Odds Display Keeps 3 Decimal Places, 35% Read kn 1.99 as 2.00

Around 35% of Croatian sportsbook users who see odds displayed as 1.99 read it as 2.00 when asked to recall the number seconds later. That figure comes from a small internal usability panel run across three Zagreb betting shops in early 2024, and it lines up with what eye-tracking studies have shown for years: three-decimal odds are read as two-decimal odds with a rounding bias that almost always favours the prettier number. The display convention is not neutral. It changes what bettors think they are being paid.

Why three decimals exist at all

Bookmakers did not adopt three-decimal pricing because punters demanded precision. They adopted it because margins got tight enough that the third decimal started to matter commercially. On a main-line football market, the difference between 1.95 and 1.953 is roughly 0.15% of margin. Multiply that across a full book's turnover and it is real money.

There is also a competitive angle. When two operators both price a favourite at "around 1.95", the one showing 1.953 looks marginally better in a comparison widget even though the underlying price is effectively identical. The third decimal becomes a marketing surface, not a mathematical one.

Croatian licensed operators under HZJZ-adjacent oversight have generally followed the regional norm set by Maltese and Austrian platforms, which is a three-decimal default on most pre-match football, basketball, and handball markets. Live markets often switch to two decimals because the price moves faster than the display can refresh. That inconsistency is itself a source of confusion: the same bettor who reads 1.99 as 2.00 pre-match reads 1.99 live as 2.00 too, but the live version moves before the bet lands.

The rounding bias is asymmetric

The 35% figure is not evenly distributed. When the third decimal is 1 through 4, misreading is rare — people round down correctly or just truncate. When the third decimal pushes the price to the edge of a round number, the error rate climbs sharply. Prices ending in .99, .98, and .95 are the worst offenders.

In the panel data, 1.99 was misread as 2.00 by 35% of respondents. 1.98 was misread as 2.00 by 22%. 1.95 was misread as 2.00 by 9%. The pattern is clear: the closer the displayed price is to a psychologically round number, the more likely the brain completes the rounding on its own.

This is not a Croatian phenomenon. It replicates in German, Italian, and Polish samples. But Croatian bettors have a specific quirk: because the kuna-to-euro conversion in 2023 introduced a generation of users to decimal-shifting at speed, there is some evidence that three-decimal odds are parsed slightly faster than in markets that never went through a currency change. Faster parsing means more rounding errors, not fewer.

What the third decimal actually costs the bettor

Take a €10 single on a price displayed as 1.99. If the bettor believes they are getting 2.00, they expect €20 back. At 1.99 they get €19.90. The 10-cent gap is trivial on one bet. It is not trivial on a 500-bet year.

At an average stake of €10 and 500 bets, the difference between a true 2.00 average price and a true 1.99 average price is €50 over the year, assuming a 50% strike rate on even-money-ish selections. That is before the operator's margin, which is already baked into the fact that the "true" price is closer to 1.95 than either number.

The more interesting cost is behavioural. A bettor who believes they are getting 2.00 on a coin-flip-ish event is more likely to take the bet than one who correctly reads 1.99. The threshold effect is well documented in retail pricing: 1.99 feels like "under 2" and 2.00 feels like "2". But in betting, the direction is reversed from retail. In a shop, 1.99 is cheaper than 2.00 and the consumer wins. In betting, 1.99 pays less than 2.00 and the consumer loses. The same psychological trick that makes 1.99 attractive in a supermarket makes it attractive in a sportsbook for the wrong reason.

The 2.00 barrier in Croatian bettor behaviour

Croatian bettors show a pronounced preference for selections at or above 2.00. In aggregated stake data from a mid-size domestic operator covering 2023, the share of single stakes placed on selections priced between 1.95 and 1.99 was 8.4%, while the share on selections priced at exactly 2.00 was 11.2%. Selections at 2.01 to 2.05 drew 6.1%.

That is a spike, not a smooth curve. The 2.00 price point captures stakes that the underlying probability does not justify. Some of that is genuine preference for even-money. Some of it is the display making 1.99 look like 2.00, so bettors do not distinguish between the two when scanning a coupon.

If the display showed 1.99 as 1.99 and 2.00 as 2.00 with equal visual weight, you would expect the stake distribution to be smoother. The spike at 2.00 suggests the third decimal is being visually suppressed, not just numerically misread.

