Two years ago, before the start of the last NCAA Tournament, we presented a system that could help narrow down the bracket, using advanced stats from recent seasons.
It worked — barely.
Virginia, one of six teams that passed all of our statistical tests, did in fact win the national championship. All it took was a last-second shot to force overtime in the Elite Eight, three last-second free throws in the semifinal and another late 3-pointer that forced OT in the title game.
Never a doubt.
We could quit while we're ahead, but where's the fun in that? Instead, it's time to bring back the same system for another attempt to narrow the field.
To review: This method uses six stats in which every champion since 2008 has ranked in the top half of Division I. All these stats are available at barttorvik.com — and we'll use that information to pare down this year's contenders.
Stat No. 1: Overall Ranking.
T-Rank is the main measure of overall team strength at barttorvik.com. Every champ since 2008 was in the top 23 in this ranking, based on the team's performance prior to the NCAAs.
That gives us the following initial list of title contenders: Gonzaga, Houston, Baylor, Illinois, Michigan, Iowa, Ohio State, Alabama, Florida State, San Diego State, Connecticut, Wisconsin, Loyola Chicago, Villanova, Virginia, Tennessee, Texas, Texas Tech, Southern California, West Virginia, Colorado, Creighton and Arkansas.
Stat No. 2: Adjusted Offensive Efficiency.
A stat based on points per 100 possessions, adjusted to account for opponent strength. The last 12 national champs all ranked in the top 55.
That means two teams are eliminated from this year's list: Loyola Chicago (ranked 65th in this stat) and Tennessee (56th).
Stat No. 3: Adjusted Defensive Efficiency.
Each national champ in our sample was in the top 40, so that's the cutoff point. That takes out several teams, including a No. 1 seed.
Baylor (47th), Iowa (61st), Ohio State (78th), Villanova (72nd), West Virginia (60th) and Creighton (42nd) all fall out of consideration.
Stat No. 4: Defensive eFG%.
Effective field goal percentage (eFG%) is a measure of field goal percentage in which 3-pointers are given extra weight. Each national champion since 2008 entered the tournament in the top 102 in defensive eFG%.
This stat actually doesn't change anything this year. Every team that falls short of that standard was already removed because of stat No. 2.
Stat No. 5: TO%.
Turnover percentage measures how often a team turns the ball over, while accounting for the number of possessions. You don't have to be great in this category to win it all, but each of the past 12 champions has at least ranked in the top 125.
So we — nervously, in some cases — say goodbye to Illinois (129th), Alabama (157th), Florida State (231st) and Texas (239th).
USC (124th) barely stays in.
Stat No. 6: Defensive Free Throw Rate.
A measure of how often a team puts its opponents on the line. Every champ since 2008 was in the top 121.
That knocks out Houston (333rd), San Diego State (198th), Connecticut (292nd) and Texas Tech (284th).
It's also worth pointing out that Baylor (175th) and Illinois (152nd) would have fallen short of this standard, if they hadn't already been dropped. Those two No. 1 seeds certainly have a lot of strengths, but there are some areas where they could be vulnerable.
Nobody needs to be told that this tournament can produce strange results, and in this challenging season, the advanced stats may be a bit less reliable than usual — but the seven teams that made it through this process without failing any of the tests are: Gonzaga, Michigan, Wisconsin, Virginia, USC, Colorado and Arkansas.
Michigan could be limited by Isaiah Livers' injury, and Virginia has been dealing with COVID-19 issues. Maybe this really is Gonzaga's year.
Especially if fortune smiles on the Bulldogs the way it finally did for those Virginia Cavaliers a couple years ago.
Follow Noah Trister at www.Twitter.com/noahtrister
More AP college basketball: https://apnews.com/Collegebasketball and https://twitter.com/AP_Top25