Apparently, trying to seed over half the field between 4pm EST and 5 pm EST was not the most effective way to do bracketology.
I got 66 of 68 teams – missing on UCLA and Drake. And all the media bracketologist – and the majority of those posting brackets on the Bracket Matrix did a better job of seeding teams than I did.
I was all prepared to write a long rant about how I don’t understand how the ACC was seeded. For example, lets take a look at Virginia Tech and Clemson.
If you look at the AP Poll going into Champ Week, Virginia Tech was ranked 22nd in the AP Poll and 21st in the Coaches Poll. Meanwhile Clemson was 29th in the AP rankings and 30th in the Coaches. So, while the polls are certainly not the Selection Committee, most of them AP writers and Coaches who rank the teams thought Virginia Tech was the better team last week. In the ACC tournament, VT was the 3 seed and lost to 6 seed North Carolina by 8. Meanwhile, Clemson lost the day before by 3 to 13 seed Miami FL. So if many felt VT was better than Clemson, and while both lost in the tournament, Clemson’s loss was clearly worse. So, how did Clemson end up a 7 seed and Virginia Tech end up a 10 seed. For that matter, how did the Selection Committee find 18 teams to push over the Hokies.
I was then going to go into ranting about Louisville being in over Syracuse.
So, I thought that I could use my Bradley-Terry model to prove my point. Apparently, that was a mistake. But it means I have to give the NCAA Selection Committee so much more credit. Because the ACC makes no absolutely sense. I am going to provide you the top 8 teams, and lets look at the following: the conference standings, the conf tournament, the seeds from the Selection Committee (the only ranking that matters), my bracketology seed, and a rank ordering of the teams based on my Bradley-Terry model predictions.
| TEAM | STANDINGS | TOURNEY | NCAA | LUNATIC | MODEL |
| Florida St | 2 (11-4) #15 | Lost Champ | 4 | 4 | 3 |
| Virginia | 1 (13-4) #16 | Forfeit Semis | 4 | 4 | 1 |
| Clemson | 5 (10-6) #29 | Lost 2nd Rd | 7 | 9 | 2 |
| North Carolina | 6 (10-6) NR | Lost Semis | 8 | 9 | 5 |
| Georgia Tech | 4 (11-6) NR | WON | 9 | 7 | 4 |
| Virginia Tech | 3 (9-4) #22 | Lost Quarters | 10 | 6 | 6 |
| Syracuse | 8 (9-7) NR | Lost Quarters | 11 | 11 | 8 |
| Louisville | 7 (8-5) NR | Lost 2nd Rd | OUT | 10 | 7 |
I figured that for the most part the conference standings would stand. Most of the conference tournament didn’t give me any reason to not keep those rankings. Georgia Tech won the tournament, but was unranked. UNC and Clemson had the same conference record, and UNC made it to the semis. Couldn’t move Syracuse over Louisville because the Orange beat NC State and the Cardinals lost to Duke.
The committee saw much differently. I focused on Clemson’s losses instead of their impressive early victories against Alabama, Purdue, and Florida State. I focused on Virginia Tech’s record and AP ranking, but they don’t have quite the victories that Clemson has (though Virginia, Villanova and Clemson isn’t bad).
The Bradley-Terry model pushed Clemson over Florida State and Virginia Tech fell behind both Georgia Tech and UNC. Its clearly a mess. Games getting cancelled made it hard to compare a 9-4 team to a 10-6 team. Conference schedules aren’t balanced to begin with – but add into games getting cancelled including ACC tournament games, and it gets crazy complicated.
Guess this was my rant saying that I think my logic for my rankings made sense, but the analytics that favor pushing Virginia Tech all the way down to below UNC and GT also suggested Louisville was better than Syracuse. But if you look at those two teams closer, it is still hard to distinguish – especially since both the games between Louisville and Syracuse got cancelled due to Covid.
At the end, I have to give the credit to the Selection Committee, because there are so many ways to compare these teams, and many of them lead to different rankings. You have to give them a lot of respect for ranking a season that had so many cancellations, so few non-conference games, and many teams with only 1 or 2 top wins that still have the better records and deserve to be in the tournament.