What is a break-even win rate calculator?
A break-even win rate calculator shows the share of bets you must win to get your money back at a given price. It divides 100 by the decimal odds. At decimal odds of 1.91, which is -110 in American format, the answer is 52.4%. Win less often than that, and your stakes shrink over time.
The reasoning is short. A winning bet at decimal odds d returns the stake times d, so the profit is d minus 1 for every 1 staked. A losing bet loses the 1. The two balance when w × (d – 1) = (1 – w) × 1, which simplifies to w = 1 ÷ d. That is the break-even win rate, written as a percentage by multiplying by 100.
It is the same number as the implied probability of the price, but you read it differently. Implied probability describes what the price says about the event, margin included. The break-even win rate is the hit rate your own record has to reach. The tool assumes every bet is placed at the same odds, a win returns the stake times the decimal odds, a loss costs the stake, and fees are ignored.
How do you use the break-even win rate calculator?
Type the odds you bet at into the Your odds box, in decimal (1.91), fractional (10/11) or American (-110) form. The result is the percentage of bets you need to win to break even. It updates as you type, runs in your browser and stores nothing.
- Find the price. Use the odds you actually took, not the odds you saw earlier. If the bet is a parlay, use the combined odds.
- Enter it. Any format works. Put the minus sign on American odds: -110 is read correctly, but 110 is read as decimal 110.
- Read the percentage. The large number is the break-even win rate. The bar beneath shows it on a scale of 0 to 100%.
- Open the Formula line. It shows the division the tool did, to two decimal places.
- Compare it with your record. Divide wins by bets, but only for bets placed at about this price.
Example: Enter 1.91. The tool divides 100 by 1.91 to get 52.36%, shown as 52.4%. Suppose you place 100 bets of 10 units at 1.91. A win returns 19.10 (9.10 profit). With 52 wins you get back 52 × 19.10 = 993.20 against 1,000 staked, a loss of 6.80. With 53 wins you get back 1,012.30, a gain of 12.30. So 53 of 100 is the first count that returns more than was staked.
What break-even win rate does each price need?
Shorter prices need higher win rates, and longer prices need lower ones. A -200 price needs 66.7% wins, even money (2.00 or +100) needs 50%, and a +300 price needs 25%. Each figure below is 100 divided by the decimal odds, so you can check any price yourself.
| American | Decimal | Fractional | Break-even win rate |
|---|---|---|---|
| -500 | 1.20 | 1/5 | 83.33% |
| -300 | 1.33 | 1/3 | 75.00% |
| -200 | 1.50 | 1/2 | 66.67% |
| -150 | 1.67 | 2/3 | 60.00% |
| -120 | 1.83 | 5/6 | 54.55% |
| -110 | 1.91 | 10/11 | 52.38% |
| -105 | 1.95 | 20/21 | 51.22% |
| +100 | 2.00 | 1/1 | 50.00% |
| +120 | 2.20 | 6/5 | 45.45% |
| +150 | 2.50 | 3/2 | 40.00% |
| +200 | 3.00 | 2/1 | 33.33% |
| +300 | 4.00 | 3/1 | 25.00% |
| +400 | 5.00 | 4/1 | 20.00% |
| +1000 | 11.00 | 10/1 | 9.09% |
Small price changes matter more than they look. Moving from -105 to -110 raises the break-even rate from 51.22% to 52.38%, a rise of 1.16 points. Moving from -110 to -120 raises it to 54.55%, a further 2.16 points. A slightly better price lowers the win rate you need.
The default 1.91 and exact -110 give 52.36% and 52.38%. Both round to 52.4% in the tool. Decimal 1.91 is the usual rounded way of writing -110, whose exact decimal value is 1.9091.
How does the break-even win rate connect to the bookmaker’s margin?
The margin is the gap between a fair price and the price on offer, and it is what pushes the break-even rate above 50%. At -110 on both sides of a market, the two break-even rates add to 104.76%. A bettor who wins half of their bets at -110 loses about 4.5% of everything staked.
Here is the maths for a 50% win rate at -110 (decimal 1.9091). Expected result per 1 staked = w × d – 1 = 0.5 × 1.9091 – 1 = -0.0455, a loss of 4.55%. Over 100 bets of 11 units (1,100 staked), 50 wins return 10 profit each and 50 losses cost 11 each: 500 – 550 = -50 units. That is 4.55% of 1,100.
