How Single-Elimination Tournaments Work: The Mathematics of Brackets

The Madness of March: A Numbers Game

Every March, millions of people fill out brackets for the NCAA basketball tournament. They predict winners, cheer for upsets, and lament when their perfect bracket crumbles by the second round. But here’s the secret: the same mathematical skeleton that organizes 64 teams into one champion also shapes everything from corporate promotion ladders to evolutionary competition. A single-elimination bracket is more than a sports ritual—it’s a model of how we sort winners from losers in a world full of contenders.

Why Brackets Matter Beyond Sports

You might think brackets are only for athletics, but they show up in surprising places. Economists use “tournament theory” to study how people compete for promotions. Picture a company with dozens of employees all angling for one manager position. Each project or performance review acts like a game: the best performers advance, and only one ultimately gets the prize. This bracket-style structure affects how hard people work, how they take risks, and even how much they earn.

In venture capital, startups pitch to investors in elimination rounds—only a few make it to the final funding. In biology, organisms compete for resources in a similar win-or-go-home dynamic. Understanding the math behind brackets helps you see the hidden design in these systems. It’s not just about who wins; it’s about how the rules of elimination shape behavior and outcomes.

The Simple Math That Governs Every Bracket

Let’s start with the core idea. In a single-elimination tournament, every game eliminates exactly one team. If you begin with 64 teams and need one champion, you must eliminate 63 teams. That means 63 games. This holds for any number: with n teams, you play n – 1 games. It’s that straightforward.

But the structure has a neat twist. For all teams to be paired perfectly in each round, the total number must be a power of two: 2, 4, 8, 16, 32, 64, and so on. If the number isn’t a power of two, you need “byes”—free passes for some teams to the next round. Think of it like filling a bracket as a binary tree. Each game is a node that combines two winners. The depth of the tree—the number of rounds—is simply log₂(n). For 64 teams, log₂(64) = 6 rounds. This is why adding more teams doesn’t balloon the tournament: doubling the field only adds one extra round.

From 64 to 1: The Elimination Process

Let’s walk through March Madness, the classic example. With 64 teams, round one has 32 games, sending 32 winners forward. Round two: 16 games. Round three: 8. Then 4, then 2, and finally the championship. Six rounds total. Each round exactly halves the field. No team plays twice in the same round; every game is do-or-die.

This simple process hides a clever tool: seeding. Seeding ranks teams before the tournament (say, 1 to 16 in each region) and places them in the bracket so that the strongest teams don’t meet until later rounds. The top seed plays the weakest opponent first. Without seeding, a random draw could pit the two best teams against each other in the opening game—which might be exciting for fans but defeats the purpose of a fair competition that aims to crown the best overall team. Seeding increases the chance that the final match features the two most deserving contenders.

Sports, Economics, and Beyond: Brackets in Action

Sports give us the most visible examples. The FIFA World Cup knockout stage, Wimbledon tennis championships, and the NHL playoffs all use single-elimination brackets after an initial group or round-robin phase. In each case, the number of teams is engineered to fit powers of two or adjusted with byes.

But brackets also pop up in economics. Consider a sales competition where the top performer gets a big bonus. Each sales quarter is like a game; only those with the best results advance. This “winner-take-all” setup motivates effort but can also encourage risky behavior if the prize is too large relative to the runner-up. Tournament theory helps companies design these incentives so employees work hard without undermining cooperation.

In computer science, binary trees (the same structure as brackets) are used in search algorithms and data compression. Even evolution can be seen as a massive single-elimination tournament where genes compete for survival, and each generation weeds out less fit variants.

What Your Bracket Intuition Gets Wrong

Our brains play tricks on us when it comes to brackets. One common belief is that the best team almost always wins. But in a 64-team tournament with random matchups, the top seed actually wins only about 45% of the time. Upsets are baked into the system. A single bad game or a lucky bounce can knock out a favorite. The fewer rounds, the more randomness matters. That’s why longer series (like seven games in the NBA playoffs) reduce upsets, but single elimination magnifies them.

Another misconception: more teams mean many extra rounds. Not true. Going from 64 to 128 teams adds just one more round (from 6 to 7). The number of rounds grows logarithmically, not linearly. People often overestimate how much a larger bracket stretches the schedule.

People also assume all brackets are equally fair without seeding. In reality, an unseeded bracket is a lottery: the strongest teams might meet early, making the tournament more volatile. Seeding levels the playing field, but it never guarantees the “right” winner. And remember, a single-elimination tournament is not the same as double elimination. Double elimination gives each team a second life in a losers’ bracket, which changes the odds dramatically—it’s not just a mulligan.

Beyond Single Elimination: What to Explore Next

Single elimination is elegant, but it’s only one format. Double-elimination tournaments reduce the sting of a single upset by giving teams a second chance from a losers’ bracket. Swiss-system tournaments, common in chess, pair players with similar win-loss records without eliminating anyone; the winner is the one with the best record after a predetermined number of rounds. Round-robin tournaments have every competitor play every other, guaranteeing a fair ranking but requiring many games.

Each format trades off fairness, time, and excitement. Single elimination is fast and tense, but high-variance. Double elimination is fairer but longer. Swiss system works well for large fields with limited time. Understanding these alternatives gives you a toolkit for designing competitions at work, in games, or even in personal goal-setting.

Three Numbers to Remember for Life

1. n – 1 games: The total number of games in any single-elimination tournament with n teams. Simple and universal.
2. log₂(n) rounds: The number of rounds needed; doubling the teams adds only one round.
3. 45%: The approximate probability that the top seed wins in a random 64-team bracket under plausible upset rates. A reminder that luck always plays a part.

Brackets aren’t just for March. They’re a window into how we structure competition, reward winners, and manage uncertainty. Once you see the math, you’ll never look at a tournament the same way again.