The AI Lab
Every betting system, tested
We ran every common betting system through 27,909 real GB races, settled honestly to the industry starting price, with fallers and pulled-up horses counted as the losers they are. The short version: none of them makes a profit. Below is every test we have run, grouped by type. Open any one for the full story, the data, and the verdict.
Prefer it interactive? The Pascal vs the Professor league table ranks them all by what they cost. New to odds? What a price really means.
Staking systems (2)
Does the Martingale system work?
No — Martingale doesn't beat the odds. It just delays the wipe-out.
Read the full analysis →Should you bet more when you're confident?
No — the model's 'confidence' doesn't predict profit, so sizing your stakes by it doesn't help.
Read the full analysis →Backing & laying strategies (12)
Are the AI's most-confident picks profitable?
No, the AI's most-confident picks are not profitable. They are the least-bad bet on the board and still a loss, just one slow enough to hide inside the noise.
Read the full analysis →Do favourites win in big fields?
No, big fields do not rescue the favourite: it is a bigger price and wins less often, yet still loses 11.53%, almost identical to small fields, because the leak is the margin, not the size of the field.
Read the full analysis →Do favourites win in handicaps?
No, the handicap favourite loses 12.4p in every pound, because the handicapper builds the race to weight that favourite back towards the pack and the bookmaker still takes its cut on top.
Read the full analysis →Do favourites win in small fields?
No, favourites in small fields do not make money: they are shorter and win more often, then hand that higher strike rate straight back in a price that is too short, for a 12.68% loss.
Read the full analysis →Do Flat favourites make money?
No, backing every Flat favourite loses 11.26% to Starting Price, because an efficient market plus a built-in margin beats you even when you back the right horse.
Read the full analysis →Do jumps favourites make money?
No, the jumps favourite is the worst favourite bet of the lot, not a profit, losing 14.4p in every pound because so many jumpers fall or pull up and you lose those stakes in full.
Read the full analysis →Do odds-on favourites make money?
No, odds-on favourites are the least-bad bet on the whole board and still lose about 7p in the pound.
Read the full analysis →Does backing the favourite make money?
No, backing the favourite does not make money: it wins most often, then hands the bookmaker's margin straight back, for a 12.48% loss.
Read the full analysis →Does backing the second favourite work?
No, backing the second favourite loses 15.7p in every pound and bleeds faster than backing the favourite, not slower.
Read the full analysis →Does backing the top-rated horse make money?
No, backing the top-rated horse loses 19.3% to SP, more than backing the favourite or the second favourite, because the official rating is already in the price.
Read the full analysis →Does following an AI tipster work?
No, following an AI's top pick blind loses about 11% to SP, much the same leak as backing the favourite, because the market has already priced in everything the model knows.
Read the full analysis →Does backing the top-rated horse in handicaps make money?
No, backing the top-rated horse in handicaps loses about a fifth of your turnover, one of the heaviest leaks on this page.
Read the full analysis →Bet types (7)
Does each-way betting actually pay in the big-field handicaps?
Yes — with the right place terms (6+ places at 1/4 odds) and the right races, each-way at the big handicaps is one of the few bets where the maths is on your side.
Read the full analysis →Do accumulators pay?
No, a four-fold on favourites loses 41p in every pound, far more than backing those same favourites singly, because the bookmaker's margin is charged four times over.
Read the full analysis →Do doubles on two favourites work?
No, a double on two favourites is the worst version of a favourite bet, because it pays the bookmaker's margin twice and loses 23.4% to Starting Price.
Read the full analysis →Do random accumulators ever win?
No, the random four-fold is the heaviest loser on the whole board, handing back barely a third of your stake because the bookmaker's margin gets multiplied four times over.
Read the full analysis →Does the Lucky 15 work?
No, a Lucky 15 on favourites loses 24.23% to SP, roughly twice the bleed of backing the same four favourites as singles, because it charges you the bookmaker's margin fifteen times over.
Read the full analysis →Is each-way an outsider worth it?
No, each-waying an outsider is one of the heaviest losers on the whole page, because you pay the bookmaker's biggest margin twice over.
Read the full analysis →Is each-way betting on the favourite good value?
No, each-way the favourite is two losing bets stapled together, and at -12.36% to SP it loses about the same as backing that favourite to win.
Read the full analysis →Meeting-specific (1)
This is research, not tips. The numbers describe what each system has done over a large, honest sample; they are not a signal to stake. The model is a calibrated read on a race and does not beat the market. 18+, please bet responsibly.