Players who keep a slot's RTP label visible on screen during UK night sessions — roughly 00:30 to 04:00 — play sessions 21% shorter than players on the same games with the label hidden, according to a 14-week behavioural study of 8,412 verified UK accounts run across three mid-sized operators. The effect holds when you control for stake size, game volatility and deposit limits, which suggests the label itself is doing something to session length rather than simply tagging along with a different type of player.

That is a strange result, because the RTP figure is the one piece of information a slot player can look at that is both accurate and useless in the moment. A 96.2% RTP tells you what happens across millions of spins. It tells you nothing about the next 200. Yet the data suggests that putting it on screen changes how long people stay.

What the study actually measured

The sample covered 8,412 accounts that had opted into a responsible gambling dashboard between January and April 2024, all UK-licensed, all with at least 40 night sessions in the period. A night session was defined as any continuous play block starting between 00:30 and 04:00 with no gap longer than 12 minutes. Median night session across the whole sample was 38 minutes. Median for the label-visible group was 31 minutes; label-hidden, 39 minutes. That gap is where the 21% comes from.

Worth being blunt about the limitations. This is observational, not randomised. Players who toggle the RTP display on are self-selecting, and the dashboard opt-in skews toward people who already think about control. The operators involved also can't tell us whether the player was reading the label or just had it on because it was the default on that game's UI. And 14 weeks in one spring is not a full year — night session behaviour in January is not night session behaviour in July.

Still, the effect survived a few robustness checks that matter. When the researchers split by volatility band, the gap narrowed on high-volatility games (18%) and widened on low-volatility ones (26%). When they split by stake, it held at every band from 20p to £5. When they looked only at sessions that ended without hitting a bonus feature, the gap was 24%. That last one is interesting, because it removes the obvious confound — that label-visible players were simply luckier and quit while ahead.

The timing detail that complicates things

Session length dropped most sharply between 02:00 and 03:30. Sessions ending in that window were 34% shorter in the label-visible group than in the hidden group, against 14% for sessions ending between 00:30 and 02:00. That pattern doesn't fit a simple "information makes people more rational" story, because rationality shouldn't be time-dependent. It fits something closer to fatigue plus a visible number.

There's a plausible mechanism, and it's not flattering to anyone. At 2am, a player's working memory is degraded. A 96.2% figure sitting on screen is a small, concrete, graspable fact in a situation where almost nothing else is concrete. It may function less as a probability statement and more as a stopping cue — a nudge that says "this is a game with a house edge, you know this, you can stop now." The label isn't teaching anyone anything at 3am. It's reminding them of something they already knew when they were less tired.

If that's right, the effect should fade in players who play mostly in the daytime. It did, partially. Daytime sessions in the same sample showed a 9% gap, not 21%. Same players, same games, different hour.

Why operators have been slow to make this standard

UK-licensed operators are required to display RTP, but the requirement is about disclosure, not prominence. In practice that means a figure buried in a game info panel, or on a paytable screen two taps deep, or in the help file. Very few put it on the main game screen during play.

The commercial logic for that is straightforward and rarely stated out loud. An RTP label on the main screen is a persistent reminder that the game is -EV for the player. It doesn't stop anyone playing — the study doesn't show that, and no serious study has — but it does appear to shorten sessions, and shorter sessions mean less handle. If you're running a 4.5% margin on slots, a 21% reduction in session length is not a rounding error.

There's a counter-argument that operators should be making louder: shorter sessions from label-visible players may not reduce lifetime value at all. A player who plays 31 minutes at 3am and logs off may be more likely to come back the next night than one who plays 39 minutes and wakes up feeling worse about it. The study didn't measure retention, which is the single biggest gap in it. Anyone citing these numbers to argue either side of the policy question is overreaching until someone runs the 12-month version.

What the Gambling Commission has said, and hasn't

The Commission's 2023 consultation on remote gambling and software technical standards touched on RTP display but stopped short of mandating on-screen placement during play. The current LCCP requirement (RTS 13) is that RTP must be available to the player before they stake, which the info-panel approach satisfies. There's no requirement that it stays visible.

That's a gap worth naming. "Available before they stake" is a disclosure standard written for a player who is reading carefully and deciding whether to start. It's not written for someone 90 minutes into a session at 3am who needs a reason to stop. Those are different design problems, and the study suggests the second one is where the behavioural effect lives.

The uncomfortable version of this finding

If a visible RTP label shortens night sessions by a fifth, the obvious question is whether hiding it is a deliberate design choice with a measurable harm. The study can't answer that, and the operators who participated did so on condition of anonymity, which tells you something about how they read their own results.

But there's a second reading that's less comfortable for the harm-reduction side. If a number on a screen can shorten a session by 21%, then the number isn't doing the work — the interruption is. Any sufficiently concrete on-screen element might do the same thing. A session timer. A running net position. A deposit-remaining figure. The RTP label might just be the socially acceptable version of a stop sign, and the reason it works is that it's a stop sign, not because it's RTP.

That would mean the policy debate about RTP display prominence is somewhat beside the point. The real question is whether UK operators should be required to keep any persistent, concrete, session-relevant number on the main game screen during play — and if so, which one produces the best outcome for players who are tired, chasing, and not reading help files.

Nobody has run that study. It's the one worth funding.