Virtual Horse Racing Odds and RTP
Every price in the game comes from one formula. This page shows the formula, the margin on each market and the test that checks both.
What RTP means
Return to player (RTP) is the share of all stakes that comes back as payouts over a very long run. With an RTP of 92%, every 100 staked returns 92 on average and the house keeps 8.
RTP says nothing about one race or one evening. Short runs swing far above and below it. It describes what happens over thousands of bets.
How virtual horse racing odds are calculated
odds = RTP ÷ probability of the outcome
The probability comes from the Harville model explained on the how it works page. Take a horse with a 25% chance in the win market, where RTP is 92%: 0.92 ÷ 0.25 = 3.68. A fair price with no margin would be 4.00.
Rounding
Odds always round down: in steps of 0.01 below 10, 0.1 from 10 to 100, and 1 above 100. Rounding costs the player between 0.1% and 0.4% of RTP, which is why measured figures sit a little under the targets.
Minimum and maximum odds
A selection priced under 1.02 is not offered. Each market also has a ceiling: Win 50, Place 20, Show 10, Forecast 1,000, Quinella 500, Tricast 5,000 and Trio 1,000. A single bet pays at most 100,000.
Virtual horse racing RTP by market
| Market | Target RTP | Measured RTP | Selections per race |
|---|---|---|---|
| Win | 92% | 91.99% | 8 |
| Place (top 2) | 92% | 91.86% | 8 |
| Show (top 3) | 92% | 91.81% | 8 |
| Lay | 92% | 91.67% | about 4 |
| Forecast | 88% | 88.02% | 56 |
| Quinella | 88% | 88.00% | 28 |
| Tricast | 85% | 84.27% | 336 |
| Trio | 85% | 84.49% | 56 |
| Odd/Even | 93% | 92.90% | 2 |
| Low/High | 93% | 92.93% | 2 |
| Top-3 total | 93% | 92.75% | 2 |
Measured RTP comes from 100,000 simulated races with one credit on every selection in every market.
Why exotic bets carry a bigger margin
Forecast and tricast prices run high, and the ceiling trims the longest of them. Commercial suppliers price exotic markets with wider margins for the same reason, and we follow that practice.
About 3% of tricast combinations hit the 5,000 ceiling. Their real return sits below 85%, which pulls a randomly chosen tricast down to about 84%. A tricast on one of the likelier combinations returns close to the full 85%.
How we tested the odds
We ran 100,000 races through the same code the server uses. For every market we compared how often selections won with how often the model said they would. The ratio came out at 1.000 for all 11 markets: outcomes land as often as the prices assume.
The same run measured the field itself. The favourite won 28.3% of races. Its median win price was 3.43, and the outsider's was 23.4.
What a session looks like
We also simulated 3,000 players for each of four betting styles. Each started with 1,000 credits and played 300 races, about 5.4 hours of play.
| Style | RTP | Median balance after 300 races | Finished ahead | Ran out of credits |
|---|---|---|---|---|
| 10 a race, mixed markets | 89.7% | 427 | 24% | 26% |
| 20 a race on the favourite | 91.3% | 466 | 16% | 28% |
| 5 a race on tricasts | 83.1% | 267 | 28% | 41% |
| 100 a race to win | 91.7% | 49 | 8% | 91% |
One casual player in four finished ahead. That comes from variance, not skill. Give the same players more races and the share keeps falling.
Fixed virtual horse racing odds
The price on your bet slip is the price you get. Odds do not drift with other players' money, unlike pool betting, and there is no starting price. Check the dividends of past races in the virtual horse racing results, or compare markets on the bet types page.