The same formula gives the profit per bet at any win rate. Expected result per 1 staked = w × d – 1. It is zero exactly when w = 1 ÷ d, which is the break-even point.
| Win rate at -110 | Average result per unit staked |
|---|---|
| 40.00% | -23.64% |
| 45.00% | -14.09% |
| 50.00% | -4.55% |
| 52.38% | 0.00% |
| 55.00% | +5.00% |
The 4.76% and the 4.55% are different measures of one margin. The 4.76 points is the overround, the amount by which the two break-even rates exceed 100%. The 4.55% is the hold, the share of all money staked that the bookmaker keeps on average when bets split evenly. Both come from the same price. A separate page on the overround is planned.
Casino bets work the same way. A straight-up number on a single-zero roulette wheel pays 35 to 1, which is decimal odds of 36. The break-even rate is 100 ÷ 36 = 2.78%, but the wheel has 37 pockets, so a number hits 2.70% of the time. The gap shows up as the 2.70% house edge: (1 ÷ 37) × 36 – 1 = -0.027. The page on roulette odds and payouts covers every bet.
How many bets does it take before a win rate means anything?
Far more than most people expect. At roughly even odds, a record of 100 bets has a 95% margin of error of about 10 percentage points, and it takes about 2,400 bets to narrow that to 2 points. A short record cannot separate skill from luck, so a win rate is only a rough guide until the sample is large.
The number of wins in n bets at a fixed win chance p follows a binomial distribution. The NIST Engineering Statistics Handbook gives its standard deviation as √(np(1 – p)). Divide by n and the win rate itself has a standard error of √(p(1 – p) ÷ n). The usual large-sample 95% interval is the win rate plus or minus 1.96 times that, the form shown on the NIST handbook page on confidence intervals for a proportion, which also notes better methods exist for small samples. The law of large numbers says averages settle toward the true value as bets accumulate, but slowly.
| Number of bets | Standard error | 95% margin of error |
|---|---|---|
| 50 | 7.06 points | plus or minus 13.8 points |
| 100 | 4.99 points | plus or minus 9.8 points |
| 250 | 3.16 points | plus or minus 6.2 points |
| 500 | 2.23 points | plus or minus 4.4 points |
| 1,000 | 1.58 points | plus or minus 3.1 points |
| 2,500 | 1.00 points | plus or minus 2.0 points |
| 5,000 | 0.71 points | plus or minus 1.4 points |
| 10,000 | 0.50 points | plus or minus 1.0 points |
Luck alone can look like skill. Take a bettor whose true win rate at -110 is exactly 50%, so they lose about 4.5% of stakes on average. In 100 equal bets they finish ahead if they win 53 or more, which happens 31% of the time. For 500 bets (262 wins needed) it is 15%, and for 1,000 bets (524 wins needed) it is 6.9%. These are exact binomial figures, not rough guesses.
A second question is how many bets it takes to tell a small real edge from break-even. A common planning formula is n = (z₁ + z₂)² × p(1 – p) ÷ δ², where δ is the gap between the true and break-even win rates. With a one-sided 5% significance level (z₁ = 1.645) and 80% power (z₂ = 0.8416), the results at a break-even rate of 52.38% are below.
| True win rate above break-even | Bets needed | Matching true win rate |
|---|---|---|
| 3 points | 1,714 | 55.38% |
| 2 points | 3,855 | 54.38% |
| 1 points | 15,422 | 53.38% |
Real studies point the same way. A study of Australian horse race betting in the Journal of Gambling Studies simulated random equal bets on 211 races and found that a bettor would need more than 10,000 bets with net returns above 9% to be reasonably called an expert, under that model’s assumptions. Your own tracking will be noisier still if prices vary, which leads to the next point.
Why can a high win rate still lose money?
Because the break-even rate depends on the price. Winning 60% of bets at decimal odds of 1.40 (-250) loses about 16% of everything staked, since that price needs 71.4% wins. A win rate only means something next to the odds it was achieved at. Compare the two, never the win rate alone.
Here is the same case in units. Ten bets of 10 units at 1.40, with 6 wins and 4 losses. The wins return 6 × 14 = 84 units against 100 staked, a loss of 16. A single loss at that price wipes out about two and a half wins, so the break-even rate sits well above 50%.
| Win rate achieved | Decimal odds | Break-even win rate | Average result per unit staked |
|---|---|---|---|
| 60% | 1.40 (-250) | 71.43% | -16.0% |
| 65% | 1.50 (-200) | 66.67% | -2.5% |
| 30% | 3.00 (+200) | 33.33% | -10.0% |
| 45% | 2.50 (+150) | 40.00% | +12.5% |
A low win rate is not a loss either, for the same reason in reverse. Longer prices pay more per win, so they need fewer wins. But a price already reflects the market’s view of the chance, margin included, so a lower break-even rate does not make a long price a better deal.
If your bets are at different prices, one win rate hides a lot. With equal stakes, and wins equally likely at every price, the single break-even rate is 100 divided by the average decimal odds. Five bets at 1.50 and five at 3.00 average 2.25, giving 44.4%. If you win more of the short prices than the long ones, that figure misleads, so track each price band separately.
