Get inside March Madness with SI.com's Luke Winn in the Tourney Blog, a daily journal of college basketball commentary, on-site reporting and reader-driven discussions.
3/15/2006 10:57:00 PM
A Quality 12-Pack?
Texas A&M, practicing on Wednesday night in Jacksonville.
TOURNEY TOWN, Fla. -- Here on the eve of the dance, from my vantage point high up in the stands of Jacksonville Veterans Memorial Coliseum, I had this thought while Texas A&M practiced below: If the Aggies don't beat Syracuse tomorrow night, then the No. 12 seeds are doomed. Twelve-over-five has become a "traditional" upset pick, yet out of a great crop of 12s in 2005 -- UW-Milwaukee, George Washington, New Mexico and Old Dominion -- only the Panthers pulled off the shocker, taking down Alabama 83-73.
The 12s in 2006 are not nearly as strong, and there's a good chance they could pull an 0-fer this tournament. A&M has a shot -- if its lock-down defense (one of the nation's best) can keep G-Mac and the Gang quiet. All of the others -- Utah State (vs. Washington), Kent State (vs. Pitt) and Montana (vs. Nevada) -- seem overmatched. Readers, which No. 12s do you think can play Cinderella?
This may be the Year of Efficiency at Sports Illustrated, with repeated stat-posts here and in this week's magazine, the definitive story on college basketball's version of Moneyball, written by Grant Wahl. Called the "Possession Obsession," it charts the source of points per possession (a.k.a. efficiency) to a surprising source in the 1950s, gives (much deserved) credit to blogger Ken Pomeroy for taking the stats to fans nationwide, and offers unique bracket analysis that factors in both tempo (which we aren't covering here) and efficiency of tourney teams.
If you believe in the power of the possession, and want to apply it to your picks, we break down the efficiency ratings (points per 100 possessions) of the top four seeds in each region, and the margin between their offensive and defensive scores. Note that we're using Pomeroy's adjusted numbers, which take into account level of competition, game location and give more weight to recent games. (Complete data can be found on kenpom.com.)
The squads with the biggest adjusted efficiency margins in each region -- and your Final Four teams, if you're obsessed with this stuff -- are UConn, Villanova, Texas and Kansas. And the least efficient of the top four seeds in each region are Tennessee, Boston College, Iowa and Gonzaga; not surprisingly, three of those teams showed up in our inefficient defense post from Monday.
While I'm not advising that you go overboard and make every pick based on efficiency (do you really think that the Longhorns will win it all?), and everyone suggests that you go with your gut when making tourney picks, ignore these stats at your own risk.
(Readers -- remember, you only have one day left to join the first-ever Tourney Blog Pool.)
Center Paul Miller and the Shockers have an edge over Seton Hall in the numbers game.
If you want to get hardcore with your bracket picks, two self-professed "Stanford nerds" are running advanced statistical models on every tourney game at bracketbrains.com. Their most robust (and most expensive) game-prediction system allows a user to assign weights to 10 different categories, ranging from the site's predictive power ratings to a momentum index for both teams.
Let's try it out on one first-round game that doesn't have a massive favorite: No. 7 Wichita State vs. No. 10 Seton Hall, in the Washington D.C. region. In Bracket Brains, I assigned the following weights to these categories:
HIGH Power rating (predictive)
AVERAGE Power rating overall Distance traveled to game W-L percentage, last 10 Momentum index Average margin of victory
LOW Respective tournament seeds Round of game W-L percentage W-L percentage, close games
The result the machine spit out? Wichita State by 3.3 points. It also -- in a feature reminiscent of baseballreference.com's "10 Most Similar" lists" -- indicates that the 2005 NCAA tournament game most similar to WSU-Seton Hall was No. 7 Cincinnati 76, No. 10 Iowa 64. Will the Shockers make like last season's Bearcats, and make good on their statistical advantage?