Every common betting system, tested on 26,839 real British races, ranked by real-world return.
Most betting-system advice comes from someone with something to sell. This doesn't. We ran every common system over 26,839real races, counted every result honestly, fallers and all, and published the method. Here is how each one actually performs, and why the bookmaker's margin is so hard to beat. Tap a strategy to read its full experiment.
Each week Pascal's Panel puts a new system to the test; once tested it graduates here, to the permanent record. See this week's test or browse all experiments.
24 systems tested · 26,839real races · none beat the bookmaker's margin. The best real return is −0.8pin the pound, so even the strongest system is a net loss. The bookmaker's margin is the reason.
| # | Strategy | Who proposed it | Real return (per £1) | Sample | Strike | £200/week → today |
|---|---|---|---|---|---|---|
| 1 | The AI's most-confident picksThe Professor | Tested system | −0.8p | 361 | — | — |
| 2 | Odds-on favourites onlyThe systems | Guest: BarryAI persona, modelled | −4.6p | 4,547 | wins 60% | −£762 |
| 3 | Favourite in a big fieldThe systems | Tested system | −6.0p | 605 | wins 24% | — |
| 4 | Favourite in a small fieldThe systems | Tested system | −7.4p | 3,772 | wins 46% | — |
| 5 | Favourite over jumpsThe systems | Tested system | −7.8p | 10,194 | wins 37% | −£2,115 |
| 6 | Festival favouritesThe systems | Tested system | −8.3p | 2,098 | wins 32% | — |
| 7 | Back the favouriteThe basics | Tested system | −8.7p | 27,421 | wins 35% | −£1,851 |
| 8 | Favourite on soft groundThe systems | Tested system | −8.8p | 10,672 | wins 35% | — |
| 9 | Favourite in handicapsThe systems | Tested system | −8.8p | 18,383 | wins 30% | — |
| 10 | Favourite on the FlatThe systems | Tested system | −9.3p | 17,227 | wins 34% | — |
| 11 | Follow the AIThe Professor | Tested system | −11.4p | 6,601 | — | — |
| 12 | Back the second favouriteThe basics | Guest: SandraAI persona, modelled | −12.0p | 27,419 | wins 21% | −£3,080 |
| 13 | Top-rated horseThe basics | Tested system | −15.9p | 24,222 | wins 19% | −£3,806 |
| 14 | Top-rated in handicapsThe systems | Tested system | −16.4p | 18,368 | wins 16% | — |
| 15 | Lucky 15 on favouritesThe multiples | Pascal, the punter | −17.5p | — | — | — |
| 16 | A random horseThe basics | Tested system | −21.6p | 27,421 | wins 13% | −£5,215 |
| 17 | Back the lowest drawThe systems | Tested system | −25.1p | 17,208 | wins 12% | — |
| 18 | Back the old stagersThe systems | Tested system | −27.2p | 8,571 | wins 8% | — |
| 19 | Each-way an outsiderThe basics | Tested system | −29.9p | 25,800 | wins 7% | — |
| 20 | Four-fold on favouritesThe multiples | Pascal, the punter | −30.6p | — | — | — |
| 21 | Back the outsiderThe basics | Tested system | −34.5p | 27,421 | wins 3% | −£4,338 |
| 22 | Four-fold on random horsesThe multiples | Tested system | −62.1p | — | — | — |
The bookmaker builds a margin (the over-round) into every price, so the implied chances in a race add up to more than 100% (around 12% over, on average). That slice sits inside every system here, and the favourite-longshot bias makes the longest prices the worst value of all, so no fixed selection or staking rule we have tested clears it. That is the real insight of this leaderboard: it shows precisely where the edge goes. The strategies are proposed by openly-synthetic AI personas (a way to generate punter ideas to test at scale) and the verdict is always the real race data. For whether anyone holds a real edge, see our track record. 18+, please bet responsibly.