If you’re expecting the NET to catch your favorite college basketball team, don’t be surprised if they slip through a tear.
When the NCAA Tournament selection committee delivered its annual “Bracket Reveal” preview, roughly three weeks before Selection Sunday (and well clear of the attention swallowed by the Super Bowl), all 16 teams included were among those ranked in the top 25 by the NCAA Evaluation Tool metric that is at the core of the selection process.
But the four seeds per region assigned this week — they are the seeds as they would be today, and are likely to change through the remaining games — were not identical to the NET rankings nor any other metric featured on the “team sheets” the members use to guide their decisions.
BRACKET PREDICTIONS: Bill Bender projects the Field of 68
For instance, No. 1 overall seed Alabama is second in the NET and the predictive ratings compiled by statisticians Ken Pomeroy and Jeff Sagarin, and third according to ESPN’S BPI. The Crimson Tide are No. 1 in the two results-based metrics, analyst Kevin Pauga’s KPI and ESPN’s Strength of Record. But if you think those are more accurate ratings to follow, you’ll find SOR No. 14 Tennessee is the ninth overall seed on the bracket reveal, and SOR No. 17 Iowa State is ranked six spots higher by the committee at 11th.
What the committee told us, more than anything, is that while there are mathematical elements to the NCAA Tournament seeding process, it’s still more art than science.
MORE: Breaking down the updated odds to win the 2023 NCAA Tournament
The committee always has insisted the NET — and its predecessor, the Ratings Percentage Index (RPI) — are tools used by the committee to group teams for the purpose of assessing their resumes. In the old RPI-based system, the selectors gave credit to teams that accumulated wins against opponents ranked among the top 50 or top 100 of the RPI standings.
At a meeting in January 2017, though, Pomeroy, Sagarin and others convinced the NCAA that they weren’t accurately capturing the increased difficulty of winning road games or neutral-court games as opposed to home games. The math indicated a win against the 75th-ranked team on the road was roughly equivalent to beating the 30th-ranked team at home. That led to the formulation of the Quadrant system.
MORE: Powerful Big 12 has chance to set March Madness records
Aware the RPI metric also was flawed, the NCAA also decided to develop a new metric. Although the NCAA insisted it wanted a team’s margin of victory to matter only marginally, the NET has turned out to track very closely with the results-based metrics, notably KenPom. Many fans have become convinced that a high NET ranking will automatically lead to a tournament berth or an advantageous seeding for the team they follow.
But Rutgers made the field last year with a NET ranking of 77. In 2019, N.C. State was excluded with a NET of 33.
It’s not a magic formula. It’s a tool. The committee members have insisted on this since the NET was introduced for the 2018-19 season, as they did for years before regarding the RPI. Just as winning enough games to make the NCAA field is not a simple matter, though, neither is it as easy as looking at computer ratings to figure out whether a team is worthy of inclusion in the field.
Here were the seeds revealed by the committee during a half-hour program Saturday on CBS, with their position on the NCAA’s 1-16 seed list in parentheses:
1. Purdue (3)
2. UCLA (8)
3. Iowa State (11)
4. Marquette (14)
1. Alabama (1)
2. Baylor (7)
3. Virginia (10)
4. Indiana (13)
1. Houston (2)
2. Texas (5)
3. Tennessee (9)
4. Xavier (16)
1. Kansas (4)
2. Arizona (6)
3. Kansas State (12)
4. Gonzaga (15)
Some minor surprises, and some things we learned:
— Arizona earning the No. 2 seed in the West over UCLA indicates the committee is valuing achievement against top opposition more so than statistical excellence. Arizona is No. 11 in the NET and KenPom (at least partly because of a less-than-stellar defense that impacts its predictive metrics) and No. 8 in Strength of Record (because of losses to four opponents likely to miss the tournament). But the Wildcats own victories over three of the top 16 seeds, including the Bruins. UCLA has been better analytically (No. 3 in KenPom, No. 5 in the NET) but hasn’t beaten anyone seeded higher than No. 7 in the composite of 94 online brackets at BracketMatrix.com.
— The committee is valuing Big 12 achievement. There were three of the conference’s members among the top seven teams, and five presented top-four seeds. That includes Kansas State, which received a No. 3 seed despite having lost four of its past five games, including to Texas Tech and Oklahoma, neither of which is included in any projected field at the Bracket Matrix.
— The proximity of the seed ranking for Kansas (fourth overall) and Texas (fifth) is an indication they are competing almost head-to-head for a spot on the top line, although Baylor (seventh) also might have a chance to force itself into the competition. The Bears entered Saturday with a game against each of them. KU and Texas play again March 4, the final day of the regular season for each.
— For those who still believe recency has an impact, Xavier received a No. 4 seed in the Midwest Region despite consecutive losses at sub-.500 Butler and Big East leader Marquette. That’s why losing three times in four games did not knock Purdue off the No. 1 seed line; the Boilermakers still were among the four teams the committee valued most.