Overround and Round Robin: How Bookmaker Margins Compound

The overround is the bookmaker’s built-in advantage — the gap between true probability and the odds on offer. On a single bet, this margin is visible and relatively modest. On a round robin, it compounds across ten components in a way that is neither visible nor modest. Margins multiply, not add — and the difference between those two words is the difference between a 5 percent edge and something substantially larger. This article explains the mechanics of that compounding, step by step, from a single race market to the full ten-bet round robin.

Overround on a Single Market

Every horse racing market carries an overround. To calculate it, convert each horse’s odds to an implied probability (1 ÷ decimal odds), sum them, and subtract 100 percent. The result is the book percentage — the bookmaker’s theoretical margin.

A six-runner race with all horses priced at 5/1 (decimal 6.00) has implied probabilities of 16.67 percent each, totalling exactly 100 percent — a “fair” book with no margin. In reality, bookmakers compress the odds. The same six runners might be priced at 9/2 (decimal 5.50), giving implied probabilities of 18.18 percent each and a total of 109.1 percent. The overround is 9.1 percent.

British horse racing markets typically carry an overround between 10 and 25 percent, depending on field size, race quality, and the bookmaker. Larger fields tend to have higher overrounds because more runners create more opportunities to embed margin. Smaller fields — five or six runners — usually carry tighter books.

Research by Hegarty and Whelan, published in Applied Economics in 2025, showed that even this overround measure understates real losses. Bettors do not spread money evenly across all runners — they concentrate on favourites and popular selections where margins are thinnest. The actual loss rate exceeds the headline overround by 20 to 40 percent depending on the sport. For horse racing, with its wide fields and varied prices, the understatement is likely at the upper end of that range.

How Overround Compounds in Multiples

When two legs combine in a double, the overround on each leg multiplies rather than adds. Consider two races, each with a 15 percent overround. The implied probability sum in each market is 115 percent (or a factor of 1.15). The combined overround factor for the double is 1.15 × 1.15 = 1.3225, meaning the effective overround on the double is 32.25 percent — more than double the single-leg figure.

For a treble drawing from three races at 15 percent overround each: 1.15 × 1.15 × 1.15 = 1.5209. The effective overround is 52 percent. This does not mean the bookmaker takes 52 pence of every pound you stake on a treble — the overround is a measure of the total implied probability exceeding 100 percent, not a direct loss rate. But it does mean the gap between the odds you receive and the fair odds widens dramatically with each additional leg.

Philip Newall’s 2015 study in Judgment and Decision Making provides empirical support for this pattern. Analysing World Cup betting markets, Newall found that margins averaged roughly 5 percent on simple match-result bets but reached 28 percent on correct-score bets and 48 percent on first-goalscorer markets — multi-outcome constructions where the compounding effect is most severe. Horse racing multiples follow the same mathematical logic, with overround compounding through every leg of every component.

The SSA component introduces an additional twist. Each SSA pair involves two sequential bets: the first leg and, if it wins, the conditional second leg. Each leg passes through its own race’s overround. The conditional nature means the second leg only fires when the first wins, but when it does fire, it carries the full margin of its market. The SSA pair’s effective margin is not the sum or product of two markets’ overrounds — it is a weighted combination, where the weight depends on the probability of the first leg winning. In practice, the effective margin on an SSA pair sits between the single-leg overround and the double’s compounded overround.

Across the six SSA pairs in a round robin, the aggregate margin is substantial. Six pairs, each carrying an effective overround somewhere between 15 and 32 percent (using our 15-percent-per-race example), produce a collective margin that accounts for the majority of the round robin’s total expected loss — because the SSA component accounts for 60 percent of the total outlay.

Practical Impact on a 10-Bet Round Robin

Let us put approximate numbers to a £1 unit stake round robin drawing from three races, each with a 15 percent overround.

The three doubles each carry an effective overround of approximately 32 percent. The treble carries approximately 52 percent. The six SSA pairs carry effective overrounds in the range of 20 to 30 percent each (weighted by the conditional trigger). The blended effective overround across all ten bets — weighted by component type and outlay — sits somewhere around 28 to 35 percent.

A 30 percent blended overround does not translate directly to a 30 percent loss on every round robin. The overround inflates the total implied probability of all possible outcomes within each market, but individual bets can still win. What the blended overround tells you is the long-run expected loss as a proportion of turnover. At 30 percent, you can expect to lose roughly £3 for every £10 staked on round robins over a large sample — three times worse than the loss rate on singles from the same races.

This compounding effect is invisible on the bet slip. The odds you see on each horse look reasonable in isolation. The doubling and trebling of those odds in the multiples section still looks attractive. But the margin has already been embedded in each price, and when ten bets compound those embedded margins, the total drag far exceeds what any individual price would suggest.

The practical response is not to avoid round robins but to choose your spots. Races with tighter overrounds — typically Premier fixtures with deep, competitive fields — give the compounding less raw material to work with. A round robin from three tight-book races compounds less aggressively than one from three wide-book races. Shopping for the best available odds on each selection, and using best-odds-guaranteed where available, further reduces the per-leg margin. These are small improvements, but across ten components they add up.

To illustrate: if you reduce the per-race overround from 15 percent to 10 percent by selecting tighter markets, the blended effective overround on the round robin drops from roughly 30 percent to roughly 22 percent. On a £10 outlay, that is the difference between an expected loss of £3 and an expected loss of £2.20 — a saving of 80p per round robin. Over a season of twenty bets, that is £16 in reduced margin drag. Not transformative, but meaningful — and entirely within the punter’s control.

Summary

Margins multiply, not add. A 15 percent overround on each of three races becomes a 32 percent effective overround on the doubles, 52 percent on the treble, and 20 to 30 percent on the SSA pairs. The blended effect across a ten-bet round robin produces an expected loss rate roughly three times worse than placing singles from the same races.

Understanding this compounding does not change the odds. It changes your expectations — and that is worth more than any betting system, because a bettor with accurate expectations makes better decisions about when to bet, how much to stake, and when to step away.