An each-way bet is the most misunderstood bet in British racing, and it is misunderstood in a very particular way. Most punters know roughly what it does — you get something back if the horse places — but very few could tell you what the place half of the bet actually costs, what it pays, or whether it is any good. They treat "each-way" as a single soft-edged thing, a hedge, a bit of insurance. It is not. It is two separate bets stapled together, and one of them is priced by a formula rather than a market.
That last part is the whole story. The win half of an each-way bet is priced by a bookmaker watching money and form. The place half is not priced at all — it is derived, mechanically, as a fixed fraction of the win odds, with the fraction and the number of places set by the size and type of the race. A formula that rigid cannot be right for every race. In some races it is mean. In others it is loose, and the trade has a name for the people who spend their lives looking for the loose ones: each-way thieves.
This piece covers the machinery first — the two bets, the standard terms table, a worked example with the arithmetic shown — and then the judgement: which races the formula misprices, what extra-places offers really change, and how to tell whether the place half of your bet is carrying a win half that couldn't stand on its own. None of it makes betting profitable, and most punters lose over time whatever bet type they use. It will, at least, tell you what you are actually buying.
Two bets stapled together
"£5 each-way" is not a £5 bet. It is two £5 bets — £10 out of your pocket. The first £5 goes on the horse to win at the full odds. The second £5 goes on the horse to place — to finish in the first two, three or four, depending on the race — at a fixed fraction of the win odds.
The two halves settle independently. If the horse wins, both pay: it has, after all, also placed. If it finishes second in a race paying three places, the win half loses in full and the place half pays. If it finishes outside the places, both halves lose and the whole £10 is gone.
Notice what you can never do: choose the place terms. You cannot ask for a quarter of the odds instead of a fifth, or four places instead of three. The terms are set by the field size and the race type, printed on the card before you bet. Your only decision is whether to accept them.
The standard place terms, and where they come from
The standard terms used across UK and Irish racing are as follows. This table is worth memorising — it is the same at essentially every mainstream firm, and both Sporting Life's each-way guide and bet365's published horse racing rules state it identically.
| Runners | Race type | Places paid | Place fraction |
|---|---|---|---|
| 2–4 | Any | None — win only | — |
| 5–7 | Any | First 2 | 1/4 of the odds |
| 8+ | Non-handicap | First 3 | 1/5 of the odds |
| 8–11 | Handicap | First 3 | 1/5 of the odds |
| 12–15 | Handicap | First 3 | 1/4 of the odds |
| 16+ | Handicap | First 4 | 1/4 of the odds |
Three things to absorb. First, with fewer than five runners there is no each-way betting at all: place money is settled as a win bet. Second, the handicap distinction only bites at twelve runners and above — handicaps are more open by design, so the terms improve, from a fifth to a quarter of the odds and eventually to four places. Third, the fraction applies to the odds, not the return: a place at 1/4 odds on a 10/1 (11.00) shot pays at 5/2 (3.50), not at a quarter of £11.
These are minimum standard terms. Bookmakers frequently pay more places as a promotion — the Grand National is often settled at five or more places by the big firms — and we will come to what those offers really change below.
A worked example, every number shown
Take £5 each-way on a horse at 10/1 (11.00) in a 16-runner handicap. Standard terms: 1/4 the odds, first four places. Total outlay £10. The place half is settled at 10/1 ÷ 4 = 5/2 (3.50).
| Result | Win half (£5 at 10/1) | Place half (£5 at 5/2) | Total returned | Net |
|---|---|---|---|---|
| Wins | £55.00 | £17.50 | £72.50 | +£62.50 |
| Finishes 2nd–4th | £0 | £17.50 | £17.50 | +£7.50 |
| Finishes 5th or worse | £0 | £0 | £0 | −£10.00 |
Check the middle row, because it is the one that surprises people. The horse runs a fine race, finishes third — and you get £17.50 back from a £10 outlay. A profit of £7.50. Perfectly decent, but a long way from the £62.50 the win delivers, and it only takes one non-runner or one price notch to turn that middle row negative. The same £5 each-way at 7/1 in the same race returns £13.75 for a place — £3.75 net. At 4/1, a place returns £10 exactly: your money back, nothing more. Below 4/1 under quarter-odds terms, a placed horse loses you money. That single line of arithmetic — the fraction times the odds versus the lost win stake — is the first filter for any each-way bet.
Split the bet: what each half implies
Here is the discipline that separates people who understand each-way from people who merely use it. Unstaple the bets and price each one alone.
The win half above is £5 at 10/1 — an implied chance of 1 in 11, about 9.1%. The place half is £5 at 5/2 — an implied chance of 2 in 7, about 28.6%. So the combined bet is only sound if you believe both things: that the horse wins at least roughly 9% of the time, and that it finishes in the first four at least roughly 29% of the time. Those are separate claims about the world, and a horse can satisfy one while failing the other. A hold-up horse in a strongly run race (see pace maps) might place far more than 29% of the time yet win far less than 9%. A tearaway front-runner might show the opposite profile.
