Short version: the NCAA Division I committee ranks the 68-team field, splits it into four regions of 16, and hands out seeds 1 through 16 in each region so the best team in each region gets the No. 1. Thirty-two teams arrive with automatic bids, 36 are picked at-large, and a set of published rules then decides who lands where. Knowing the seed tells you a team’s first-round opponent and the road ahead. It does not tell you who wins.
I spent enough March Madness weekends arguing about my team’s seed to know exactly which questions fan forums throw around: does the AP poll matter, does one huge win in February move a line, is this whole thing math? I’ll get to all three. First, the basic chain, which is genuinely simple once you see the pieces are separate from each other.
Table of Contents
- How College Basketball March Madness Bracket Seeding Works: Quick Guide
- What Is a Seed in the March Madness Bracket?
- Who Decides the Tournament Field and Rankings?
- How Does the Committee Evaluate Teams for Seeding?
- How Are the 68 Teams Placed Into the Four Regions?
- How Does the Bracket Assign First-Round Matchups and Byes?
- Why Can a Lower Seed Beat a Higher Seed?
- What Happens After Selection Sunday?
- Frequently Asked Questions
- Conclusion
How College Basketball March Madness Bracket Seeding Works: Quick Guide
The single most common confusion is treating the committee’s ranking, the NET ranking, the seed line and the bracket position as one thing. They are four different things, produced at four different moments.
| Term | What it is | Who or what sets it |
|---|---|---|
| Committee ranking | The committee’s own 1-through-68 ordering of the field, used for the Final Four | The 12-member selection committee |
| NET rank | A public 1-through-366 ranking of every Division I team | A formula published by the NCAA using four components |
| Seed line | One of 16 slots in a region, numbered 1 to 16 | The committee, applying the seed-line assignment process |
| Region | One of four 16-team sections of the bracket | The committee, with the overall No. 1 seed picking first |
| Bye | A first-round game the team does not play | Automatic for the 8 best seeds; the rest come from the First Four |
How college basketball March Madness bracket seeding works step by step
- 32 teams win their conference tournament on Selection Sunday and lock in an automatic bid.
- The committee adds 36 at-large bids to reach 68.
- The committee ranks all 68 teams and then splits them into four regions of 16.
- Each region receives one team at every seed line, from No. 1 down to No. 16.
- Placement rules handle rematches, travel, and home-crowd advantages.
- First-round sites are assigned, and the four First Four play-in games decide the last four open spots.
Keep that sequence in your head and the rest of this guide is just detail on each link.
What Is a Seed in the March Madness Bracket?
A seed is a team’s position within its region, expressed as a number from 1 to 16. In the East region, one team is the No. 1, one is the No. 2, and so on down to the No. 16.
What a seed controls
A seed decides three things: your first-round opponent, the seed line you play against again if you win, and where your elimination game happens in the regional bracket. It is a statement about expected strength, not a reward for effort or a judgment about how entertaining your team is.
What a seed is not
- Not the NET rank. The NET ranks all 366 Division I teams. A seed exists only among the 68 that got in, and only inside one region.
- Not the committee’s overall ranking. Overall seeds 1 through 4 become the No. 1 seeds. Overall seeds 5 through 8 become the No. 2 seeds. The two lists are related, not identical.
- Not your conference. Two teams from the same conference can end up on different seed lines, in different regions, with no seeding penalty for it.
- Not the first-round opponent by itself. A No. 5 knows who it plays in round one; the No. 12 opponent is assigned by the 5-12 pairing, not by anyone choosing.
- Not a promise. The gap between a 1 and a 16 is enormous on paper and in practice. The gaps at 8-9 and 5-12 are not.
Two teams can share a seed line number in different regions, and they will usually face very different second-round opponents. Seed 8 in one region can be the toughest draw in the tournament if it lands next to a No. 1.
Who Decides the Tournament Field and Rankings?
