How much does history matter when it comes to building the bracket?
In bracketology, it’s very easy to get muddled in the idea of precedent, simply because that’s what bracketology is mostly based on: analyzing data from previous years’ brackets to help determine what measures the selection committee most values, and apply that backward-looking analysis into a forward-looking manner.
This sort of thinking provides a helpful baseline for bracketology, as it allows you to recognize pretty obvious patterns, like noticing that teams with good résumé metrics and strong schedules typically get rewarded with better seeds.
But precedent can also be a dangerous thing to get caught up on because the selection committee breaks precedent nearly every year.
One stat that I frequently parroted a few years ago was the self-proclaimed “95% confidence interval,” which stated that no team that had ever been found in 95% of Bracket Matrix projections (since the Matrix expanded to over 100 bracketologists in 2012) had ever been left out of the field. This was even true in 2022 when Texas A&M, one of the most notorious snubs of all time, was included in 200 of 211 brackets, or 94.8%. Seems like a sure thing then, right?
Welp, the 95% confidence interval was promptly broken in 2023, when Rutgers appeared in 218 of 229 brackets (95.2%)… Only to be the committee’s second team out. And then the interval was shattered again last season, when West Virginia (244/250, 97.6%) was inexplicably kept out of the dance.
These sort of things that you believe are surefire end up broken more often than you’d think, with no fewer than three major breaks last year alone. No team in the NET era had ever gotten an at-large bid with a worse Quad 1-3 record than 2023 Providence at 10-11… Until 2025 Texas made the field at 12-15. No team with a quality metric average of 10.0 and under had ever been worse than a 5-seed… Until 2025 Gonzaga landed all the way at the 8-line. No conference had ever had more than 11 of its teams selected to the field of 68… Until the 2025 SEC sent 14 of its 16 teams to the dance.
You see what I mean? Sometimes we bracketologists get so bogged down in the little details that we apply historical patterns that never should have been applied in the first place. Keeping an open mind that things can and will break what we thought we knew, while still being wise to what factors are important for selection and seeding, is a critical part of tackling the mental challenge that the bracketology process creates.
So I’m treading lightly over the idea of “historical precedence” in 2026. You could say it’s that very thing that’s keeping me from putting Nebraska on the 1-line, despite me being so very tempted to do so, what with their 3.7 résumé metric average, four Quad 1A victories, and flawless 20-0 record. It’s just that no team that’s ever earned a 1-seed has ever had a quality metric average worse than 10 (2024 North Carolina); the Cornhuskers would obliterate that with a mean BPI/KenPom/T-Rank ranking of 14. But if Nebraska goes into Ann Arbor and beats fellow 1-seed Michigan on Tuesday night? History has to go out the window—the Huskers would undoubtedly be on the top line.
Elsewhere, I’m breaking precedent already; no conference has ever had five teams in the top-9 overall seeds. (The SEC got darn close last year with four.) But the Big Ten is stacked at the top this season, with each of Michigan, Nebraska, Illinois, Michigan State, and Purdue currently residing in single-digit true seed territory, at least for me.
So if you want to get into the habit of saying “you can’t put that team there because no team like that has ever been put there before,” …just tread carefully.
Anyway, on to the bracket! As a reminder, my automatic bids for each conference are chosen based on who the leader is in my résumé metric BRCT as of yesterday afternoon. (Apologies to Merrimack and UMES, current conference standings leaders who would probably be more sensible choices to include over Quinnipiac and Howard. But, since BRCT includes non-conference results in its calculations, the latter two get the benefit for now.)
Enjoy your Monday morning with a new Bauertology projection, and see you again on Friday!
