How it works

1. Our model

Every team has a strength rating. From those ratings we work out Nebraska's chance of winning each game left on the schedule, then play out the rest of the season 10,000 times, Big Ten tournament included.

Nobody knows exactly how good a team really is, least of all in November. So each played-out season also makes Nebraska a little better or worse than its current rating, at random. That allowance is large early and shrinks as games are played, which keeps early numbers from being too sure of themselves.

Each played-out season is scored by one question: did Nebraska win more games than a typical bubble team would have against the same schedule? Quad 1 and Quad 2 wins earn extra credit and Quad 3 and Quad 4 losses cost extra, the way the selection committee reads a team's record. Comfortably more means in, clearly fewer means out, and winning the Big Ten tournament always means in. The share of seasons that end with a bid is the model's number.

2. The bracket consensus

Dozens of people publish projected tournament brackets. The second number is simply the share of those brackets that have Nebraska in the field today.

3. The blend

The headline number mixes the two. Early in the season it leans on our model, because brackets only describe today and the model looks ahead. As Selection Sunday gets closer the brackets become more accurate, so their share grows from 20% to 80%. Right now the brackets count for 0%.

Until the NCAA publishes its first NET rankings in early December, the number is labelled an early estimate. It is our model on its own, built on team ratings that still lean on preseason expectations, so treat it as a rough guide.

Where the data comes from

• NCAA: NET ranking and quadrant records
• Bart Torvik: team strength ratings
• Bracket Matrix: the bracket consensus
• ESPN: schedule and scores
• Warren Nolan: strength of schedule

This is an estimate, not a promise. The selection committee is made of people, and people surprise models every March.