The Duke Blue Devils ran through the ACC Tournament, winning four games in four days. The last three came against teams that will be in the NCAA Tournament, and seeded very highly. In doing so, I now believe they have moved into the final #1 seed, surpassing North Carolina and holding off Kentucky (who still plays in the SEC final today) and Arizona.
But what about the regular season? Shouldn’t it matter? Of course, and it does. Duke, though, added three great wins, wins better than most tournament teams have cumulatively over the course of three months. There will be cries of Duke bias if they get the top seed. It comes down to simple comparison.
Yes, North Carolina won the ACC. But this talking point about Duke finishing 5th is a non-factor. League schedules were imbalanced. Duke was one game from 2nd in the standings, and were the only one of the top five teams to play the bottom five teams (who will not be in consideration for an at-large bid) only once each in the regular season. They played North Carolina head-to-head twice in the rivalry games (more on that in a minute), something most other contenders didn’t have to do.
The Selection Committee does not care about titles, per se, without looking at the underlying details. And it is comparing those details that I think swings it to Duke.
Here is a breakdown of the potential candidates for a top seed, with a general look at how they did against teams likely to be seeded in the top 6, seeded 7 to 12, against all others in the RPI Top 150, and those outside the Top 150.
You can see why Villanova should be the #1 overall seed. Gonzaga may not play as many tough games in the West Coast Conference, but they beat the Big 12 tourney champ, the Pac-12 tourney champ, and Florida. Kansas has the two losses to non-tourney teams but has otherwise done enough.
Which leaves the last spot. Duke has the same number of losses against the top teams, but the most wins against potential top seeds of anyone after the run through the ACC tournament. They do have those three losses to other teams that could be in the at-large pool. Even with those, the win percentage against tournament teams is still right in line with Kentucky and North Carolina–they played more games against tournament teams.
Arizona, meanwhile, I am excluding here. They have the same number of wins against likely tourney teams (6) all season, that Duke has against top 4 seeds in the last six weeks.
So let’s get more granular.
Duke beat North Carolina twice. Going 2-1 wouldn’t swing it if it wasn’t otherwise close, but since Duke has more top wins and has played more games against tournament teams (20 to 16), I do think it should factor here. North Carolina also has more losses outside of tournament teams (Georgia Tech and Indiana, versus NC State, so the difference in record is based on schedule differences.
That brings us to Kentucky. The Wildcats also beat North Carolina head-to-head in December (which is why I think North Carolina is out here as well, 1-3 against the other contenders). Kentucky, though, is 2-4 against the top teams, including home losses to both Kansas and UCLA. Either of those wins likely would have sealed it for the Wildcats with the way the rest of the season has played out. Add in the loss to Louisville (who Duke just beat), and I’m going with Duke (13-7 against tourney teams) versus Kentucky (8-4), as both have one loss outside that group.
After sleeping on it, I am going with Duke as the #1 seed in Memphis, opposite Kentucky as the #2 seed. North Carolina will fall to the #2 seed in the East most likely, opposite Villanova.