Pascal's Panel #13: Four-fold on favourites

This week on the Panel

Welcome to the Panel, where every betting idea gets a proper test. Two strategies in front of the Professor this week, both run over real races with the method laid bare, so you can see for yourself how each one holds up.

First, Pascal, the eternal optimist. He after a run of close calls, Pascal is chasing a big return this week, so this week he is stacking favourites into an accumulator: Four-fold on favourites, Four favourites in an accumulator. He is certain he has cracked it. He usually is, and that is exactly why his ideas are worth testing.

Second, a new idea. It comes from Rishi, one of our AI punter personas: the AI-truster, the sort who would back the computer's number one, backed blind. The persona is modelled, not a real reader (more on that just below), but the strategy it suggests is a genuine, testable one: Follow the AI, back the computer's top pick in every race.

The Professor tested both the only way that counts: against 26,839 real British races, flat stakes to Starting Price, with the fallers and pulled-up horses counted honestly as losses. Here is how each one actually performs.

Pascal

the eternal optimist

−31.3p / £1

This week's pick Four-fold on favourites

“Each favourite wins more often than it loses, so stack four of them together and the returns balloon while the risk barely shifts. One tidy Saturday slip and a tenner becomes a proper payday.”

Meet Pascal

Pascal is the Lab's resident optimist and Professor Furlong's foil, certain every single week that he has finally cracked it. He turns up with a fresh angle, usually an accumulator or a way to chase last week's loss back, pitches it with total confidence, and watches the Professor test it properly on real races. He never wins and never learns, and he is the warmest reason to keep coming back.

See the full experiment →

Rishi

the AI-truster

−11.4p / £1

This week's pick Follow the AI

“It is an AI, so the computer must know more than I do. If I just back its number one in every race and let the maths do the work, surely a clever model beats a mug punter squinting at the form.”

Meet Rishi

Rishi trusts the machine over any human hunch, so he follows the model's most confident picks, sure that a clever computer must know. The honest twist is that our own model, backed blind, loses about as much as the favourite, because naming the likely winner is not the same as beating the price.

Their tell: the computer's number one, backed blind

See the full experiment →

Professor Furlong's verdict

On Pascal's pick It loses, and heavily: four favourites in an accumulator comes out at -31.30% to Starting Price, so for every £100 staked you get back only about £68.70. Here is why. A favourite is no near-certainty, it wins barely a third of the time, and even backed on its own to SP it already loses 9.0p in the pound because the bookmaker's margin is baked into the price. When you multiply four of those prices together you multiply that margin four times over too, so the four-fold carries a far bigger built-in edge against you than any single leg. All four favourites must win or the whole slip is dead, and the run of days where just one gets turned over swamps the rare day they all land, which is exactly why the loss is this deep.

On Rishi's pick Honest answer is no. Back my top pick blind in every race and you lose money: over 6,601 real bets to starting price the return is -11.36%, so for every £100 staked you get back about £88.64, a little worse than simply backing the favourite, which now loses about 8.96%. Here is why, and it has nothing to do with the model being clever. My top pick is usually the favourite or close to it, and the bookmaker has already shaved value off those short prices to bake in their margin, the over-round. Naming the most likely winner is not the same as being paid enough when it wins. I can rank horses better than a coin toss, but I am not handing you odds bigger than the true risk, so the house edge grinds the bankroll down race after race. Worth knowing too: this is mostly a jumps card, where fallers and pulled-up horses cost you full stakes on top of the margin.

The standings

Both of this week's strategies sit on the leaderboard already. Here is where everything we have tested stands right now, ranked by real-world return, the least-negative at the top. Every system is a net loss, so the top row is the best real performer, not a winner.

