The history of the one-seed: how often the top seeds win
The No. 1 seed is the bracket's safest pick and still a long shot to win the title. This is the history of the top line: how seeding began, what it buys a team, and why even the best teams so often fall short.
Fill in a bracket without thinking and you will pick the one-seeds. That is not laziness; it is the strongest single habit in the sport. The committee gives its four best teams the number one next to their names, sends them to the regions with the softest draws it can build, and starts them against a 16 seed who, until recently, had never won. If any teams deserve to be favored in March, they are those four.
And yet the one-seed's story is full of contradictions. It is the most successful seed in tournament history by a wide margin, and also a seed that most years loses at least one of its four representatives before the second weekend. It is almost invulnerable in the first round and strangely fragile in the second. It has produced the most dominant teams the tournament has seen, and the most famous collapses. A team can finish a season at 38-0 and fail to reach the final. Another can lose to a 16 seed one year and win the national championship the next.
This article follows the one-seed through the tournament's history: how seeding began, what the top line actually gets, how often it wins, the stories of its most memorable failures, and what the mathematics says about why. For the process that creates the seeds, see our explanations of how the bracket is built and of how the field is chosen.
Where the number comes from
For the first forty years of the tournament there were no seeds as we know them. Teams were placed in regions according to geography and the committee's judgment, and brackets were drawn so that nearby teams did not meet too early. The idea of ranking the entire field and numbering the teams was introduced in 1979, the same year Magic Johnson's Michigan State met Larry Bird's Indiana State in the title game. From then on, every team in the field had a number, and every tournament since has been framed by the seed list.
The format took its familiar shape in 1985, when the field grew to 64 teams: four regions, each with seeds 1 through 16. Within each region the bracket pairs teams so that the top seed meets the bottom one, the second meets the next-to-bottom, and so on. The committee does not simply assign seeds region by region. It ranks the whole field from 1 to 68, a process known as the S-curve, so that the best team overall is placed in a region with the weakest of the four two-seeds, the second-best team with the next-weakest, and so on down the line. The overall number one seed therefore gets the most favorable path available. That advantage is small but real, and it is one reason committees spend so much effort on the top four spots.
Here is what the one-seed's road looks like, round by round, in terms of the opponents it usually meets.
| Round | Usual opponent | What it means |
|---|---|---|
| First round | A 16 seed | The easiest game in the bracket |
| Second round | An 8 or 9 seed | A solid team from the middle of the field, not a pushover |
| Sweet 16 | A 4 or 5 seed, often | A strong team, but usually not a great one |
| Elite Eight | A 2 or 3 seed, often | A strong, high-seeded opponent |
| Final Four | The champion of another region | Another top team |
| Final | The other semifinal winner | Often a top seed |
Seeds do not always line up this neatly, because upsets scramble the opponents, but the table shows why the draw has such value. A one-seed that is playing well should reach the Sweet 16 comfortably and should reach the Elite Eight more often than not.
How often the top line wins
Since seeding began, a clear majority of national champions have been one-seeds, roughly three in every five. It is the one number about the one-seed that everyone remembers, and it is correct as far as it goes.
But look at it from the other direction. There are four one-seeds each year, so a one-seed's chance of winning the title in a given year is, on average, in the neighborhood of one in seven. That is much higher than the chance for any other seed, and much lower than casual fans tend to assume. If you pick one of the four top seeds to win it all, you are likely to be wrong in any given year. You are just less wrong than if you chose anything else.
The gap between "most champions are one-seeds" and "any given one-seed is unlikely to win" is easy to resolve. A one-seed is better than anyone else in the bracket, but it is not so much better that it beats six opponents in a row very often. The mathematics of six consecutive games makes the point.
The compounding problem
Suppose a team has a fixed chance of winning each of its games. The probability of winning all six is that chance multiplied by itself six times.
- At 80 percent per game, the chance of winning six in a row is about 26 percent.
- At 85 percent, it is about 38 percent.