Two bodies matter here. Conferences hand out automatic bids on their own. The NCAA Division I Men’s Basketball Committee fills the rest.
| Automatic bid | At-large bid | |
|---|---|---|
| How many | 32 | 36 |
| Who grants it | The conference, when a team wins its conference tournament | The NCAA selection committee |
| Seed range | Can be anywhere from No. 1 to No. 16 | Anywhere from No. 1 to No. 16 |
| Lock-in point | Automatic once the conference tournament is won | Only when the field is unveiled |
The committee is made up of 12 members drawn from Division I institutions, a mix of athletic directors and conference commissioners, who serve five-year terms. Membership rotates, which is why any article listing “the committee” from a specific year goes stale fast. Treat the role, not a roster.
Members submit confidential ballots ranking teams they believe belong in the field. Those ballots are discussed, revised and re-voted over several days, with the chairs sitting in while anyone with a conflict recuses themselves and leaves the room. The final product, released on Selection Sunday, includes the 68-team field, the seed lines, the regions and the first-round and second-round sites.
Committee chairs have said the process is closer to consensus than arithmetic. Bubba Cunningham, a former chair, framed the question members argue about as who did you play, where did you play and how did you do.
How Does the Committee Evaluate Teams for Seeding?
The NET is one input, not the decision
The NCAA Evaluation Tool, or NET, is the one metric with an official role in selection. It is a public ranking of all 366 Division I teams built from four components: efficiency, which looks at offensive and defensive margin adjusted for opponent; opponent quality; game location, meaning home, neutral or road; and a game-margin component that credits close wins and punishes narrow losses.
NET rank matters, and it shapes the conversation every year. It does not sort itself into seeds. Plenty of teams get seeded a line or two away from their NET position, and the NET order frequently disagrees with what the committee publishes.
Quadrant wins and losses come straight from NET
Because the NET ranks every team, it also tells you who each team beat. A quadrant win is a win over a team in a given NET range at a given location. NCAA.org publishes the men’s thresholds in NET rank:
| Result | Home site | Neutral site | Road |
|---|---|---|---|
| Quadrant 1 win | 1 to 30 | 1 to 50 | 1 to 75 |
| Quadrant 2 loss | 31 to 75 | 51 to 100 | 76 to 125 |
| Quadrant 3 loss | 76 to 125 | 101 to 175 | 126 to 200 |
| Quadrant 4 loss | 126 to 225 | 176 to 350 | 201 to 275 |
Read that carefully, because the direction flips for wins and losses. Beating a top-30 team at home is a Quadrant 1 win. Losing to a 200th-ranked team at home is a Quadrant 4 loss, the kind that hurts a resume badly. Quadrant numbers can also shift during the season because they depend on where the opponent sits in the NET that day.
Results metrics versus predictive metrics
Analysts who model the tournament rank teams differently from the NCAA, and the two families of metrics answer different questions. Results-based metrics describe what already happened. Predictive metrics estimate who is better right now and who would win a game played tomorrow.
| Results-based | Predictive |
|---|---|
| NET | KenPom |
| Wins Above Bubble (WAB) | Torvik |
| Kevin Pauga Index (KPI) | Basketball Power Index (BPI) |
| Strength of record | Adjusted efficiency projections |
Members look at more than one kind of number. Predictive systems and results systems disagree about plenty of teams, and the committee uses that disagreement as information rather than settling it.
The published selection criteria
The NCAA publishes the criteria the committee weighs, in alphabetical order. Condensed, the committee looks at bad losses, common opponents, how competitive a team’s losses were, whether a team peaked early or late, head-to-head results, an observable component that covers intangibles like character and leadership, significant wins, strength of schedule, the NET and Wins Above Bubble.
Head-to-head results matter, but as a contributing factor and often a final tiebreaker, not as a formula. A long forum thread about seeding debate on a UConn board landed on the same conclusion: one head-to-head result is a data point inside a full-season resume.
