Selection Sunday is a month away. On March 17, the choices of the NCAA tournament selection committee will be revealed.
Four years ago, I attended the NCAA's two-day Mock Bracket Selection seminar in Indianapolis and learned how the committee does its job. After studying the Rating Percentage Index for 20 years and interviewing selection committee chairs and members, I thought I knew the system before I went. But it was an eye-opener. Enlightening in its dismissal of misconceptions. Numbing in its demand of diligence. Exhausting in its comprehensiveness.
The following is an NCAA tournament selection primer. It won't attempt to summarize all I learned in Indy but will touch on some common themes that seem to arise every March and commonly lead to misinformation. Along the way, I'll try to debunk some widely held beliefs that are outdated, inflated in importance or simply never have been true at all.
1. The RPI is a starting line, not a destination
Don't think a 10- or 20-place difference between RPI ranks means much -- it doesn't. If your school is known to be in a tenuous spot for NCAA consideration, its RPI rank won't help it much. And using it before March to judge likelihoods is pointless. Even once we get there, it's more of an initial barometer than a clear indicator.
Once the committee convenes on the Wednesday before the brackets are announced, you can use the RPI to judge precedent. If your team is in the teens or better, you're a lock. If your team is in the 80s or worse, you can pretty much forget it. But schools with as a high a rank as 21 have been spurned and with as low a rank as 74 have been granted at-large bids. The mid-40s is usually about the break-even point. But that can waver, wholly depending upon the resumes of teams involved.
Use the RPI as a preliminary gauge. Yes, the 60s are dicey and the 30s are encouraging. But then move on to more specific data. The RPI is usually a much more accurate predictor of seeding than of who will receive at-large bids.
2. Impressive road wins are more important than ugly home losses
Weeding the pretenders from the contenders involves an exhausting exercise in which the 10 committee members are asked repeatedly to “list and rank” groups of at-large candidates. Over and over, this happens: 25 teams pop up on your computer screen. You're asked to list your favorite eight and hit enter. The computer tabulates the aggregate vote and then up pops a list of eight teams. You're asked to rank them best to worst.
After consulting the resumes of said candidates dozens of times, your brain morphs into Hostess crème filling and you tend to make certain decisions. The resumes that pop out are the ones in which candidates succeeded away from the comfort of their home courts. After a while, you latch onto these scraps of meaningful data like gold nuggets. You are willing to forgive twice as many transgressions at home if a team shows it's tough on the road against a cluster of top-100s or a couple of top-25s.
We'll get to Temple's specific case in a later post. But don't believe that just because the Owls have three ugly losses to triple-digit-RPI teams (including a particularly smelly one Thursday at home to Duquesne) that they are lost at sea. Because they have a non-con road scalp at Syracuse, that will stand out like a beacon. Another one like it would be nice, of course. But consider that Southern California had six such losses to teams with 100-plus RPI teams in 2011 and got in, as did Iona last year with five such losses. A couple of great road wins can compensate for a Superfund site full of toxic losses.
3. Large numbers of wins mean nothing if the schedule stinks
We won't have such a Philadelphia-area case this year. But last year's Drexel team was a perfect example. Because the Dragons won 27 games, many assumed they were something close to a lock.
When I made an appearance on Daily News Live a few days before the brackets' release, Michael Barkann asked me during a Quick Six: “One to 10, Drexel's chances to make the NCAA tournament?” I said “two” and stunned him. But I had figures to back up that prediction later in the show. The CAA was in a down year and Drexel simply had not beaten anyone. The committee has repeatedly discarded such resumes.
The same thing happened to Penn State in 2009. The Lions had nothing to show in a flimsy non-conference schedule, and their only top-25 wins were at home. They were above .500 in the Big Ten (10-8), usually a bellwether for a bid. They had 22 wins and would make it 27 after a big NIT run. But that didn't matter. No road scalps and looking like you're not trying in November and December is a big problem with most committees.
4. Conferences have very little influence on selections
You'll have to repeat this to yourself because the chatter to the contrary is so incessant. At-large candidates are compared against each other based on their overall resumes. What conference they play in has no bearing.
Now, do candidates who play in highly competitive leagues get more chances to beat top teams? Sure. But leagues don't get more of their teams into the NCAAs just because they have strong seasons. Each at-large candidate is an island. In fact, the committee is told to disregard any correlation between at-large candidates and their conferences. There are no conference quotas.
So, just because the Big East is having a strong season, that won't help Villanova if it ends up smack on the bubble. Same for Temple, La Salle and Saint Joseph's just because the Atlantic 10 is on an uptick. It doesn't matter where you finish in your league standings. It only matters who you beat and where you beat them -- in all of your games.
5. As long as everyone's healthy, how you're playing in March gets no more weight than how you did in November
How you finished actually was a factor until several years ago. The committee's computers were fed a category headed “Last 12” that showed each at-large candidate's record in its final 12 games.
But for two good reasons, that stat was removed in 2008: It could be skewed arbitrarily depending upon each team's late schedule strength. And it was deemed unfair to emphasize February and March at the expense of a total four-month body of work.
So, while you never want to finish badly, it means no more now than starting badly. A game is a game is a game.
Hope this helps. We'll delve into specific cases of the local teams with regular updates on their resumes in the coming weeks. I'll refer to these guidelines and raise some others as the need arises.