Do odds-on favourites make money?

Pascal only touches the bankers, the odds-on favourites the whole world fancies. His logic is simple, and it is the logic most punters reach for. If a horse is shorter than even money, the market is screaming that it cannot lose, so back it and treat it like money in the bank. A price of 4/6 or 8/13 looks like the bookmaker waving a flag that says safe. The crowd has piled in, the form reads well, the horse is clear of the field on the racecard. What could go wrong with siding with the one the market is this sure about?

It is the most natural instinct in betting. We are taught that the favourite is favourite for a reason, and an odds-on favourite is that reason turned up to full volume. These horses do win plenty. Three in five odds-on favourites oblige, which is a far higher hit-rate than any other angle on the board. Sit and watch them come in, weekend after weekend, and the bet feels less like gambling and more like a savings account.

That comfort is exactly the claim we set out to test. Not whether odds-on favourites win often, because they plainly do, but whether backing every single one of them actually makes money. Those are two completely different questions, and the gap between them is where the bookmaker lives. A horse can win most of the time and still cost you money over a season, if the price you took was never quite big enough to pay for the times it lost. So we ran the bankers blind, every odds-on favourite, flat stakes, to the official Starting Price, across thousands of real British races, and let the arithmetic answer Pascal honestly.

Winning feels like proof, and odds-on favourites win a lot. That is the whole hook. When the short-priced banker keeps obliging, the system looks like it is working, and a punter rarely sits down afterwards to tot up that the small returns on the winners are quietly outweighed by the near-evens losers. The wins are frequent and vivid, the losses are forgotten, and the running total never gets the cold audit it needs.

There is a deeper comfort underneath the numbers. Backing the horse everyone expects to win feels sensible rather than reckless. You are siding with the wisdom of the crowd, not chancing your arm on a hunch, and that emotional cover is worth a lot to most punters. Nobody feels foolish backing the 4/6 shot that the whole paddock fancies, especially when three in every five of them oblige. If it loses, well, so did everyone else, and that shared disappointment is far easier to swallow than a busted longshot you picked alone.

The losing runs help the illusion along, because they stay short. An odds-on favourite that gets turned over rarely starts a long barren streak the way a longshot system does. A couple of beaten bankers read as bad luck, the next one wins, and the curve looks like it has recovered. So the bad patches register as variance, the noise around a sound plan, rather than as the plan itself quietly carrying a cost.

And a good weekend cements all of it. Land four or five short-priced winners in a row and the belief hardens into something close to faith. The word banker does its quiet work, telling you the bet is as good as money in hand. What that word never tells you is whether the odds were ever big enough to cover the days the banker breaks, and that is the only question that decides whether you finish ahead.

Here is the interesting part: the whole result comes down to one small, steady cost, −4.6p in every pound, and once you can see it you can see exactly where it comes from. There is no spectacular blow-up, no losing run long enough to set off alarms. The cost is gentle and consistent, and it adds up the longer you play and the bigger you stake. Back odds-on favourites at 10 pounds a bet across a full season of qualifiers and that small margin works away quietly in the background, because the favourite wins more often than it loses and the bad runs stay mercifully short.

The mechanism is the overround, the bookmaker's built-in margin. Add up the chances implied by every price in a race and they come to more than 100%, about 12% per race extra on a typical British field, and that surplus is the house edge you pay on every single bet. An odds-on favourite is genuinely probable, so the layer only needs to shade its price a fraction below its true chance to keep itself in front, and the short prices are exactly where the public money piles in, so it can shade and still take every bet it wants. You are backing real winners at prices clipped just enough to keep you on the wrong side of break-even.

The strike rate cannot save you, and this is the heart of it. Three in five of these horses win, but at odds-on each winner pays you a pittance, while the two in five that get beaten still cost you the full stake. A horse that falls or is pulled up loses the lot the same as any other. Tot it up and the small returns never cover the losers plus the margin shaved off every winner. A high hit-rate does not make a system profitable. The price does, and the price is fractionally short every single time.