24 systems tested · 29,483real 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)	£/week	Sample	Strike	£200/week → today
1	The AI's most-confident picksThe Professor	Guest: RishiAI persona, modelled	−0.8p	−£2	361	—	—
2	Odds-on favourites onlyThe systems	Guest: BarryAI persona, modelled	−5.1p	−£10	4,942	wins 60%	−£945
3	Favourite in a big fieldThe systems	Tested system	−5.7p	−£11	662	wins 24%	—
4	Favourite in a small fieldThe systems	Guest: CarolAI persona, modelled	−7.7p	−£15	4,173	wins 46%	—
5	Festival favouritesThe systems	Tested system	−7.8p	−£16	2,297	wins 32%	—
6	Favourite over jumpsThe systems	Tested system	−7.9p	−£16	10,446	wins 37%	−£2,021
7	Each-way the favouriteThe basics	Guest: EddieAI persona, modelled	−8.7p	−£17	29,483	places 62%	—
8	Favourite on soft groundThe systems	Guest: CarolAI persona, modelled	−8.9p	−£18	10,993	wins 35%	—
9	Back the favouriteThe basics	Tested system	−9.0p	−£18	29,483	wins 35%	−£1,780
10	Favourite in handicapsThe systems	Tested system	−9.1p	−£18	19,737	wins 31%	—
11	Favourite on the FlatThe systems	Tested system	−9.5p	−£19	19,037	wins 34%	—
12	Follow the AIThe Professor	Guest: RishiAI persona, modelled	−11.4p	−£23	6,601	—	—
13	Back the second favouriteThe basics	Guest: SandraAI persona, modelled	−12.0p	−£24	29,481	wins 21%	−£3,114
14	Top-rated horseThe basics	Guest: FrankAI persona, modelled	−15.4p	−£31	26,121	wins 19%	−£3,790
15	Top-rated in handicapsThe systems	Guest: FrankAI persona, modelled	−16.0p	−£32	19,719	wins 16%	—
16	Double on two favouritesThe multiples	Pascal, the punter	−17.1p	−£34	—	—	—
17	Lucky 15 on favouritesThe multiples	Pascal, the punter	−17.9p	−£36	—	—	—
18	A random horseThe basics	Guest: DawnAI persona, modelled	−21.4p	−£43	29,483	wins 13%	−£5,225
19	Back the lowest drawThe systems	Guest: StanAI persona, modelled	−25.2p	−£50	19,015	wins 12%	—
20	Back the old stagersThe systems	Guest: CarolAI persona, modelled	−26.7p	−£53	8,821	wins 9%	—
21	Each-way an outsiderThe basics	Tested system	−30.2p	−£60	27,714	wins 7%	—
22	Four-fold on favouritesThe multiples	Pascal, the punter	−31.3p	−£63	—	—	—
23	Back the outsiderThe basics	Guest: LornaAI persona, modelled	−34.9p	−£70	29,483	wins 3%	−£5,514
24	Four-fold on random horsesThe multiples	Pascal, the punter	−61.8p	−£124	—	—	—

24 systems tested · 26,839 real races · none beat the bookmaker's margin. The best real return is -4p in the pound. The bookmaker's margin, the over-round, the slice baked into every price so the odds add up to more than 100%, sits in all of them, and no selection rule we have tested clears it. That is the genuinely useful part: the leaderboard shows exactly where the edge goes.

See the full Betting Systems Leaderboard →

What we're testing next week

Pascal will be Pascal will be looking for a long shot next time. And another of the personas has floated our next guest considers trainer form, and we will run it the same way, against real races, with the method shown and the real result reported.

If you take one thing from the Panel, take this: picking a likely winner is the easy part. Getting paid more than the true risk is the hard part, and the over-round is what stands in the way. That is the insight every row on the leaderboard is really measuring. Bet for fun, with money you can afford to lose, and read our track record for whether anyone, us included, holds a real edge at all.

More Pascal's Panel issues

Week 12 · 14 Sept 2026

Carol vs Pascal

Week 11 · 7 Sept 2026

Dawn vs Pascal

Week 10 · 31 Aug 2026

Frank vs Pascal