Basketball Prospectus: Unfiltered Everything Else is Fluff.

March 23, 2011

The RVA Revolution and the RPI Test

Filed under: Uncategorized — Kyle Whelliston @ 3:36 pm

Richmond, Virginia isn’t just a place to get a 21st Century dinner on the way back from Colonial Williamsburg anymore. Its erstwhile role as seat of the Confederacy has been superseded by its new status as the capital of NCAA Tournament’s Southwest region. Virginia Commonwealth and University of Richmond have put the River City on the national college basketball map. It might also end up as a crucial battleground in the fight for college basketball’s future.

Much has been written in this space, and by fellow Prospectors elsewhere, about the university game’s odd relationship with the past, most specifically with the Ratings Percentage Index equation that was introduced to the selection process in 1981. There was no “eye test” back then (“Bette Davis Eyes”, however? On 8-track? Yes.) The formula was designed to go where the TV cameras couldn’t, to measure all teams equally. It was as applicable to any of the NCAA’s games — volleyball, soccer, lacrosse, and so on — and ended up being as soullessly borderline-functional as any other multi-sport contraption. There were 22 at-larges in a 48-team men’s basketball bracket that year. This sport’s selection process has become the most intricate and scrutinized (and leads to the most profitable event the NCAA has), and obviously deserves more detailed metrics. That shouldn’t even be a debate with an opposite side.

It took nearly three decades for the NCAA to provide a measure of transparency about their selection process (beyond 1. Regular Season 2.3. Bracket!). I have attended a mock selection exercise in Indianapolis, and I have tasted the Kool-Aid. I have found it to be tart and refreshing. There is no possible logical way that the selection committee could conspire to keep any team out or let any other team in, and no tinfoil hat conductive enough to make me believe otherwise. There are too many people in that room, too much need for consensus and compromise, too much due process, and too many team sheets.

The following is a reasonable facsimile of the official team résumé that allowed VCU into the NCAA Tournament as one of the last four at-larges, despite a title game loss to Old Dominion in the Richmond-based CAA tourney. Greg Shaheen and the other wonderful folks at the NCAA were kind enough to allow me to replicate the template for my website.

Whatever theory you have about why VCU was included in the field, this is how they got in — the one-sheet that serves as the starting point of all discussion in that room on the 15th floor of the Indianapolis Westin. Though Shaheen and others insist that the RPI is just one tool in the process, note how much of this sheet has something to do with the forumla. In the top corner, the average RPI win and loss. The level of wins’ quality is displayed left to right, in order by opponent RPI. RPI, RPI, RPI. The sheet would fall apart without it.

By this measure, VCU did have two “signature” top 25 wins — against fellow Colonial Athletic Association teams. Teams from the CAA won 57 percent of its non-conference regular season games in 2010-11 (82-60), but went 6-16 on aggregate against teams from the largest six leagues. They did finish 9-7 against the Atlantic 10, though, and 3-2 against Conference USA, their peer leagues on the East Coast.

And by this formula, citymates Richmond scored a “top 50 win,” one of their three against three losses, by beating VCU on December 11 in the Black and Blue Classic. We’ll never know if the Spiders would have been admitted as an at-large, because they won their title game, at the Atlantic 10 tourney. They beat Purdue, just like VCU did in the Tournament — the No. 12 team in the RPI. Their next best win against a Big Six power-conference team was in the third column: Seton Hall (No. 102).

The CAA’s success is strikingly similar to the four-bid Valley of 2006: a losing record against the six best conferences (12-15) but solid performance against peer leagues (6-3 vs. the Horizon). When conference play came along, the best teams split series and shared RPI power. Then, as now, some in the industry could not accept non-televised games against top conference teams as important, and trotted out vague terms like “eye test.”

But sooner or later, leagues at this level will figure this out. Theoretically, CAA and Atlantic 10 and Horizon League and Missouri Valley and Conference USA teams could ignore the top leagues altogether, beat each other all winter before league season starts, share the RPI strength, and flood the 10-13 seed lines with mid-major at-larges in March. We’d have 11 over 6 and 12 over 5 every year, and lots of $1.2 million NCAA Tournament conference win shares for everybody. I hereby call for a meeting among the heads of the five families, to build a scheduling consortium to help make this happen.

And then we’d find out if there really is an NCAA conspiracy against mid-majors. If there is, they’d go ahead and use that opportunity to ditch the RPI.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress