How to calculate winning percentage, including draws

By StratPollen Studio · Updated 25 October 2026 · 4 min read

Winning percentage answers a simple question: out of all the games played, how often did this team or player win? It's the most common way to compare records when teams have played different numbers of games, and it works for almost anything you can win or lose: league football, basketball, chess, esports, pickleball, board game nights.

The only tricky part is draws. This guide walks through the basic formula, the four common ways of handling draws, and a few worked examples you can check with a calculator.

The basic formula

When there are no draws, winning percentage is wins divided by games played:

Winning % = Wins ÷ (Wins + Losses) × 100

A team that wins 12 games and loses 8 has played 20 games. 12 ÷ 20 = 0.60, so the winning percentage is 60%.

Multiplying by 100 only turns the decimal into a percentage. Some sports leave it as a decimal, which is covered further down.

Four ways to count draws

As soon as games can end level, you have to decide what a draw is worth. There is no single right answer. Different sports and leagues pick different rules, and the choice can change the result a lot. These are the four you'll meet most often.

1. A draw counts as half a win

Winning % = (Wins + ½ × Draws) ÷ Games played × 100

This is the most widely used general convention. It treats a draw as half a win and half a loss, which is fair to both sides. The NFL has used it for ties since 1972, and it's the same idea as chess scoring, where a draw earns half a point.

2. A draw counts as not a win

Win rate = Wins ÷ Games played × 100

Here a draw is simply a game you didn't win. Football (soccer) managers' records are usually quoted this way, as a "win rate". It's stricter: a team that draws a lot will look worse than under method 1.

3. Ignore draws completely

Winning % = Wins ÷ (Wins + Losses) × 100

Draws are taken out of the count altogether, as if those games never happened. Before 1972 the NFL calculated it this way. In baseball, the rare tied game is also left out of winning percentage.

4. League points as a share of the maximum

Points % = Points won ÷ (3 × Games played) × 100

Most football leagues give 3 points for a win, 1 for a draw and 0 for a loss. Points percentage tells you how many of the available points a team actually took. It's a good way to compare teams in a league table that have played different numbers of matches.

Worked examples

Take one record and run it through all four methods: 10 wins, 6 draws and 4 losses, so 20 games played.

The same record gives anything from 50% to 71.4%. That's why it matters to say which method you used whenever you quote a winning percentage with draws in it.

Example: a chess club season

A player finishes a club season with 7 wins, 5 draws and 3 losses from 15 games. Chess gives 1 point for a win and ½ for a draw, so the score is 7 + 2.5 = 9.5 points out of 15. 9.5 ÷ 15 = 0.633, or 63.3%. This is exactly method 1.

Example: comparing two teams

Team A is 18-6 with no draws (75%). Team B is 14-3 with 3 draws, 20 games in total. Counting draws as half a win, Team B scores (14 + 1.5) ÷ 20 = 77.5%. Team B edges ahead even though it has fewer wins, because it has lost only 3 times.

Rounding and the ".600" style

For everyday use, one decimal place is enough: 65.0%, 71.4%. Two teams can look level at whole numbers and still differ in the first decimal.

Baseball and basketball standings usually show winning percentage as a three-digit decimal with no percent sign: a 60% record is written .600 and spoken as "six hundred". A perfect record is 1.000. If you see a team listed at .547, that means it has won 54.7% of its games.

Common mistakes

• Dividing wins by losses. 12 wins and 8 losses is not 12 ÷ 8 = 1.5. Always divide by games played.
• Forgetting draws in the total. With methods 1, 2 and 4, draws are still games played and belong in the bottom of the fraction.
• Mixing methods. Comparing one team's win rate (method 2) with another team's half-win figure (method 1) isn't a fair comparison.
• Reading too much into small samples. 3 wins from 3 games is 100%, but it says very little. The more games played, the more a percentage means.