How operators and regulators handle it

There is no Croatian rule requiring three decimals. The relevant framework, inherited from the pre-2023 gambling law and updated in the 2023 Act on Games of Chance, requires that odds be displayed "clearly and unambiguously" but does not specify decimal places. In practice, operators choose their own convention, and most choose three because that is what the platform vendor defaults to.

Some operators offer a display toggle in account settings: two decimals, three decimals, or fractional. Take-up on the two-decimal option is low. In data from one operator, fewer than 4% of active users had changed the default. The default wins, which means the vendor's choice becomes the de facto standard for the market.

A handful of operators have experimented with "clean" pricing, where every price is shown to two decimals and the underlying price is rounded to the nearest 0.01. The commercial effect is mixed. Margins compress slightly on markets where the third decimal was being used to shave value, but bettor satisfaction scores on clarity questions rise. One operator reported a 12% drop in "confusing odds" complaints in the quarter after switching, with no measurable drop in turnover.

What a fairer display might look like

Three options are on the table, each with trade-offs.

Option 1: Two decimals, honest rounding. Show 1.99 as 1.99 and 2.00 as 2.00, and round 1.953 to 1.95. This is the simplest fix. The cost is that operators lose the ability to price at 1.953, which slightly reduces their margin flexibility. Bettors get a clearer picture.

Option 2: Three decimals, but with the third decimal visually de-emphasised. Grey out the third digit or reduce its font size. This preserves pricing precision while signalling that the number is not the headline. Eye-tracking suggests this reduces misreading by roughly half, but it does not eliminate the 2.00 completion effect.

Option 3: Show the implied probability alongside the odds. A price of 1.99 is a 50.25% implied probability. A price of 2.00 is 50%. Displaying both makes the difference explicit. This is the most informative option and the least popular with operators, because it draws attention to the margin.

None of these are required by law. All three have been tested in other markets with varying results. The Croatian market, being small and vendor-dependent, will likely follow whatever the dominant platform does rather than lead.

The live betting problem is worse

Pre-match misreading is annoying. Live misreading is expensive.

In live markets, prices move in seconds. A bettor who reads 1.99 as 2.00 and taps the bet slip may find the price has moved to 1.96 by the time the bet is confirmed. The confirmation screen shows the new price, but the bettor's mental model is still anchored on 2.00. The result is a bet the bettor did not intend to place at a price they did not agree to.

Croatian operators are required to display a price-change confirmation when the odds move beyond a certain threshold before bet acceptance. The threshold varies by operator, typically 0.02 to 0.05. A move from 1.99 to 1.96 is exactly the kind of change that sits near the boundary: large enough to matter, small enough that some operators accept the bet without a second confirmation.

The 35% misreading figure was measured on pre-match coupons. In live markets, where cognitive load is higher and time pressure is greater, the misreading rate is almost certainly higher. No Croatian operator has published live-specific data, which is itself telling.

What the bettor can do

There is no setting that fixes this entirely, but there are habits that reduce the damage.

Read the third decimal deliberately. If the price is 1.99, say "one ninety-nine" in your head, not "two". The act of verbalising the third digit reduces the completion effect.

Check the bet slip before confirming. The bet slip shows the exact price and the exact potential return. If the return does not match what you expected, the odds were not what you thought.

Use the two-decimal display if your operator offers it. It will not change the price you get, but it will change what you see, and what you see is what you bet on.

Set a stake limit. Not because three-decimal odds are dangerous in themselves, but because the misreading effect is strongest when bets are placed quickly and without reflection. A stake limit forces a pause.

The wider question about display neutrality

The 35% figure is small-sample and should be treated as indicative, not definitive. But the direction is consistent with decades of pricing psychology research, and the mechanism is well understood. The question is not whether three-decimal odds cause misreading. They do. The question is whether that misreading is an acceptable cost of pricing precision, or whether it is a design choice that quietly transfers value from bettors to operators.

Croatian regulators have shown appetite for display rules in other areas — warning labels on slot machines, mandatory session reminders, restrictions on bonus advertising. Odds display has not been a priority. That may change if the market grows or if a consumer protection case forces the issue.

In the meantime, the third decimal sits there on every coupon, doing two jobs at once: pricing the market to a fraction of a percent, and nudging a third of readers toward a number that is not quite real. Whether that is clever pricing or a small, legal deception depends on which side of the bet slip you are standing on.