an NCAA championship. "I worry about them fouling," he said. "The games are called a lot tighter during the tournament than they are in our league."
If you're looking for vulnerabilities among the top contenders (as I always do), check out Kansas' free-throw shooting. The Jayhawks are ranked 10th in the Big 12 at 66.8 percent. Even while they were whupping up on Colorado, they still made just 12 of their 21 attempts.
It's amazing how Michigan State's Kalin Lucas looks like a completely different player than he did even a month ago. He has that old explosiveness and confidence back. I've been saying all season that people have underestimated the difficulty of coming back from a ruptured Achilles tendon. And Lucas suffered his injury in the second round of the NCAA tournament, which means it hasn't even been a year. It's no coincidence that the Spartans have played their way back into the NCAA tournament just as Lucas has gotten back to his old self. Couldn't happen to a nicer kid, by the way.