Break-even win rate: how often a bet must win at given odds
Divide 1 by the decimal odds. At 2.00 a bet must win 1 time in 2 (50%) to get back what was staked; at 4.00 it must win 1 time in 4 (25%). Fractional a/b needs b / (a + b). Any win rate below that figure loses money over a long run.
The short answer
Every price carries a hurdle: the share of bets that must win just to get your money back. That hurdle is the break-even win rate. It is the same number as the implied probability, and the method for reading it from any format is in the guide on converting betting odds into implied probability.
This note looks at the hurdle from the bettor's side of the table. Instead of asking "what does the price imply?", it asks "how many wins out of how many bets do I need, and how far is that from what I can reasonably expect?"
The formula and where it comes from
D is the decimal odds, the total returned per shilling staked when the bet wins. Suppose you place n bets of one shilling each at odds D and w of them win. You staked n and received w × D.
The result is zero when w × D = n, that is, when w ÷ n = 1 ÷ D. Any win rate lower than that leaves you short, and any rate higher leaves you ahead. The stake size does not matter, because it multiplies both sides equally.
Illustrative example
Thirty bets of KES 100 at decimal odds of 3.00. Total staked: 30 × 100 = KES 3,000. A winning bet returns 100 × 3.00 = KES 300.
With 10 wins: 10 × 300 = KES 3,000, a result of zero. That is 10 ÷ 30 = 33.33%, which equals 1 ÷ 3.00. With 9 wins the return is KES 2,700, a loss of KES 300. With 11 wins it is KES 3,300, a gain of KES 300.
Break-even win rates at common prices
The table applies the formula to ten prices. The fourth column converts the rate into wins over a series of 100 bets of KES 100 (KES 10,000 staked in total). Where the exact figure is not a whole number it is rounded up, because a win cannot be split.
| Decimal odds | Fractional | Break-even win rate | Wins needed per 100 bets | Return per winning KES 100 bet |
|---|---|---|---|---|
| 1.25 | 1/4 | 80.00% | 80 | KES 125 |
| 1.50 | 1/2 | 66.67% | 67 | KES 150 |
| 1.80 | 4/5 | 55.56% | 56 | KES 180 |
| 1.90 | 9/10 | 52.63% | 53 | KES 190 |
| 2.00 | 1/1 | 50.00% | 50 | KES 200 |
| 2.50 | 3/2 | 40.00% | 40 | KES 250 |
| 3.00 | 2/1 | 33.33% | 34 | KES 300 |
| 4.00 | 3/1 | 25.00% | 25 | KES 400 |
| 5.00 | 4/1 | 20.00% | 20 | KES 500 |
| 10.00 | 9/1 | 10.00% | 10 | KES 1,000 |
The pattern is a trade. Each step up in odds lowers the hurdle, but a bet that wins only 1 time in 10 at 10.00 spends most of its life losing. Nothing in the table says how likely the event really is; it only states what the price needs.
Compare the hurdle with the real chance
The break-even rate is only useful next to a second figure: your estimate of the true probability, p. If p is below the hurdle, the bet costs money on average. The size of that cost per shilling is p × D − 1, the quantity worked through in the guide on the expected value of a bet.
Illustrative example
Two evenly matched sides, so the true chance of each result is 50%. The price offered is 1.90 on each side. The hurdle is 1 ÷ 1.90 = 52.63%, so the bet needs 2.63 percentage points more than the event delivers.
On KES 100: 0.50 × 190 = KES 95 returned on average, a cost of KES 5 per bet. Over 100 bets of KES 100 that is KES 500 out of KES 10,000 staked. At a fair 2.00 the hurdle and the chance would match at 50% and the average cost would be zero.
The gap between the hurdle and the real chance is the bookmaker's margin showing up in one bet. Where it comes from, and how to measure it across a whole market, is set out in how to calculate the bookmaker's margin.
What a break-even bettor still experiences
Suppose, for argument, that a bet at 5.00 wins exactly as often as its hurdle, 20%. That bettor has zero long-run result but loses 4 bets in every 5 on average. The chance of ten losses in a row in a given set of ten bets is 0.80 × 0.80 and so on ten times: 0.8010 = 0.1074, about 10.7%.
So even a perfectly neutral price produces long dry spells, and a price carrying a margin produces them more often. The guide on variance and losing streaks in betting explains how wide those swings are before they settle toward the average.
Working it out in four steps
- Write the price as decimal. For 5/2 use 5 ÷ 2 + 1 = 3.50. The odds converter does this for fractional and American prices.
- Divide 1 by it. 1 ÷ 3.50 = 0.2857, or 28.57%.
- Turn it into a count. 28.57% is about 2 wins in 7 bets. Expressing it as "one in N" (N = D) is the quick form: at 3.50, one in 3.5.
- Set it against your own estimate. If you honestly think the event happens less often than that, the price is working against you before the bet is placed.
Common slips
- Confusing the hurdle with the odds of winning. A 25% break-even rate at 4.00 does not mean the bet wins 25% of the time.
- Counting profit as return. At 2.00 the return on KES 100 is KES 200 and the profit is KES 100. Only the profit is new money.
- Judging by a short run. Wins at or above the hurdle over ten or twenty bets say very little, as the losing-run arithmetic above shows.
- Forgetting the margin sits inside the hurdle. The hurdle at 1.90 is already above 50%, so an even-chance event cannot reach it on average. The idea behind this is covered in the note on what overround means.
Where the law touches the arithmetic
Regulation 23(1) of the Gambling Control (Conduct of Gambling Operations) Regulations, 2026 requires a licensee to clearly display betting odds, rules and terms applicable to each bet, so the price you divide into 1 should be visible before you stake. Section 71(1) of the Gambling Control Act, 2025 sets the smallest online bet at twenty shillings, so every figure above scales from a stake of at least KES 20; see the note on the minimum online stake.
The break-even rate is a way of seeing the cost of a bet, not a way of finding winners. Play responsibly, and set a limit on what you stake before you start.
Questions and answers
What win rate do I need at odds of 2.00?
50%. A bet at 2.00 returns twice the stake when it wins, so one win in every two bets returns exactly what was staked across the pair. Fewer wins than that means a net loss.
Is the break-even win rate the same as the chance the bet will win?
No. It is the win rate the price demands, worked out from the odds alone. The real chance of the event is a separate figure, and with a bookmaker's margin in the price it is usually lower than the break-even rate.
How do I get the break-even rate from fractional odds?
For a/b, divide b by the sum of a and b. Odds of 5/2 give 2 / (5 + 2) = 2 / 7, which is 28.57%. This is the same as converting to decimal (3.50) and dividing 1 by it.
Why do high odds need such a low win rate and still lose money?
A low break-even rate only means wins can be rare. The real chance of the event is also low, and it is judged against the break-even rate, not against how small that rate looks. Long odds also produce long runs of losses, which is hard on a budget.
Does a bet that reaches its break-even win rate make a profit?
No. Reaching it means the total returned equals the total staked, so the result is zero. Profit needs a win rate above the break-even figure.