We took every odds-on favourite, that is every horse returned at shorter than even money, and backed it blind to the official Starting Price across 27,421 real British races. No cherry-picking the good days, no skipping the ones that looked dodgy on paper, no shopping around for a better price. The rule was mechanical: if the horse went off odds-on, it went in the sample. That left 4,547 qualifying bets, a serious sample rather than a handful of feel-good results.

Stakes were flat, one level unit on every selection, because that is the only honest way to measure whether a selection method makes money. Staking plans like doubling up after a loss change how a loss arrives, never whether it arrives, so layering one on top would only have hidden the underlying edge. We wanted the edge itself, naked.

The settling is where most rosy betting-system numbers fall down, and where we were deliberately strict. Fallers and pulled-up horses are counted as the losing bets they are, because a horse that does not complete still costs you your full stake. Earlier versions of this kind of analysis quietly dropped those non-finishers, which flattered every favourite figure and once even conjured a fake profit for jumps favourites. We do not drop them. Joint-favourites were split so no result leaks in and inflates the count.

Everything is measured to industry Starting Price with no commission and no allowance for the price drifting against you before the off. That makes the test generous to the system, not harsh, because a real punter taking real prices and paying exchange commission bleeds a touch faster than the headline. So if anything the true cost is worse than what we report. The figure that came back is the cleanest, most flattering version of backing odds-on favourites that honest accounting allows.

Backing every odds-on favourite returned -4.64% to Starting Price across 4,547 real British races. In plain money, put £100 through the system and you are left with about £95.36 over the long run, a leak of −4.6p in every pound you bet. The 95% range on the sample is [-7.0,-2.4], so even on the kindest reading the result sits firmly in the red. There is no version of this sample where the long-run figure turns positive.

The bankers wins 60% of the time. Three in five of them oblige, which is comfortably the highest hit-rate of any system we tested, and it is exactly the number that fools people. Read it on its own and the bet looks unstoppable. But a three-in-five strike rate at odds-on is not enough, because at those short prices each winner pays you so little that the two in five that get beaten, every one costing the full stake, drag the whole thing under. The arithmetic is settled and it does not care how often you win.

What makes this result worth dwelling on is its place on the board. This is the bet that holds up best of all 24 systems we measured. Backing the favourite blind across all races returns -8.7%. The favourite over jumps, once wrongly thought to be the one profitable angle, comes in at -7.8% once you count the fallers, no exception to the rule. Chasing outsiders sheds -34.5%. Against that company, odds-on favourites at -4.6% genuinely hold up best on the page.

And that is the whole lesson in one number. The bet that holds up best, the highest strike rate, the shortest prices, the safest-feeling wager in racing, still leaks −4.6p in every pound. Short prices mean a smaller margin, not no margin. If even the bankers cannot break even, the board has no winner on it anywhere, and 0 of the 24 systems we tested make a profit.

So the honest answer is no. Odds-on favourites only is the best of a losing bunch and still a loser: -4.64% to Starting Price across 4,547 real British races, flat stakes, with fallers and pulled-up horses counted as the losing bets they are. It suits nobody as a money-maker. At most it is the gentlest way to lose if you are determined to back short-priced favourites anyway, which is damage limitation, not a strategy.

This is the most important page on the whole board to read slowly, because it is the one that comes closest to working and still does not. The strike rate is seductive, the comfort is real, and a couple of good weekends will tell you the system is sound. But the arithmetic is settled and it has been settled for a long time. The overround, about 12% per race, is baked into every price, and a high hit-rate cannot overcome a price that is fractionally short every single time. Past results are not a forecast.

The genuinely useful finding is the favourite-longshot bias in plain sight. Short prices lose least, long prices lose most, and somewhere along that line is the bet that holds up best. We found it. It is odds-on favourites, and even that one still leaks −4.6p in every pound. If the bet that holds up best on the board loses, every other angle loses harder, which is exactly why 0 of the 24 systems we tested make a profit.

The straight takeaway is the one the whole Lab keeps proving. Naming the most likely winner is not the same as being paid enough when it wins, and no selection method, however safe it feels, removes the tax built into the odds. It just hides it. Treat the bankers as proof of how hard the market is to beat, never as a system to follow. This is not a way to beat the bookies.