If Nothing Works, How Am I Meant to Pick a Horse?

If you want one thing to act on, pick the race before you pick the horse. What the bookmakers charge varies far more from race to race than anything you can turn up by studying the horses inside a single one.

That is the short answer, and it is worth being plain about why it is the answer. Our own testing keeps finding that backing favourites loses, that form study does not beat the market, that staking plans do not help and that accumulators make it worse. At which point the reasonable reply is: fine, so what am I supposed to do on Saturday?

The honest answer has two halves. The first is that we cannot tell you which horse will win, and neither can anyone else who is being straight with you. The second is that quite a lot of what determines your afternoon has nothing to do with which horse you pick, and that part we can help with.

The things that are actually within your control

Which race. The charge in the biggest fields runs to nearly three times what it costs in the smallest. That is the largest single difference we have measured, it is visible on the racecard before anyone opens a form book, and it requires no judgement at all.

How many bets. The charge lands on every one. Six bets on a card costs six times what one bet costs, and the entertainment does not scale the same way.

What price you take. Taking the best available price rather than an average one is real money, and it is the one area where we have seen a genuine signal in our own testing. How some people make a living at this sets out what that looks like, and why we still will not call it proven.

What kind of bet. Singles rather than accumulators, on our numbers: -9.10% against -31.73% on the same four horses.

None of that is a tipping service. It is the difference between losing slowly and losing quickly, and over a season it is a larger number than most selection methods would move.

And the horse itself

Pick it however you enjoy picking it.

We mean that seriously. Since no method we have tested produces a profit, and since the favourite already contains the market's best guess, there is no measurable penalty for choosing a horse because you like its name, its colours or its trainer. A horse picked at random returns -21.42% against -9.10% for the favourite, so the difference is real, but it is a difference between two ways of losing.

If you want the smallest loss available on a simple method, back the favourite. If you want the most enjoyment, back the one you fancy. If you want to be able to explain the result afterwards, read the form, which is genuinely interesting and will make you better company at the rail.

What we would ask is that you do not let anyone sell you the third option as the first. Reading form is worth doing. It is not a way of beating the charge, and anybody telling you their selections are is selling something.

Five AI tipsters are working on the same problem in public on the Silicon Tipster League, and none of them has cracked it either.

A word on all of this

None of these pages is a tip, and none describes a way to win. They describe what betting costs, which is a different and more reliable subject. If your betting has stopped being fun, BeGambleAware has free, confidential help, and the National Gambling Helpline is on 0808 8020 133.

Common questions

If no system works, why bother reading form?

Because it makes racing more interesting and stops you making the worst bets. It is not a route to profit on any evidence we have.

Is backing the favourite the best simple approach?

It is the least expensive of the simple approaches we have tested, at -9.10%. It still loses.

Does picking randomly really cost that much more?

Yes. A random pick returns -21.42% against -9.10% for the favourite. Both lose, but the gap is large.

What is the single most useful thing I can change?

Which race you bet in. The charge nearly triples between small fields and big ones, and you can see the field size on the card.

Do you sell tips?

No. Everything on this site is published research, including every losing figure, and there is nothing to subscribe to.