How stats and history would pick the 2021 NCAA Tournament

Round of 32

West Region

(1) Gonzaga over (8) Oklahoma. Gonzaga has the highest odds of Sweet 16 advancement (90%) since 2000-01 Duke. No 1 seed at 82.5% or better to make the Sweet Sixteen has missed it.

(5) Creighton over (13) Ohio. This was originally Creighton over Virginia, but applying a small COVID pause deduction to Virginia’s metrics (thanks to Evan Miyakawa for his research) allowed us to just pick Ohio instead.

(6) USC over (3) Kansas. I would’ve made this pick even before Jalen Wilson was announced as out, but USC has the highest Sweet Sixteen odds of any 6 seed by miles at nearly 44%. My general 6 seed Sweet Sixteen cutline is 34%, and they’re way above that.

(2) Iowa over (7) Oregon. Oregon doesn’t meet our 20% Sweet Sixteen odds cutline for 7 seeds. 7 seeds at 20% or worse to get there are just 4-for-42, and 2 seeds at 63% or better (Iowa is around 70%) are 30-for-36. Play the percentages.

South Region

(1) Baylor over (9) Wisconsin. I don’t like this pick, as Wisconsin is one of the strongest 9 seeds ever and Baylor is below our 70% Sweet Sixteen odds cutline. 1 seeds below 70% are just 18-for-25 in making the Sweet Sixteen since 2000; the rest are 49-for-55. Only two of the 25 (2003 Texas & 2018 Kansas) have made the Final Four. Not only that, we average about two 1 seed Round of 32 upsets every three years. It would be wonderful to be picking Wisconsin here. But I can’t pull the trigger and be comfortable with it. Wisconsin’s Sweet Sixteen odds of 19.5% are the fifth-best of the last 20 years for any 9 seed, but the issue is this: only five 9 seeds have even made the Sweet Sixteen since 2000. If you’re rolling with Wisconsin, not only are you hoping they beat a tough 8 seed in the first round, you’re also hoping they beat the only team that was consistently ranked alongside Gonzaga this season until a late cold streak.

(5) Villanova over (4) Purdue. No Gillespie for Villanova, but they’re still better even with that penalty applied. Villanova’s Sweet Sixteen odds still sit at about 40-41%; 5 seeds at 41% or better to make the next round are 19-for-26 in making it. This is also a fade of Purdue, who has a really tough first round game with North Texas and is a neutral-court underdog to Villanova.

(6) Texas Tech over (3) Arkansas. Another value play based on Arkansas’ Round of 64 matchup and Texas Tech being Texas Tech. The Red Raiders sit at 35% to make the Sweet Sixteen; 6 seeds whose odds are 34% or better are 10 for 20 in making it historically. Along with that, 58 teams in the 64/65/68-team eras have started the preseason unranked yet finished it in the top 10. Precisely two of those teams have made the Final Four, and almost all went out prior to the Elite Eight. Fade Alabama and Arkansas.

(2) Ohio State over (7) Florida. Ohio State beats the 63% cutline pretty easily. Plus, you’re not gonna bet on Mike White, are you?

Midwest Region

(8) Loyola Chicago over (1) Illinois. Hoo boy. Picking against the second-hottest team (behind Gonzaga) in the entire field is a frightening proposition, and I am sure this will not end well. Again, though: we’re acting as if this is for a moderately-sized bracket group, and you need to take risks. The most calculated of these is picking a Loyola Chicago team that’s ranked top 10 in KenPom, has a 22% chance to make the second weekend (8 seeds at 12% or better are 9-for-46; all others 0-for-34), and would only be 4.5-point underdogs to the Illini. Illinois has the lowest Sweet Sixteen odds by a hair of the 1 seeds. This year, out of the eight 1 & 2 seeds, the odds are in the favor of about 5.8 of them making the Sweet Sixteen. Only three teams don’t meet the numbers I traditionally look for in early exits: Baylor (explored above), Alabama (explored later), and Illinois. If I’m going by the numbers, I have to take two huge swings and hope they hit. Here’s one.