Three myths worth dropping
- The AP poll has no formal role. AP and coaches’ polls are not part of selection. Fans still watch them move, which is exactly why a team dropping from No. 2 to No. 4 feels like an insult.
- One big win rarely moves a seed line by itself. The NET already prices in opponent quality and margin, so a huge win against a weak team adds little. Big wins against top-30 teams move resumes, and fans often overestimate the effect.
- Seeds are not pure math. The observable component is written into the criteria. Two teams with nearly identical profiles can be separated by committee judgment, and that is the design, not a bug.
How Are the 68 Teams Placed Into the Four Regions?
Placement happens in a fixed order. The four No. 1 seeds go into four different regions, then the four No. 2 seeds are placed so they do not share a region with a No. 1, and so on down the list. The overall No. 1 seed chooses which region it wants first, including its first-round site.
After that basic distribution, the committee balances regions by the sum of their seeds. Sixteen teams numbered 1 through 16 add up to 136, so the ideal split is 34 per region. A region carrying three No. 1 seeds and no No. 4 seeds is not balanced, and the committee will move teams around to correct it.
That balancing is why a team can land a seed line it did not project, and why the committee is allowed to move a team up or down one seed line, or two in extraordinary circumstances. A forum poster asking why their team jumped to the No. 5 line instead of the No. 6 got the unsatisfying but accurate answer: regional balance usually explains it better than any single game result.
Three placement rules do most of the work:
- Rematch limits. Two teams from the same conference cannot meet in the first round, and in some bracket structures they cannot meet in the second round either. Conference rivals who fans expect to keep colliding get separated.
- Home-crowd disadvantage. Seeds 1 through 4 are placed away from their home site when they would otherwise play close enough to home to draw a road-heavy crowd. The committee treats an expected home-crowd edge as something to offset, not reward.
- Travel and areas of natural interest. Assignments try to limit cross-country trips in the opening round and keep a team’s region within a reasonable distance of its fan base.
None of these turn geography into a right. A No. 1 seed can absolutely be sent three time zones from campus, and there is no appeal.
How Does the Bracket Assign First-Round Matchups and Byes?
Once regions are set, the pairings are automatic. Only two of the sixteen first-round games in each region are undecided at Selection Sunday.
| Round of 64 | Round of 32 (if the seeds hold) |
|---|---|
| No. 1 vs. No. 16 | Winners: No. 1 vs. No. 8 |
| No. 8 vs. No. 9 | Winners: No. 4 vs. No. 5 |
| No. 4 vs. No. 13 | Winners: No. 2 vs. No. 7 |
| No. 5 vs. No. 12 | Winners: No. 3 vs. No. 6 |
| No. 2 vs. No. 15 | Winners: No. 1 vs. No. 4 |
| No. 7 vs. No. 10 | |
| No. 3 vs. No. 14 | |
| No. 6 vs. No. 11 |
The 8-9 game has one wrinkle: if both teams came from the same conference, the higher seed is placed in the 8 slot instead, so they cannot play each other.
Who gets a bye
The eight best seeds in each region, No. 1 through No. 8, play no first-round game. They enter in the round of 64. The other eight slots in each region are filled by the four First Four winners and four automatic-qualifying teams.
What the First Four actually is
The First Four exists because the field grew from 64 to 68. Four teams that would otherwise miss the tournament play two early games, and those two winners play each other, leaving one team to take the final No. 16 seed in one region. It is a short tournament with one prize: a first-round game instead of no tournament at all.
The No. 15 and No. 16 seed lines are provisional until those games are played. Bracket pools that lock before the First Four is settled are picking under a known asterisk.
Why Can a Lower Seed Beat a Higher Seed?
Because a seed measures expected strength, and expected strength is a distribution, not a switch. The gap between a No. 15 and a No. 2 is the gap between two teams that played very different schedules. The gap between a No. 9 and a No. 8 is almost nothing.