An each-way bet is two bets. Split it, price each half on its own, and ask whether you would place either one by itself. If the answer is "the place half yes, the win half no", the place half is subsidising a bad bet — and there is usually a cleaner way to buy the part you actually like.
The cleaner way exists. Exchanges run standalone place markets, and some firms offer "betting without the favourite". If your real opinion is "this horse hits the frame but doesn't win", you can often express exactly that, at a market price rather than a formula price — worth checking against the exchange before defaulting to each-way.
Where the formula runs loose: the each-way thief's races
Because the place fraction is chained to the win odds, it goes wrong whenever the win market and the place reality come apart. The classic case is a small field with one dominant horse.
Picture a six-runner handicap: an odds-on favourite at 4/6 (1.67), a clear second-string at 8/1 (9.00), and four others priced in double figures. Terms: 1/4 the odds, first two places. The formula prices the 8/1 shot's place half at 2/1 (3.00) — an implied one-in-three chance of finishing in the first two. But look at the shape of the race. The favourite is priced as roughly a 60% winner, so one of the two place slots is spoken for most of the time; the genuine contest is for the slot behind it, and the 8/1 shot is comfortably the most likely horse to fill it. Its true chance of finishing top two is plainly better than the one-in-three the formula charges — not because anyone has misjudged the horse, but because the fraction is keyed to a win price that a monster favourite has stretched.
That gap is what the each-way thief harvests: races where a short favourite inflates everyone else's win odds, and the rigid fraction drags the place odds up with them. Two caveats, both real. First, the win half of such a bet is usually poor — you are buying a mispriced place leg bundled with a fairly hopeless win leg, and the bundle as a whole may still be no bargain. Second, bookmakers know this pattern at least as well as you do; accounts that bet this way persistently tend to find their stakes restricted. And be careful the "dominant" favourite deserves its price at all — a fake favourite collapses the whole logic.
The mirror image is just as important. In a wide-open twenty-runner handicap with no strong favourite, the formula often flatters the bookmaker instead: a fifth of the odds on a 25/1 chance can be a thin price for a place in a race where a dozen horses hold genuine claims. The formula is not systematically generous or systematically mean. It is rigid, and rigidity cuts both ways.
What extra places really change
"Extra places" — five, six, sometimes more places on a big handicap instead of the standard four — is the most heavily promoted each-way offer in racing, and it genuinely alters the arithmetic. Each additional place is an extra winning outcome for the place half at no change in stake. On a 20/1 (21.00) shot at 1/4 odds, the place half pays 5/1 whether the terms cover four places or six; six simply makes the event much more likely.
But read the terms, because the places and the fraction move together. A common shape is six places at 1/5 the odds where the standard book pays four at 1/4. On that 20/1 shot, 1/4 odds gives a place price of 5/1 (implied 16.7%); 1/5 odds gives 4/1 (implied 20%). You are being asked to accept a shorter place price in exchange for two extra finishing positions. For big-priced horses, whose chance of sneaking fifth or sixth dwarfs their chance of winning, the trade is usually favourable. For horses near the head of the market, the shaved fraction bites harder and the extra places add little they weren't already likely to reach. And bookmakers do not run these offers out of kindness: the win prices on heavily promoted races are often shaded to pay for them, which is a job for reading the overround — on the win book and, mentally, on the place book too.
Non-runners move the goalposts
One trapdoor catches even experienced each-way backers: the terms are set by the number of horses that actually run, not the number declared. A 16-runner handicap advertising four places becomes a 15-runner race paying three the moment one horse is withdrawn. An 8-runner race paying three places becomes a 7-runner race paying two. If your horse finishes fourth in the first case or third in the second, a bet that was winning on the card you studied is a losing one at the settle.
The same withdrawal usually triggers a deduction from your odds as well, under Rule 4 — so a single non-runner can shorten your price and delete your place in one stroke. Before any each-way bet in a field near a terms boundary — 8 runners, 12, 16 — it is worth asking what the bet looks like if one comes out. Some firms' "extra place" offers also quietly revert to standard terms below a stated field size; the detail lives in the small print, and the small print settles the bet.
What to take from all this
Each-way betting is neither a mug's habit nor a hidden edge. It is a bundle: a market-priced win bet tied to a formula-priced place bet, on terms you cannot negotiate, only accept or refuse. The formula is mean on short-priced horses, roughly fair in the middle, occasionally loose in small fields behind big favourites, and genuinely improved — sometimes — by extra-places offers whose cost is hidden elsewhere in the book.
The honest summary is the same one that applies to every page on this site: understanding the terms will not make you a winning punter, and the normal long-run outcome for each-way backers is the same as for everyone else — a loss, at roughly the rate the margin dictates. What understanding does is stop you paying for things you didn't mean to buy: place halves that lose money on placed horses, win halves smuggled in behind a place price you liked, four places that became three while you were queueing. Bet to stakes that assume losing — bankroll and staking covers how — and treat each-way as what it is. Two bets. Price them both.