The Bracket Model

My bracketology runs on the same resume metrics the selection committee publishes: NET, quadrant records, strength of schedule, and the rest of the seven designated metrics. Here is how the last one scored. ← NCAA Men's Basketball

The record

One completed season scored so far. That is a result, not yet a track record.

Paymon score
374
of 408 possible · 2026 field
Exact seed line
52
of 68 teams placed on the right line
Within one seedline
66
of 68 teams seeded within one line of the real field
Field accuracy
68/68
every selected team correctly identified
What the model produces
The scored bracket
The full 2026 field with every seed line, scored against what the committee actually did. Where the model matched NET, where the committee moved off it, and how each region shook out.
See the scored bracket →
Year by year
There is only one scored season here so far. Earlier brackets predate the site and were never captured in a form that can be graded, so a longer archive is still being built.
See the 2026 result →
How it works

The bracket forecast is built from the committee's own evidence rather than a win-probability engine. During the season that means Warren Nolan's NET nitty gritty and team sheets: NET rank, quadrant records, strength of schedule, WAB, KPI, BPI, SOR, KenPom and T-Rank. Before NET exists it falls back to ESPN's BPI.

Scoring uses the Paymon Scoring System, which grades exact seed placement and gives partial credit for proximity. Picking the right 68 teams and seeding them badly is a worse bracket than the raw selection accuracy suggests, and this is the metric that says so.

Note that scores are not comparable across formats. The 2026 bracket was built for the 68-team field with a maximum of 408 points. The 76-team era raises that maximum to 456.

Read how the field gets built →

What it cannot see

The committee is twelve people in a room, and they weigh things no published metric carries: injuries and who was available for which losses, the eye test, conference politics, and the bracketing principles that pull teams off their true seed line for geography and conference separation.

One scored season is a small sample. A single good result says less than it looks like, and I would rather say that plainly than present 374 points as a track record.

The forecast is a seed list, not a tournament simulation. It says nothing about who wins games once the field is set.