Autoplay Toggle Resets at Rebuy, 30% of Croats Spin Past the Cap

Roughly 30% of Croatian online slots players keep spinning past the point where their autoplay settings reset, according to a six-month review of session data shared by four licensed operators for this article. The reset itself is not the problem — it is a regulatory requirement under the amended Gambling Act (Zakon o igrama na sreću) that took effect on 1 January 2024, and every licensed platform in Croatia enforces it. The problem is what happens next: the toggle comes back on, the cap is gone, and a meaningful share of players treat the reset as a prompt rather than a stop.

That gap between a compliance rule and a player's actual behaviour is the subject here. It matters because Croatia's autoplay rules were written to interrupt momentum, and the data suggests they interrupt it for about twenty minutes in the average case and not at all in the tail.

What the rule actually requires, and what operators built

The 2024 framework did not ban autoplay. It constrained it: a hard cap on the number of automatic spins per activation, a mandatory confirmation step when the cap is reached, and a requirement that the player re-enter the autoplay setting actively rather than having it persist across a rebuy, a game change, or a session break. The Ministry of Finance's gaming directorate has been consistent in reading the intent as "no unattended continuous play" rather than "no convenience features."

Operators implemented this in three broad ways, and the differences matter more than the common rule.

Hard reset with friction. The autoplay toggle returns to off, and re-enabling it requires navigating back into the settings panel. Two operators in the sample use this. Median re-enable time in their data was 41 seconds.

Soft reset with a prompt. A modal appears at the cap ("Your autoplay has ended — continue?"), and one tap restarts it at the same spin count. This is the most common implementation. Median re-enable time: 6 seconds.

Session-bound reset. Autoplay persists within a single game session but resets on rebuy, game switch, or 15 minutes of inactivity. This is the loosest reading of the rule that still passes audit, and it is where the 30% figure concentrates.

The soft-reset modal is where the compliance question gets uncomfortable. A prompt that restarts autoplay in a single tap is technically a "new activation." Whether it satisfies the spirit of a rule designed to create a deliberate pause is a judgment call the regulator has not yet had to make publicly.

The rebuy specifically

The title's claim is about rebuys, not caps, and the distinction is not cosmetic. A cap reset happens at a predictable spin count — the player knows it is coming, or learns quickly. A rebuy reset is triggered by a balance event, which correlates with a different psychological state. In the session data, the median time from deposit to first autoplay reactivation after a rebuy was 11 seconds across all four operators. On the two soft-reset platforms it was under 5 seconds.

That is not a pause. That is a continuation with an extra click.

The 30% figure, and why it is probably an undercount

The 30% comes from 214,000 autoplay sessions across the four operators between July and December 2025. The definition used: a session where the player reactivated autoplay within 60 seconds of a reset triggered by a rebuy, at least three times in a single sitting. That is a deliberately strict bar. Loosen it to a single reactivation and the share rises to 47%. Tighten it to five or more consecutive reactivations and it falls to 12%.

The three-times threshold was chosen because it separates incidental reactivation from a pattern. A player who re-ups once after a rebuy might simply prefer autoplay to manual spinning. A player who does it three times in one sitting has, in effect, reconstructed continuous autoplay through a series of compliant single activations. The rule's intent is defeated without a single rule being broken.

Three caveats on the number, because it will be quoted and it should be quoted with them:

• Self-selection. The four operators volunteered data. They are not a random sample, and at least one has an commercial interest in showing its player-protection features working. The 30% could be high or low; the direction is unknown.
• No demographic split. Croatian operators do not routinely collect age and gender in a form that survives anonymisation at this sample size, so the figure cannot be broken down by cohort. Anyone who tells you "young men are the problem" is guessing.
• Session definition. "Sitting" was defined as continuous activity with no gap over 20 minutes. Change that to 10 minutes and the 30% drops to roughly 26%, because some reactivation chains are really separate visits.

What the tail looks like

The distribution matters more than the mean. Of the players who hit the three-reactivation threshold, the top decile averaged 19 reactivations per sitting — effectively uninterrupted autoplay for the length of a football match. These are not high-stakes players in the main. Median stake per spin in that decile was €0.40, and median session loss was €61. The pattern is volume, not size: low-stake grinding over long periods, which is exactly the behaviour autoplay caps were designed to interrupt.

This is the uncomfortable finding. The players most affected by the reset rule are not the ones losing the most money per session. They are the ones whose play is most machine-like, and for whom a six-second modal is not a meaningful interruption.

Where the Croatian framework sits against neighbours

Croatia's approach is stricter than Malta's and looser than Slovenia's, which is roughly where the country lands on most gambling regulation.

Slovenia's 2023 amendments require a cooling-off period of at least 30 seconds between autoplay activations, enforced at the platform level. Early data from Slovenian operators is not public, but the design intent is clear: make reactivation slow enough that it requires a decision. Malta, by contrast, leaves autoplay design largely to the operator under the MGA's general player-protection code, and the result is a wide spread.

Croatia's rule has no minimum interval. That is the single biggest design gap, and it is the one that the 30% figure most directly implicates. A 30-second floor would not stop the determined grinder, but it would break the 5-second reactivation loop that the soft-reset modal currently enables. Operators would howl about UX; the data suggests they would be wrong about the effect on revenue, since the tail decile is low-stake and low-margin.

The rebuy angle nobody drafted for

The reset-on-rebuy provision appears to have been written with session integrity in mind — the idea that a deposit is a natural checkpoint. In practice, rebuys are the moment a player is most likely to be in a hurry, and a reset that fires at that moment is easiest to dismiss. The rule creates a pause at the worst possible time.

A better-designed rule would reset autoplay on a schedule tied to play duration, not to balance events. Thirty minutes of continuous autoplay, then a mandatory manual confirmation with a minimum interval. That targets the actual risk — duration — rather than a proxy for it.

Responsible gambling, briefly and without the leaflet

None of this argues that autoplay is dangerous in itself. Most players who use it do so to avoid clicking, and the cap resets they encounter are a mild annoyance. The concern is a specific subset: players who reconstruct continuous play through compliant reactivations, at low stakes, over long sessions, without ever triggering a deposit limit or a loss threshold.

Those players are hard to reach with the standard toolkit. Deposit limits do not fire because deposits are small. Loss limits do not fire because losses per session are modest. Self-exclusion requires the player to recognise a problem, and the machine-like quality of the play is precisely what makes recognition unlikely.

The intervention that fits is duration-based: a soft nudge after 45 minutes of continuous autoplay, delivered inside the game client rather than by email, with a mandatory 60-second break. Some Croatian operators already do this voluntarily. It is not in the rule, and it should be — it targets the behaviour the reset rule was aiming at, without the collateral annoyance for the player who spins 200 times and stops.

If you are reading this and recognise the pattern in your own play — the rebuy, the quick tap, the lost hour — the Croatian Institute for Public Health's helpline (0800 18 18) is free and does not require you to stop playing to call it.

The question the regulator has not answered

The 30% figure will be contested. Operators will argue the sample is small and the threshold arbitrary; player advocates will argue it is an undercount because it excludes the players who never rebuy at all because they hit a limit first. Both are right.

The question that matters is narrower. Croatia's autoplay rule assumes that a reset creates a decision point. The data says it creates a six-second inconvenience for most players and nothing at all for the ones the rule was written to protect. So: does a reset that can be dismissed in one tap satisfy a requirement whose stated purpose is to interrupt momentum? The Ministry has not had to rule on it. The first enforcement action on a soft-reset implementation will settle it, and on current evidence that action is a matter of when, not whether.