In betting, your strike rate is the share of your settled bets that win, and the break-even strike rate for a price is 1 / decimal odds. With every bet at $3.00 you need 33.3% winners just to get your stakes back; at $1.50 you need 66.7%, and at $11.00 only 9.1%. With mixed prices at level stakes, the figure is 1 / the average of your winners' prices.
That is why a strike rate means little without the prices beside it. A punter who wins 68% of their bets can lose money at short prices, while one who wins 14% can make money at long ones. Over a few hundred bets, luck alone moves a strike rate several points either way.
What strike rate means and what counts in it
Strike rate = winners / settled bets x 100. Win rate and winning percentage mean the same thing. In Australian racing the term also describes a horse, jockey or trainer, as wins from starts. That figure carries no prices, so it cannot tell you whether backing them made money.
A strike rate only compares fairly when every record counts the same way:
| Result | How to count it | Why |
|---|---|---|
| Void, refund, or a scratching that is refunded | Leave it out | The stake came back, so the bet was never decided |
| All-in (futures) bet on a runner that is scratched, with no refund under the bet's terms | A loser | The stake was lost; check the bookmaker's futures terms |
| Dead heat for first | A winner, with the reduced return in your profit | It won, but dead heat rules cut the payout, and for an online bet those are the bookmaker's own terms |
| Each way bet | Two bets: the win half and the place half | Each half settles on its own |
| Multi | One bet | One stake and one result, however many legs |
| Bonus bet | A separate line | A winning bonus bet usually pays the winnings only, not a stake; the bookmaker's terms decide |
The break-even strike rate: 1 / odds
A winning bet returns the price per $1 staked, stake included, so your winners have to return as much as every stake put together. That happens when the share of bets that win equals 1 / decimal odds, the break-even strike rate.
Example: Illustrative prices: 150 bets of $20, every one at $3.00. Turnover = 150 x $20 = $3,000, and each winner returns $20 x 3.00 = $60. Break-even = 1 / 3.00 = 33.3%, which is 150 / 3 = 50 winners, and 50 x $60 = $3,000.
| Winners from 150 | Strike rate | Returned | Result |
|---|---|---|---|
| 49 | 32.7% | $2,940 | -$60 |
| 50 | 33.3% | $3,000 | $0 |
| 51 | 34.0% | $3,060 | +$60 |
Each winner either way moves the result by $60, which is 2% of turnover. The break-even figure is the price's implied probability under another name, so the odds converter doubles as a break-even strike rate calculator: type any price and read its implied probability.
To finish ahead rather than level, the strike rate needed for a target yield = (1 + target yield) / price. At illustrative prices:
| Price | Break-even strike rate: 1 / price | Strike rate for a 5% yield: 1.05 / price | Extra winners per 100 bets that the 5% takes |
|---|---|---|---|
| $1.50 | 66.7% | 70.0% | 3.33 |
| $1.90 | 52.6% | 55.3% | 2.63 |
| $2.50 | 40.0% | 42.0% | 2.00 |
| $3.00 | 33.3% | 35.0% | 1.67 |
| $4.20 | 23.8% | 25.0% | 1.19 |
| $6.00 | 16.7% | 17.5% | 0.83 |
| $11.00 | 9.1% | 9.5% | 0.45 |
The last column is 5 / price. At $11.00 a 5% yield takes 0.45 of an extra winner per 100 bets, less than half a winner, which is why records at long prices take so many bets to read.
Average odds: judge a record by its average winning price
Needing 33.3% at average odds of $3.00 holds when every bet is at $3.00. Most records mix prices, and then the price that matters is the average of your winners' prices. At level stakes:
- yield = strike rate x average winning price - 1
- break-even strike rate = 1 / average winning price
Shorter prices win more often, so winners usually average a shorter price than the bets as a whole, and the average price of all your bets flatters the record.
Example: A hypothetical record of 120 bets of $10 each, at three prices.
| Price | Bets | Winners | Returned |
|---|---|---|---|
| $1.70 | 40 | 22 | 22 x $17.00 = $374.00 |
| $3.10 | 50 | 15 | 15 x $31.00 = $465.00 |
| $5.50 | 30 | 5 | 5 x $55.00 = $275.00 |
| All | 120 | 42 | $1,114.00 |
The strike rate is 42 / 120 = 35.0%. The 120 bets average (40 x 1.70 + 50 x 3.10 + 30 x 5.50) / 120 = $3.233. Its break-even is 1 / 3.233 = 30.9%, so the record looks 4.1 points ahead. The 42 winners average (22 x 1.70 + 15 x 3.10 + 5 x 5.50) / 42 = 111.40 / 42 = $2.652, which needs 1 / 2.652 = 37.7%.
So the record is 2.7 points behind: $1,114 back from $1,200 staked is -$86, a yield of -7.2% (0.35 x 2.652 - 1 = -0.072).
When stakes vary, weight each winner by its stake: yield = total returned by winners / total staked - 1. Before any results are in, the yardstick for mixed prices is the average of each price's implied probability, not the implied probability of the average price.