(5) Tennessee over (4) Oklahoma State. Tennessee is the better team on every metrics site, has an easier Round of 64 draw, and, importantly for our stats, is at 45% to make the Sweet Sixteen. 5 seeds at 35% or better to make the next round are 23-for-41, and more specifically, 5 seeds at 41% or better are an astounding 24-for-26 in getting there. Oklahoma State sits at about 38%, which is low for a 4 seed. 4 seeds at 40% or worse to make it are 2-for-28 in doing so.

(3) West Virginia over (11) Syracuse. West Virginia’s Sweet Sixteen odds aren’t good enough, but I don’t feel comfortable having zero three seeds make it.

(2) Houston over (10) Rutgers. Even the least-charitable Houston odds have them well above the 63% cutline, and Rutgers sits below the 17% cutline for 10 seeds. 10 seeds at 17% or worse to make the Sweet Sixteen: just 2-for-52.

East Region

(1) Michigan over (9) St. Bonaventure. Same deal as earlier: St. Bonaventure is a very good 9 seed, but even the most charitable projection still has Michigan at 70% to make the second weekend. HOWEVER: this is one I’m watching. Michigan would be about a 7-point favorite over either the Bonnies or LSU on KenPom, but this does not include many games sans Isaiah Livers, who won’t be playing in the Tournament. A non-Livers Michigan isn’t the same team, which is why I’ll be making a certain painful, hurtful, depressing, stupid pick later.

(4) Florida State over (12) Georgetown. Seminoles are easily the best 4 seed in the field and also sit anywhere from 46-51% to make the second weekend. If you’re at 47.5% or better to make the Sweet Sixteen, you’re 24-for-29 in getting there so far.

(11) UCLA over (14) Abilene Christian. I would also take UCLA over Texas, for the record. I’m not going to take a 14 seed further than the Round of 32 ever, but we’re long overdue for a 14 seed in the Sweet Sixteen. Torvik’s historical tourney stats suggest we should’ve had three or four in the last 20 years; we’ve had none. This is probably the year it happens if it’s ever going to. UCLA’s Sweet Sixteen odds after the MSU win are now a little over 17%; 11 seeds at 17% or above are 12-for-27 in making the Sweet Sixteen, and UCLA is the only 11 seed that hits this number. UCLA and Utah State are the only two double-digit seeds in the field to meet historical requirements for Sweet Sixteen advancement, and 12 straight NCAATs have had a double-digit seed among the final 16 teams. I’m rolling with UCLA because I simply believe their path is better than Utah State’s.

(7) Connecticut over (2) Alabama. YOU BET. Alabama, depending on where you look, is anywhere from 55-58% to make the Sweet Sixteen. This sounds low, and it is, because UConn is playing at the level of a 4 seed, not a 7. The Huskies have a 25.4% chance to make the Sweet Sixteen per KenPom and a 27.6% chance per our metrics combination. Both are above the 20% cutline I have for 7 seeds (14-for-38 in getting to Sweet Sixteen). Meanwhile, by KenPom alone, Alabama’s 58% odds of getting to the second weekend do escape our 57% cutline, where 2 seeds above 57% are 44-for-58 in getting to the Sweet Sixteen.

HOWEVER: 44-for-58 is 75.9%, which means one out of four 2 seeds don’t get there. Every other 2 seed is at 66% or better, and Alabama’s easily the lowest on the 2 seed totem pole here. By a combined Torvik/KenPom metric, Alabama is at 56.8%. 2 seeds below that 57% cutline: 4-for-22 in making it.

Like I said earlier, you have to take big swings. Here’s my second.

NEXT PAGE: Sweet Sixteen, Elite Eight

One thought on “How stats and history would pick the 2021 NCAA Tournament

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s