- At 90 percent, it is about 53 percent.
These figures are an illustration of the arithmetic, not a claim about any particular team. The real chances vary from game to game, and a one-seed is a much heavier favorite in the first round than in the Elite Eight. But the lesson holds. To have even a coin-flip chance of winning the title, a team must be about 90 percent to win every one of its games, including a Final Four semifinal against one of the best three or four teams in the country. Almost no team is that good, and that is why the favorite so often fails even in tournaments that go mostly according to form.
This is also why back-to-back champions are so rare. The tournament has produced only a few repeat champions in the modern era, such as Duke in 1991 and 1992, Florida in 2006 and 2007, and Connecticut in 2023 and 2024. Repeating means going through the same gauntlet with a roster that has changed, and winning it all again at those odds is a remarkable feat. The best-known dynasty of all is UCLA's, which we cover in UCLA's dynasty: ten championships in twelve years.
The first round: almost always safe
In the 64-team era, one-seeds were almost perfectly safe against 16 seeds. They won their first 135 games against that line before something happened that no one had seen.
In 1989, Georgetown, a one-seed, beat Princeton 50-49 in a game that was close until the end. Princeton's coach Pete Carril, whose slow, methodical offense frustrated stronger opponents, had a team that nearly pulled off what no 16 had done. For years afterward, that game was the standard of the near miss. There were other close ones, but the line held.
Then, in 2018, UMBC beat Virginia 74-54 in the first round. The result was a blowout, not a close shave, and it ended the idea that the first round was a formality. Five years later, in 2023, Fairleigh Dickinson, which had reached the field through the First Four, beat top-seeded Purdue 63-58. Two losses in well over a hundred and fifty meetings is a rate of roughly one percent, which is why a bracket that picks all four one-seeds through the first round is correct almost every time. But the possibility, once proven, never went away.
The women's tournament had an earlier example. In 1998, Harvard, a 16 seed, beat top-seeded Stanford, twenty years before the men's tournament saw the same thing.
A fair reading of the record is that the 1-versus-16 game is not a contest in any normal sense, but it is not a foregone conclusion either. We look more closely at upsets as a statistical phenomenon in why upsets happen: the maths and the myths.
The second-round trap
If the first round is safe, the second is where one-seeds really lose. The opponent is usually an 8 or 9 seed, a team that the committee considered close in quality to the teams around the middle of the bracket. On paper the one-seed has a significant edge. In practice 8 and 9 seeds have beaten one-seeds in roughly one game in five, which means that with four one-seeds each year, it is common for at least one to go home in the first weekend.
Why does this round cause so much trouble? Part of the answer is simply that a team that is seeded 8 or 9 is a good team. The committee compares an 8 and a 9 to each other, but in absolute terms both have excellent records. Part is that the one-seed has just played a game in which it was a massive favorite and learned little from it, while its opponent has often come through a tight, tense first-round win. And part is that a single-elimination game does not allow the better team to rely on its depth or the law of large numbers. The better team has to win this particular night.
Dominant and disappointing
The record of the one-seed includes both the most successful teams in the sport's history and some of its most notable failures.
Indiana, 1976. Bob Knight's team finished 32-0 and won the national title, beating Michigan in the final, an all-Big Ten matchup. It was the last men's team to finish a season unbeaten. We discuss the coach and his methods in Bob Knight: motion offense, discipline and controversy.
UNLV, 1990 and 1991. The Runnin' Rebels beat Duke by 30 points in the 1990 final, 103-73, the largest margin in the history of the title game. The next year they came back unbeaten and as heavy favorites, only to lose to Duke 79-77 in the semifinals. The same Duke team went on to win the championship.
Kentucky, 2015. The Wildcats entered the Final Four at 38-0, the closest any team had come to a perfect season in decades. They lost to Wisconsin 71-64 in the semifinal.
