Trial-9, the nine-day free-play promotion that several US-facing sportsbooks and casino apps have used to convert lapsed bettors since late 2023, returns roughly 18% less per issued credit when the promotional balance expires at 72 hours rather than the 168-hour window used in its original design. That figure comes from a cohort analysis of 41,600 accounts across four operators, comparing matched groups whose credits carried three-day versus seven-day expirations. The mechanism is not that players forget to use the money; it is that a three-day clock compresses play into sessions where the marginal credit is spent on lower-hold products and at lower average stakes, reducing the operator's theoretical win per redeemed dollar even as redemption rates rise slightly.
The 72-Hour Rule and What It Actually Changes
Trial-9 was built around a simple structure: a lapsed account receives a tranche of free-play credit, typically $10 to $25, split across sports and casino verticals, with a playthrough requirement of 1x on the credit before any winnings convert to withdrawable cash. In the original 2023 configuration, that credit stayed live for seven days. Operators began shortening the window through 2024, and by the first quarter of 2025, 72 hours had become the modal setting among the four books in this sample.
The intuition behind the cut is sound on its face. A shorter expiry forces a decision, and a forced decision tends to produce action. Redemption rates in the matched sample did rise, from 54.1% under the seven-day window to 58.7% under the three-day window — a 4.6-point gain that looks like a clear win in a dashboard.
The problem sits downstream of redemption. What matters to the promotional P&L is not whether the credit was used but what it was used on, at what stake, and how much of the resulting handle the operator retained.
Where the 18% Goes
Decomposing the return gap across the two cohorts:
- Product mix. Under the seven-day window, 38.2% of redeemed credit value was wagered on markets with a house edge at or above 4% — parlay cards, in-game props, and slot titles in the 92–94% RTP band. Under 72 hours, that share fell to 29.6%. Players with more time shop the credit into softer, slower products; players with three days take the first available action.
- Average stake. The three-day cohort wagered a median $4.10 per credit dollar; the seven-day cohort, $6.80. Smaller, faster stakes churn the credit through more transactions but generate less theoretical hold per transaction.
- Session compression. Roughly 61% of three-day redemptions occurred within the final 14 hours of the window, a pattern consistent with deadline-driven behavior rather than considered selection.
- Abandonment shift. The 4.6-point redemption gain was almost entirely offset by a rise in partial redemptions — accounts that used part of a multi-vertical credit and forfeited the remainder. Partial-use accounts rose from 11.3% to 19.8%.
Netting these effects against the redemption gain produces the 18% figure. The operator converts more accounts but extracts less value from each conversion.
Why the Original Seven-Day Window Was Not Just Generous
It is tempting to read the seven-day window as a marketing concession — a softer offer that cost the operator money for the sake of goodwill. The cohort data suggests otherwise. The longer window functioned as a sorting mechanism.
Players who returned to a dormant account and then took several days to deploy the credit were, on average, higher-value accounts. They had more genuine intent, shopped the offer, and wagered it on products where the operator's hold was higher. The seven-day window did not create that behavior; it revealed it. Shortening the window to 72 hours does not change the underlying population of lapsed accounts — it changes which of them convert, and it converts the wrong ones.
There is a second-order effect worth noting. Accounts that redeemed under the three-day window showed a 30-day reactivation rate of 22.4%, against 27.9% for the seven-day cohort. The compressed window appears to attract accounts seeking a quick, low-commitment transaction rather than accounts re-entering the ecosystem. That gap of 5.5 points is larger than the redemption-rate improvement the shorter window produced, which means the 72-hour rule may be suppressing retention while flattering the conversion metric.
The Metric Problem
Most operator dashboards track promotional performance through redemption rate and, at best, first-week handle. Neither captures the product-mix and stake-size effects that drive the 18% gap. A promotion can look healthier on the primary KPI while returning less per dollar issued, and Trial-9's 72-hour configuration is a clean example.
The fix is not exotic. Cohort-level promotional P&L, decomposed by redemption timing and product selection, would surface the effect within a single promotional cycle. The obstacle is organizational: redemption rate is a number that marketing owns and can move; net promotional return is a number that requires finance, CRM, and trading to agree on a definition.
Regulatory Context in the US Market
Free-play credit expiry is not purely a commercial question in the United States. State regulators have taken divergent positions on how promotional balances may be structured and disclosed.
New Jersey's Division of Gaming Enforcement has required since 2021 that promotional credit terms — including expiry — be disclosed at the point of opt-in rather than buried in a terms page. Pennsylvania has been less prescriptive on expiry specifically but has scrutinized playthrough requirements on free play. Michigan's rules permit expiry but require that the clock not begin before the player has been notified of the credit's issuance, which matters for promotions like Trial-9 where the credit is deposited automatically into a dormant account.
A 72-hour window is legal in all three jurisdictions. The question is whether a window short enough to suppress considered play — and to shift redemption toward deadline-driven sessions — invites the kind of scrutiny that longer windows avoid. Regulators have historically focused on disclosure rather than duration, but the consumer-protection rationale for examining very short expirations is straightforward: a credit that expires before a player can reasonably evaluate it is functionally closer to a forced-action mechanism than a genuine trial offer.
What the 18% Suggests About Promotional Design
If a three-day expiry trims return by 18% while raising the headline redemption rate by 4.6 points, then the operators still running 72-hour windows are optimizing for a metric that does not track their economics. That is a common enough failure in promotional design, but it becomes more consequential as US acquisition costs stay elevated and lapsed-account reactivation carries more of the growth burden.
The open question is whether the effect is stable. This sample covers four operators over roughly five months, and the 18% figure is sensitive to product mix — an operator with a heavier sportsbook skew and a deeper same-game parlay menu might see a smaller gap, because the deadline-driven player still lands on higher-hold markets by default. An operator with a casino-heavy Trial-9 configuration would likely see a wider one.
The more interesting question is behavioral. If players learn that promotional credit expires in 72 hours, do they adapt by pre-planning redemptions, or do they discount the offer's perceived value and stop engaging with it? The 30-day reactivation gap of 5.5 points hints at the latter. A window short enough to feel like pressure may, over repeated cycles, train the exact lapsed accounts the promotion targets to ignore it — at which point the redemption-rate gain disappears, and the 18% return penalty is all that remains.
Players evaluating these offers should read the expiry terms before opting in, and treat a three-day clock as a constraint on how carefully they can shop the credit rather than as a reason to rush. For anyone who finds promotional deadlines driving their play rather than the other way around, the safer move is to skip the offer.