Strike rate vs ROI: why a high strike rate can lose money
Strike rate counts how often you win; yield counts what winning paid. Yield is profit / turnover, and many betting records call it ROI, which ROI vs yield untangles. Four illustrative records at level stakes:
| Record | Strike rate | Average winning price | Break-even: 1 / average winning price | Yield: strike rate x price - 1 |
|---|---|---|---|---|
| A | 68% | $1.42 | 70.4% | -3.4% |
| B | 52% | $1.88 | 53.2% | -2.2% |
| C | 31% | $3.48 | 28.7% | +7.9% |
| D | 14% | $7.60 | 13.2% | +6.4% |
Record A wins more than twice as often as record C and still loses money, because 68% falls short of the 70.4% its prices need. The result depends on the gap between a strike rate and its own break-even figure, never on the strike rate alone.
A bookmaker's margin sits inside every price. If a price carries a margin and is otherwise accurate, the strike rate you can expect at it sits below its break-even figure, however high that strike rate is. Price, not strike rate, is where any betting strategy finds its edge.
Risk: Betting involves risk. Records C and D are hypothetical, and a record above its break-even figure can drop below it over the next few hundred bets. See responsible gambling for limits and support.
How many bets before a strike rate means anything
Luck moves a strike rate by a standard error of square root of (p x (1 - p) / bets), where p is the chance each bet wins. For a punter with no edge at $3.00, p = 1 / 3:
| Bets | Standard error | Range luck allows about 19 times in 20 (2 standard errors) |
|---|---|---|
| 50 | 6.7 points | About 20% to 47% |
| 100 | 4.7 points | About 24% to 43% |
| 250 | 3.0 points | About 27% to 39% |
| 500 | 2.1 points | About 29% to 38% |
| 1,000 | 1.5 points | About 30% to 36% |
| 2,500 | 0.9 points | About 31% to 35% |
The working for 100 bets: square root of (0.3333 x 0.6667 / 100) = 0.0471, so 33.3% plus or minus 2 x 4.7 points runs from about 24% to 43%. Over those 100 bets the no-edge punter reaches 36% or better, an 8% yield (0.36 x 3.00 - 1 = 0.08), about 32% of the time (binomial distribution).
A real 8% edge at $3.00 lifts the expected strike rate from 33.3% to 1.08 / 3.00 = 36.0%, the mark the no-edge punter reaches by luck about 32% of the time over 100 bets. The gap is 0.08 / 3 = 2.67 points, and for it to stand two standard errors clear takes bets = 4 x (1/3) x (2/3) / (0.08 / 3) squared = 1,250. Sample size in betting gets the same count from its yield formula and shows why closing line value gives a verdict sooner.
Tipster strike rates in context
A strike rate is the easiest number for a tipster to make look good. Before reading anything into one, check five things:
- The average winning price, or the price of every tip, so you can work out break-even.
- The number of tips, counted from the first, with none dropped.
- Whether placings count as strikes, or only winners.
- Whether the prices are ones you could have taken after the tip was published.
- Results at level stakes, so a few large bets cannot carry the record.
Two hypothetical records, at one unit a tip:
| Tipster | Tips | Strike rate | Average winning price | Break-even: 1 / price | Yield |
|---|---|---|---|---|---|
| Tipster 1 | 200 | 71.0% (142 winners) | $1.38 | 72.5% | -2.0% |
| Tipster 2 | 300 | 19.0% (57 winners) | $6.10 | 16.4% | +15.9% |
Tipster 1's winners return 142 x 1.38 = 195.96 units on 200 staked, and Tipster 2's return 57 x 6.10 = 347.70 units on 300. The higher strike rate loses money.
The profitable record is too small for its prices. Treat every tip as a no-edge bet at $6.10, and luck alone gives 57 or more winners from 300 about 13% of the time, roughly 1 in 8. Checking a tipster's record covers the closing-price test that separates skill from luck.
Risk: Betting involves risk. A strike rate, a yield and a tipster's record all describe bets already settled, and past results are no guarantee of future results. See responsible gambling for limits and support.
Counting errors that distort a strike rate
| Mistake | Effect on the figure |
|---|---|
| Counting refunds as losers | 30 winners from 90 settled bets at $3.00 is break-even (33.3%), but with 10 refunds counted as bets it reads 30 from 100, or 30% |
| Counting placings as strikes | 18 winners and 27 more placings from 100 win bets reads as 45%, when the win strike rate is 18% |
| Judging a record by strike rate alone | Record A's 68% loses 3.4% of turnover |
| Using the average price of all bets | The 120-bet record looks 4.1 points ahead and is 2.7 points behind |
| Moving to shorter prices to lift the strike rate | Going from $3.00 to $1.50 doubles the strike rate needed, from 33.3% to 66.7%, and the margin is still in every price |
Lowering the strike rate a bet needs
The price you take sets the strike rate you need: the same bet at $3.30 instead of $3.00 needs 1 / 3.30 = 30.3% winners instead of 33.3%. B337's Terminal sets out racing and sports prices from 40+ bookmakers side by side, one column per bookmaker, with exchange back and lay prices beside them. It picks out the best available price on each runner or selection, which is also the price with the lowest break-even strike rate.
A free account opens a limited view of the Terminal with live odds, and the odds comparison page shows how the board is laid out. The board fetches prices in repeated sweeps, so a price you see may already have moved: check it in your own bookmaker account before you place the bet.
Risk: Betting involves risk. A higher price lowers the strike rate a bet needs to break even, but it cannot make the bet win, and there is no guarantee of profit. See responsible gambling for limits and support.