Float-to-Integer Odds Display Cuts Misread Tickets 19% in 500 Bets

A trial run across 500 settled bets found that displaying odds as integers rather than floating-point decimals reduced misread tickets by 19%, from 71 errors in the control group to 58 in the treatment group. The result comes from a small operator-side experiment conducted between 3 and 17 March 2025, using a single sportsbook interface and a panel of 250 participants recruited from an existing customer base. The finding is narrow, but it speaks to a persistent and under-discussed problem in bet-slip design: the gap between how odds are stored and how they are read.

The Problem With Decimal Odds at the Point of Confirmation

Decimal odds are the default presentation format across most regulated markets outside the United States, and for good reason. They are transparent: a 2.50 line tells the bettor the total return per unit staked, including the stake itself. Fractional odds, still common in the UK and Ireland, obscure this for casual users. American moneyline odds, standard in US sportsbooks, invert the mental arithmetic depending on whether the price is positive or negative.

But decimal odds carry a subtler cost. Floating-point representation means a price can be stored as 2.4999999 or 1.9090909, and interfaces vary in how many decimal places they display. A bettor comparing 1.91 on one screen with 1.909 on another may not register that these are the same price, or may misread a truncated 1.90 as meaningfully better than 1.91 when the true line is 1.905. The error is small per ticket, but it compounds across a slip that includes four or five selections, particularly when a bettor is trying to verify that the combined odds match their expectation.

The trial's premise was straightforward: if odds were rounded and displayed as integers — 250 instead of 2.50, 191 instead of 1.91 — the cognitive load of reading and confirming a price would fall. The mechanism is not that integers are inherently more accurate. They are not. A price of 2.50 displayed as 250 is the same number with the decimal shifted. The claim is that removing the decimal point reduces parsing errors at the confirmation stage, where bettors are most likely to misread and least likely to notice.

What Counts as a Misread

The study defined a misread ticket as one where the participant's stated understanding of the odds, recorded immediately after placing the bet, diverged from the actual odds displayed. This is a self-reported measure, which is a limitation. It captures perception, not behaviour. A bettor who misreads 1.91 as 1.97 but stakes the same amount and accepts the same risk has not suffered a material loss. The 19% reduction is therefore a reduction in reported confusion, not in financial harm.

That distinction matters. Operators have an interest in reducing misreads because confused bettors are more likely to contact support, dispute a settlement, or churn. A 19% reduction in misreads across a book that processes, say, 40,000 tickets a day would translate to roughly 7,600 fewer confused confirmations daily, assuming the effect holds at scale. Whether it does is an open question.

Why the Effect Size Is Plausible but Fragile

A 19% relative reduction from 71 to 58 errors is statistically detectable in a sample of 500, but only just. The absolute difference is 13 tickets. With a 95% confidence interval, the true effect could plausibly range from a modest 4% reduction to a substantial 34% reduction, depending on how the sample was drawn and whether the two groups were matched on betting experience, device type, and market familiarity. The study reports that participants were randomised, which helps, but randomisation does not guarantee balance in a sample this small.

There is also a novelty effect to consider. Participants in the integer-display group were using an unfamiliar format. Some may have paid closer attention precisely because the display was different, inflating the apparent benefit. A follow-up at 30 or 60 days, after the format became routine, would test whether the effect persists or decays toward zero. The study period was 14 days, which is long enough to capture initial confusion but short enough to leave habituation unmeasured.

The Rounding Question

Integer display requires a rounding convention, and the study does not specify one clearly. If 1.91 becomes 191 and 1.909 becomes 191, the two are indistinguishable. If 1.95 becomes 195 and 1.949 becomes 194, the rounding rule matters for bettors comparing prices across markets. A consistent round-half-up rule is simple but introduces a systematic upward bias of up to 0.5% on displayed odds, which is a regulatory problem in jurisdictions that require displayed odds to match settled odds exactly.

This is the practical obstacle. In the UK, the Gambling Commission's licence conditions require that odds be displayed accurately. In Australia, the Interactive Gambling Act and state-level regulations impose similar constraints. Shifting to integer display is not a design tweak; it is a compliance question. An operator cannot display 191 if the settled price is 1.909 and the difference affects payout. The workaround is to display integers only at the confirmation stage, with the precise decimal available on request, but that reintroduces the parsing problem the study set out to solve.

What the Result Does and Does Not Support

The study supports a narrow claim: in a controlled setting with 250 participants and 500 bets, integer odds display reduced self-reported misreads by 19%. It does not support the broader claim that integer display improves bettor comprehension, reduces disputes, or increases trust. Those would require a larger sample, a longer window, and outcome measures beyond self-report.

It also does not address the population most at risk of misreading odds: new bettors. The participants were drawn from an existing customer base, which means they had prior exposure to decimal odds. A first-time bettor encountering 250 instead of 2.50 may find the integer format more confusing, not less, because it departs from every other price they have seen. The study's own logic — that familiarity drives parsing ease — cuts against its conclusion for novice users.

The Regulatory Angle

If integer display were adopted widely, it would create a two-tier system: experienced bettors reading 250 as 2.50, and everyone else wondering why the number looks like a stake rather than a price. The 19% figure is interesting precisely because it is small enough to be plausible and large enough to justify further study. It is not large enough to justify a format change on its own, particularly one with compliance implications.

A more promising line of inquiry is whether the misreads cluster around specific price ranges. Odds near evens (1.90–2.10) are read more often than longshots, and the decimal point matters more at 1.95 than at 15.0. If the 19% reduction is concentrated in the 1.80–2.20 band, a targeted intervention — bolding the decimal, or displaying the implied probability alongside the price — might achieve the same effect without the rounding problem. The study does not break down its results by price range, which is the most obvious gap in its design.

An Open Question for Operators

The finding invites a question that the study cannot answer: if a 19% reduction in misreads comes from removing a single decimal point, how much of the remaining 81% comes from the way odds are framed, ordered, and contextualised on the slip? Operators spend heavily on interface design, yet the confirmation screen — the last point at which a bettor can catch an error — is often the least examined. The 500-bet trial suggests the cost of that neglect is measurable, but it also suggests the effect is modest and contingent on factors the study did not isolate. The next step is not to switch to integers. It is to find out which misreads actually cost money, and whether a display change can reach them.