A pre-commitment tool that let players set a hard stop on slot play cut average session length by 22% across a 14-week trial of 3,100 U.S.-facing accounts, while return-to-player on those sessions held flat at 96.4%. The drop came almost entirely from players who had previously logged sessions over 90 minutes. Nobody's RTP moved. That's the finding worth sitting with, because it undercuts the two loudest arguments in the responsible-gambling debate at once.
What the study actually measured
The setup was a forced-choice limit screen: on login, players picked a session cap (30, 60, or 120 minutes) or explicitly opted out. Roughly 61% set a cap; the rest declined. Researchers tracked session length, spin volume, average bet, and realized RTP per session over the full window.
The 22% figure is a blended average, and the blend matters. Players who chose the 30-minute cap cut their sessions by 41%. Players who chose 120 minutes barely changed anything — a 6% reduction. The opt-out group was flat. So the headline number is really a story about the middle of the distribution: people who picked 60 minutes and mostly stuck to it.
The RTP non-finding is the point
Realized RTP on capped sessions was 96.4% versus 96.3% on uncapped sessions — inside the noise band for a sample this size. Slot RTP is a property of the math model and the spin count, not the clock. A player who stops at 60 minutes instead of 100 has fewer spins, which means more variance in their realized return, not a better or worse expected value. Any operator or affiliate claiming a session limit "improves your odds" is selling something that doesn't exist. What limits change is exposure, and exposure is where bankrolls actually die.
Where the playtime went
The 22% cut didn't translate into a proportional revenue hit, which is the part operators will care about. Average bet size rose 8% among capped players — plausibly a substitution effect, where a player who knows the clock is running bets a bit more per spin. Net coin-in per session fell about 14%, not 22%. The house edge applied to a smaller handle is a smaller expected loss, but the gap between 22% and 14% is real money.
There's a caveat the researchers flagged: the trial ran during a promotional period at two of the participating brands, and the bet-size bump may be promotional, not behavioral. A cleaner read needs a promo-free window.
The compliance angle for U.S. operators
As of the 2024–2025 state rollouts, seven jurisdictions require some form of session-limit or time-out tool, but only two mandate that it be offered before first deposit rather than buried in account settings. The trial's design — a forced choice at login, with an explicit opt-out — is closer to the stricter model. The opt-out rate of 39% is the number regulators should study. A tool that four in ten players decline isn't a failure, but it does mean voluntary limits alone won't move population-level harm metrics much.
The open question is whether the 22% holds when the limit isn't presented as a fresh feature. Novelty tends to inflate early compliance; players who feel watched tend to route around friction. A 12-month follow-up with no promotional confound would tell us whether session caps are a durable behavioral nudge or a one-time novelty effect — and whether the bet-size substitution is a real trade-off operators will have to price in.