First round, a No. 16 beating a No. 1 is the rarest and most discussed result in the bracket, and it happens on a schedule the committee can see coming. Weak records, a thin nonconference slate, and a conference that did not test anybody. When it does happen, the higher seed usually rebounds and wins its next game against a No. 15 that had to play a 5-12 game first. Tired teams get tired.
Later rounds the math narrows. A No. 5 that survives the first weekend faces a No. 4, then a No. 1. Nothing in the bracket says that No. 5 has to lose, only that history has usually gone that way.
Seeding exists to make the early rounds predictable so the later ones matter. Treat it as a starting expectation, then update.
What Happens After Selection Sunday?
Once the bracket is public, four checks pay off before you fill anything in.
- Find your team’s region. The region matters more than the seed. A No. 3 in the region with the No. 1 you dislike faces a much harder run than the same No. 3 in a softer region.
- Look at the second-round opponent. Trace the bracket one round deeper. The team a No. 5 could meet is not decided, but the seed line it would meet is.
- Check the quadrant record. Find the team’s NET rank, then see whether its best wins are Quadrant 1 wins or only Quadrant 2 and 3 wins. That distinction explains a lot of seeding arguments.
- Treat bracketology as a forecast. Projections from sites and analysts are built before Selection Sunday and are designed to be as close to right as possible without cheating by copying the real bracket. When they differ from the committee, that gap is the interesting part, not a reason to trust either one blindly.
One more thing fans get wrong: the bracket is not a difficulty rating. Seeding controls who plays whom, and placement rules even out some regional imbalance, but nobody is optimizing for a fair run. A No. 1 seed is expected to win twice before it meets another No. 1 seed. That is the design.
Frequently Asked Questions
Are March Madness seeds the same as the final rankings?
No. The committee’s final ranking covers all 68 teams, and only its top four entries become No. 1 seeds. NET rank covers all 366 Division I teams and has no formal role in assigning seeds at all. A seed is a position inside one region, so a No. 8 seed in the South and a No. 8 seed in the West can sit very differently in the two lists.
Why do some No. 1 seeds have easier paths than others?
Each region is built with one team at every seed line, so on paper all four look even. Balance is checked using the sum of seeds, and placement rules like rematch limits and the home-crowd disadvantage can still leave one No. 1 seed with a softer second-round opponent than another. Expect talent to be the strongest predictor, not the bracket path.
How are the four No. 1 seeds chosen?
The committee ranks the field and the top four teams become the No. 1 seeds. A No. 2 seed can move up a line if the committee needs regional balance, so the four No. 1 seeds are not always the committee’s top four entries in order. The overall No. 1 seed also gets first choice of region and first-round site.
Do teams choose their March Madness region?
Only in one case. The overall No. 1 seed selects its own region and first-round site. Every other team is placed by the committee under the published placement rules, which cover same-conference rematches, travel distance, and the home-crowd disadvantage for the top four seeds in each region.
What is the purpose of the play-in game?
The First Four gives four teams a path into a tournament they would otherwise miss. Two games are played, the two winners meet in a third game, and that champion claims the remaining No. 16 seed in one region. Those games are the only reason a No. 15 or No. 16 seed line stays unsettled until the week of the tournament.
Does a higher seed always have home-court advantage?
It does not, and the committee works against it on purpose. For seeds 1 through 4, the placement rules push teams away from home when they would otherwise draw a road-heavy crowd in their area of natural interest. The reasoning is that a genuine home edge in a loud building is harder to defend than the official rules suggest.
Conclusion
Seeding is a ranking exercise followed by a placement exercise, and keeping those two apart explains almost every argument about it. The committee ranks 68 teams using NET, quadrant records, strength of schedule, head-to-head results and published judgment criteria, then spreads those teams across four regions under rematch, travel and home-crowd rules.
If you are new to this, do one thing first: find your team’s region before you look at anything else. A seed without a region is half an answer. Then check its NET rank and its best quadrant win, and you will understand more about Selection Sunday than most people arguing on a forum about it.