What is the break-even win rate for a parlay?
For a parlay, enter the combined odds. The leg rates multiply: two legs at -110 pay 3.64 in decimal, so all legs must win together 27.4% of the time to break even. Three legs need 14.4%. Each leg still needs about 52.4%, but the margin compounds with every leg added.
| Legs at -110 | Combined decimal odds | Combined American odds | Break-even rate for all legs to win | Average result per unit staked if each leg is a 50% chance |
|---|---|---|---|---|
| 1 | 1.91 | -110 | 52.38% | -4.5% |
| 2 | 3.64 | +264 | 27.44% | -8.9% |
| 3 | 6.96 | +596 | 14.37% | -13.0% |
| 4 | 13.28 | +1228 | 7.53% | -17.0% |
| 5 | 25.36 | +2436 | 3.94% | -20.8% |
The last column is the cost of the margin growing. A 50% bettor loses about 4.5% of stakes on a single bet but about 8.9% on a two-leg parlay and 13.0% on three legs. Prices quoted for a combined bet can differ from the product of the single prices, so use the combined odds you are actually offered. A parlay calculator and a post on how parlays work are planned.
What mistakes do people make with a break-even win rate?
The most common mistake is comparing a win rate with the wrong price. A record that mixes odds has no single break-even rate. Other mistakes are ignoring the margin, judging skill from a short record and treating break-even as a goal. Reaching break-even only returns your stakes, and it is not a profit.
- Mixing prices in one record. Wins at 1.30 and wins at 3.00 do not count the same. Group bets by price.
- Typing American odds without the sign. -110 gives 52.4%. Typing 110 is read as decimal 110 and gives 0.9%.
- Reading a short run as skill. As the tables above show, 100 bets leave a margin of error near 10 points.
- Forgetting costs outside the price. Fees, commission, taxes and voided bets move the real break-even rate, and the tool does not include them.
- Using break-even as a target. Chasing losses to get back to even is a recognized warning sign of gambling harm. If tracking a win rate has started to feel like a reason to bet more, the Safer play toolkit has practical limits and help options.
What should you do next?
Enter the price you actually bet at, write down the break-even rate, and compare it with a record kept by price band and counted in hundreds or thousands of bets. Treat any gap you see as an estimate with a wide margin of error. Then look at what the margin costs you over time.
To see the other side of the price, the odds calculator shows the profit and return on a stake, and the implied probability calculator shows the chance a price suggests. The expected loss calculator turns a house edge into a cost for a set number of bets. All three are planned alongside this tool, and a spend tracker is planned for keeping a record of your own bets.
How to read the result
The large percentage is the break-even win rate: the share of bets you need to win, out of all bets placed at this price, for winnings to match stakes lost. The bar beneath shows it on a scale of 0 to 100%. The Formula line shows the division to two decimal places. Short prices need high rates: decimal 1.50 needs 66.7%. Even money (2.00) needs 50.0%. Long prices need low rates: decimal 5.00 needs 20.0%. Winning more often than this rate at this price returns more than you staked. Winning less often loses money over time.
How it is calculated
Break-even win rate (%) = 100 ÷ d, where d is the decimal odds. The odds you type are first turned into decimal. Fractional a/b becomes 1 + a ÷ b. Positive American +A becomes 1 + A ÷ 100. Negative American -A becomes 1 + 100 ÷ A. Decimal odds are used as typed and must be above 1.00. American numbers between -99 and +99 are not accepted. The result is rounded to one decimal place for the large figure and shown to two in the Formula line. The rule comes from setting the average gain equal to the average loss: w × (d - 1) = (1 - w) × 1, so w = 1 ÷ d.
Worked example
Worked example
Using the default input, Your odds = 1.91 (the usual rounded form of -110). The tool calculates 100 ÷ 1.91 = 52.36% and shows 52.4% with the label "You need to win this often to break even." Over 100 bets of 10 units at 1.91, each win returns 19.10. With 52 wins the return is 993.20 against 1,000 staked, a loss of 6.80. With 53 wins the return is 1,012.30, a gain of 12.30. Entering -110 gives 52.38%, which also shows as 52.4%.
Limits
The calculator takes one price and assumes every bet is placed at that same price. A win returns the stake times the decimal odds, and a loss costs the stake. It ignores fees, commission, taxes, rounding by the operator, voided bets, dead heats, cash-outs and boosted prices. It does not use stake size, and it cannot say whether a price is fair or a bet is worth placing. For mixed prices, work by price band. For parlays, enter the combined odds you are offered. Results are arithmetic estimates, not predictions.