Guide

How a bet builder works — and where the value hides

By AdamUpdated 1 August 20269 min read

A bet builder is a single bet made of several selections from the same match — a shot on target, over 2.5 goals, a booking for someone combative — combined into one price on one slip. Every selection has to land, or the bet loses. This guide explains how that price is put together, where the margin hides inside it, and how to tell whether you're finding value or just enjoying the ride.

One match, and all of it has to land

Three things separate a bet builder from an ordinary multiple, and each one matters to what it costs you:

• One match, not many. An accumulator strings selections together across different fixtures. A bet builder — same-game multi, bet builder, whatever a given bookmaker calls it — keeps every leg inside the same ninety minutes. That single fact is what makes the pricing interesting, and it is the reason the rest of this guide exists.
• All of it, or none of it. Every leg has to come in; there is no partial payout. Four legs you'd each back at 90% is not a 90% bet, it's a 66% bet (0.9 × 0.9 × 0.9 × 0.9). Adding a leg you're fairly confident about still makes the bet harder, every time.
• The price is not the legs multiplied. Because the legs share a match, they rise and fall together, and the bookmaker prices that in — which is why the number you're offered is usually shorter than the multiplication suggests. That's covered properly further down; it is the single most misunderstood part of the product.

One practical note that catches people out: if a player named in a leg doesn't start, most bookmakers void that leg and settle the bet on the rest — but the rules genuinely differ between books and between markets, so it's worth reading the terms once rather than assuming.

A great product, and priced like one

Bet builders are the bookmaker's favourite product: advertised at half-time, pushed in the app, whole campaigns built around them. What they sell is convenience — one slip, one stake, your whole read of a match. It's a genuinely good product. It's also the most profitable thing most bookmakers sell, and those two facts are related.

Every price a bookmaker offers carries a margin — the gap between the odds you get and the odds the true probability deserves. On a single bet you pay that margin once. In a bet builder, every leg brings its own margin, and margins multiply. Three legs each priced with a typical 5% margin cost you roughly 14% before kick-off; five legs and you're near 23%. Nothing dishonest has happened. You've just paid for convenience five times in one transaction.

Value lives at the leg level

Because the margin arrives leg by leg, that's where value has to be found — one leg at a time, each judged on its own price. The definition is short enough to memorise:

edge = your probability × decimal odds − 1

Suppose a model makes a player 40% to commit two or more fouls. Fair odds for 40% are 2.50. If the bookmaker offers 2.75, the edge is 0.40 × 2.75 − 1 = +10%. If they offer 2.30, the edge is −8% and no amount of liking the player fixes it. The entire game — all of it — is having a better probability than the price implies. Everything else is decoration.

Notice what this rules out. "He's due a card" is not a probability. "He always has shots against them" is a sample of three. A value leg is a specific, falsifiable claim: we rate this 62%, the price implies 57%, and here is where our number came from.

Where a trustworthy probability comes from

You don't need to build a model to bet well, but you do need to know what one does, if only to hold your own numbers to the same standard. SharpXI's, in plain English:

• Start with the player's real rate — how often it actually happens, per 90 minutes, with recent games weighted more heavily.
• Distrust small samples — a hot month from a new signing gets pulled toward what players in that position normally do. The fewer the minutes, the harder the pull.
• Adjust for context that's measurable — the opponent's tendency to concede shots or commit fouls, and for cards, the referee, who is routinely the most mispriced factor on the slip.
• Use the right distribution — shot and foul counts don't behave like coin flips, and pretending they do quietly miscounts the tail outcomes the lines are set on.

The details are on the methodology page, and each market — shots on target, cards, fouls — has its own page explaining what drives it.

The margin you can't see

One step trips up almost everyone who tries to do this properly. To compare your probability with the bookmaker's, you have to strip the margin out of their price first — "de-vigging". The naive way is to shrink every price proportionally. The problem: bookmakers don't spread margin evenly. They load it onto longshots, because punters overpay for longshots — the favourite–longshot bias, one of the oldest and best-documented regularities in betting markets.

Player bets are longshots. De-vig them naively and you'll systematically flatter the bookmaker's implied probability on exactly the prices you care about — and manufacture edges that were never there. SharpXI uses Shin's method, which accounts for the bias. If you take one technical point from this guide, take this one: a bad de-vig doesn't just add noise, it invents value.

The correlation caveat

Multiply three leg prices together and you've assumed the legs are independent. In one match, they aren't. Goals, shots, cards and fouls are all driven by the same ninety minutes — the derby that turns feisty produces the card and the fouls; the game that opens up produces the goals and the shots. Bookmakers know this, which is why the bet builder price you're offered is often shorter than the multiplication says it should be.

The honest consequence: a combined "edge" computed by multiplying leg edges is an optimistic number, not a real one. SharpXI's builder shows the naive multiplication and then corrects it — it measures how often a player's legs have actually landed together across that player's match history, and shows the adjusted true chance and fair price next to the naive figure. It also flags redundant legs, like 1+ shots stacked under 1+ shots on target, which add margin and no value. Treat the leg edges as the signal and the adjusted combined figure as the honest estimate.

Keeping score honestly

Say you do everything above properly. What happens next is the part nobody advertises: mostly, variance. A +8% edge still loses outright more often than not at bet-builder prices. Results over a weekend tell you nothing, and even a season of them is barely conclusive. The scoreboard that actually works is closing line value: whether the prices you took consistently beat the market's final, sharpest price. Beat the close over hundreds of legs and profit tends to follow. Fail to, and a hot streak is a hot streak, however long it lasts.

Two pieces of discipline follow. Stake as if your edges are half as good as they look — a fractional Kelly approach exists precisely because model edges are estimates, not facts. And watch sample size: SharpXI puts one on every pick because a 65% built on nine matches deserves far less trust than a 65% built on ninety.

One more honest note: bookmakers restrict accounts that win. That isn't a reason not to seek value — it's evidence the value is real and an honest cap on how far it scales. Anyone selling you a bet-builder system without mentioning it is selling you the fun part only.

Frequently asked

Keep going