Virginia, 2018 and 2019. The Cavaliers were the overall top seed in 2018 and became the first one-seed to lose to a 16. A year later they won the national title, reaching the final through a close semifinal win and beating Texas Tech in overtime, which turned the story from a disaster into a redemption.
Duke, 2019. The Blue Devils were the overall top seed, with an extraordinary freshman class, and were beaten by Michigan State in the Elite Eight by a single point.
These stories share a pattern: dominance does not guarantee a title, and failure on the biggest stage does not define a program permanently. The tournament has a long memory for results, but a short memory for resentment. The same team that fell in March can sit atop the rankings the next year.
The years the top line broke
Some tournaments have borne no resemblance to the seed list. In 2008 all four one-seeds, Kansas, North Carolina, Memphis and UCLA, reached the Final Four, the first time that had ever happened. Kansas beat Memphis in overtime in an all-one-seed final. In 2011 the Final Four had no one- or two-seeds at all: Kentucky, Connecticut, Butler and VCU were seeded 4, 3, 8 and 11. In 2023, the Final Four did not include a seed better than 4, with Connecticut a 4 seed, San Diego State and Miami 5s, and Florida Atlantic a 9.
These two extremes show the range of what is possible. A tournament can unfold according to the seed list, with the best teams advancing, or it can be dominated by underdogs. Neither outcome is inconsistent with the underlying probabilities; they are just different draws from the same distribution. And in nearly every tournament there is at least one team that wins more games than anyone expected. We tell some of those stories in Cinderellas: the greatest runs by double-digit seeds, and the most unlikely champion of all, an eight-seed, appears in the 1985 Villanova upset of Georgetown. Villanova's 1985 title remains the lowest seed ever to win the tournament.
The women's top line
The women's tournament tells a different version of the same story. Because the best women's teams have often been far ahead of the rest of the field, the top line has produced a larger share of champions than it has on the men's side, and the champions have tended to come from a small group of programs. Tennessee finished 1998 at 39-0 under Pat Summitt, Connecticut has gone through whole seasons without a loss on several occasions under Geno Auriemma, and South Carolina won the 2024 title without losing a game. Connecticut's run of 111 straight victories ended only in the 2017 Final Four, when Mississippi State's Morgan William hit a shot at the end of overtime to win 66-64. The shock of that game was proportional to how rarely such a result happens.
The lesson is that a seed's meaning depends on the spread of talent in the field. When a few teams are well ahead of everyone else, the top seeds win most of the time, and when the field is more even, as it has been in the men's game for a long time, the top line is a weaker favorite. Our look at the women's Final Four's history traces how that spread has narrowed as the sport has grown.
What the one-seed does and does not mean
A final point of perspective concerns what a seed actually measures. A seed is the committee's estimate of how good a team's season has been, translated into a number. It is not a prediction of who will win. The committee does not try to forecast the tournament; it tries to rank the teams by their records. A one-seed has the best resume, not necessarily the best chance to win six games against specific opponents on specific nights.
A second point is that the seed line hides differences. The best one-seed is often much stronger than the weakest one, and the overall top seed is meaningfully better than the fourth. The same applies to 2-seeds and the rest. When you read a history of how one-seeds have performed, remember you are looking at an average of very different teams.
A third is that seeds interact with the rest of the bracket. The overall top seed is given the weakest two-seed in its region, while the fourth one-seed is given the strongest, so a soft region for one team is by definition a harder one for another. The advantage of the top line is real, but it is also relative to the other seeds around it.
What to take away
The one-seed deserves its reputation as the tournament's best bet and its reputation as a disappointment. Both are true. It wins the title more often than any other line, usually reaches the second weekend, and almost never loses in the first round. But no team is so far ahead of the field that the arithmetic of six straight games leaves it comfortable. If you are filling out a bracket, take the one-seeds through the first weekend, think hard about which ones you trust in the Elite Eight, and be willing to pick a champion who is not on the top line. Our bracket-picking strategy article follows that thinking through to the pool, and the men's brackets pages keep every year's finished bracket so that you can check the history for yourself.