Accumulators: Why the Payout Looks Bigger Than It Is

Everyone understands the appeal. Six results you already have opinions about, one small stake, a life-changing-looking return. What almost nobody does is price the thing they just built. This article does exactly that, and the arithmetic is not subtle.

How the odds are built#

Multiply the decimal odds of every leg. Three legs at 1.80, 2.00 and 1.50:

1.80 × 2.00 × 1.50 = 5.40

A 10-unit stake returns 54 units. That is the number the slip shows you, in large type.

Here is the number it does not show you. Probability multiplies too, and identically. Convert each leg: 1 ÷ 1.80 = 55.6%, 1 ÷ 2.00 = 50.0%, 1 ÷ 1.50 = 66.7%. Multiply: 0.556 × 0.500 × 0.667 = 18.5%. And 1 ÷ 0.185 = 5.40, the same number. The odds and the chance are two views of one quantity, so nothing has been created by combining the bets. All that has happened is that your money now depends on three independent things instead of one.

Build a few and watch both halves move together — the same tool lives on its own page as the accumulator calculator if you want it to hand while you read a coupon:

The margin compounds — this is the whole argument#

Every one of those legs was quoted by a bookmaker who built a cut into the price. Call it 5%, which is about right for a mainstream match result market and generous compared to what most high-street books charge.

That means the true probability of each leg is lower than its implied probability — by a factor of roughly 1.05. (If the idea of stripping margin out of a price is new, read fair odds first; everything below depends on it.) So take the fair versions: 55.6 ÷ 1.05 = 52.9%, 50.0 ÷ 1.05 = 47.6%, 66.7 ÷ 1.05 = 63.5%. Multiply those three fair probabilities: 16.0%. The fair price of this accumulator is 1 ÷ 0.160 = 6.25.

You were offered 5.40 for something worth 6.25.

The gap is 6.25 ÷ 5.40 − 1 = 15.8%. And 1.05³ − 1 = 1.157625 − 1 = 15.8%. Same number, because that is what is happening: the margin factors multiply.

Accumulator overround = (1 + m)ⁿ − 1 where m is the margin per leg and n is the number of legs.

Not m. Not m × n. It compounds, exponentially, and it does so silently — every leg you add makes the displayed payout bigger and the deal worse at the same time.

Legs	Total overround	Expected return per unit staked
1	5.0%	95.2%
2	10.3%	90.7%
3	15.8%	86.4%
4	21.6%	82.3%
6	34.0%	74.6%
8	47.7%	67.7%
10	62.9%	61.4%

Read the right-hand column slowly. A ten-fold acca built from ordinary 5% legs returns, on average, 61 units for every 100 staked. You are not paying a 5% commission. You are paying nearly 39% of your stake for the privilege of the payout shape.

Short favourites are not safer#

The most common way to build an acca is to string together six "bankers" — heavy favourites, prices around 1.10, results that feel certain.

Give each of them a genuine 90% chance, which is optimistic for a 1.10 shot after margin. The probability that all six land is:

0.9⁶ = 0.531441 → 53.1%

Just over a coin flip. Six things you would each bet your house on, and the parlay fails nearly half the time. There is no version of "safe" that survives multiplication.

And the margin has been compounding the entire time. Six 90% legs at a 5%-margin book price at about 1.058 each, so the acca pays roughly 1.40. The fair price of a 53.1% event is 1 ÷ 0.531441 = 1.88. You are being offered 1.40 for 1.88 — a 34% overround, exactly as the table says.

The intuition that fails here is treating "very likely" as "certain". Ninety per cent is not certainty; it is one failure in ten. Ask for six of them at once and you have asked for something quite hard.

When an acca is still defensible#

Not never. Three honest cases.

You want the payout shape and you know its price. There is nothing irrational about paying for a lottery-shaped outcome the way you'd pay for a lottery ticket, as long as you have done the arithmetic above and are spending entertainment money, not investment money. The failure is not betting an acca — it is betting one while believing it is good value. Keep it inside a stake plan you set in advance; bankroll management does not stop applying because the bet is fun.

The legs are genuinely mispriced. Edges compound in exactly the way margins do. If every leg carries a real 8% edge over a book charging 5%, the ratio per leg is 1.08 ÷ 1.05 and the acca is positive. The catch is severe: you need a real, demonstrated edge on every leg simultaneously, and the errors have to be independent. Correlated legs — same match, same league, same weather — break the multiplication rule entirely, which is why books price "bet builders" as a separate, fatter-margin product rather than letting you multiply their own numbers. Almost nobody has verified edges on six markets at once. Be honest about whether you do; the expected value guide is where that question gets settled.

Someone else is paying the margin. Free bets, price boosts, acca insurance and enhanced-odds offers shift part of the cost off you. That genuinely changes the sums, and it is why those promotions exist attached to accas specifically. Read the terms, work out the actual value of the offer, and treat the boost as the product rather than the bet.

Work out what strike rate your acca actually needs before you place it — the break-even calculator converts a price into the hit rate required to stand still, and the number is usually sobering.

Why the site pushes them at you#

Accumulators are the most profitable product on a bookmaker's site, per unit staked, by a distance. The margin compounds, the payouts are visible and shareable, the losing slips are forgotten, and the near-misses — five from six — generate more repeat business than wins do.

That is not a conspiracy, it is just product design, and it is worth saying plainly: the promotional weight behind accas is a reliable signal of which side the maths favours. Nobody advertises the bets that cost them money.