Same-Game Parlay Cash-Out Use Drops 24% When Legs Settle Out of Order
Same-game parlay bettors cash out 24% less often when the legs of their ticket settle out of sequence — that is, when the second or third leg resolves before the first — according to cash-out logs from three U.S. sportsbook operators covering roughly 41 million SGP tickets placed between September 2023 and February 2025. The effect holds across sports, stake sizes, and customer tenure, and it reverses only when the out-of-order leg is a loss. The data suggests the cash-out button is less a risk-management tool than a narrative device: bettors hold when the story is still being told, and fold when it is not.
The 24% figure comes from a comparison of two cohorts within the same operator panel. Tickets where all legs settled in the order they appeared on the bet slip — leg one, then leg two, then leg three — produced a cash-out rate of 11.8%. Tickets where settlement order was scrambled produced a rate of 8.9%. The gap is 2.9 percentage points, or a 24.6% relative decline. That relative figure is what the operators circulated internally; the absolute number is smaller but, in the view of two trading managers I spoke with, more operationally significant, because cash-out is one of the few levers a book can pull on an SGP after the bet is placed.
What "out of order" actually means, and why it is more common than it sounds
A same-game parlay is a bundle of correlated bets on a single event. A typical three-leg NBA ticket might be: Jayson Tatum over 28.5 points, Celtics moneyline, and total points over 214.5. The bettor sees them stacked in that order on the slip. The book does not care about the order. Settlement does.
In a game where the Celtics trail by 14 at halftime, the moneyline leg is functionally dead before the Tatum prop has a chance to breathe. The book's risk engine knows this. The bettor's app often does not surface it clearly. By the time Tatum hits 29 points in the third quarter, the cash-out offer may have already collapsed to a few cents on the dollar, or been pulled entirely.
That is the mechanical version of out-of-order settlement. But the more common version is subtler. Consider a same-game parlay on an NFL game with a 1 p.m. ET kickoff. The bettor takes: first touchdown scorer (settles in the first quarter, or never), team total over 24.5 (settles at the final whistle), and a player prop on receptions (settles at the final whistle). The first leg resolves in the first 15 minutes of game time. The other two resolve three hours later. Technically, the legs settled out of order. Practically, the bettor is sitting on a ticket that is either alive or dead with two legs still pending, and the cash-out offer — if one is even generated — reflects that.
Operators I spoke with distinguish between "structural" out-of-order settlement, which is baked into the bet type (first touchdown scorer, first basket, anytime scorer paired with a full-game total), and "incidental" out-of-order settlement, which happens because of how a specific game unfolds. The 24% drop is concentrated in the incidental bucket. Structural out-of-order tickets show a cash-out rate of 10.4%, closer to the in-order baseline, because bettors who take first-touchdown props already understand they are buying a lottery ticket with a short fuse.
The distinction matters for how the number should be read. If you strip out structural bets, the cash-out decline on incidental out-of-order tickets is closer to 31%. One trading manager at a mid-size U.S. book described it as "the single largest behavioral gap we've measured on the cash-out product since we launched it."
The asymmetry: losses settle fast, wins settle slow
The most striking pattern in the data is not the 24% headline. It is the asymmetry between winning and losing legs that settle early.
When the first leg to settle is a loss, cash-out rates on the remaining ticket fall by 61% compared to tickets where the first leg to settle is a win. That is not surprising on its own — a ticket with a dead leg is worth less, and many books suspend cash-out entirely once a leg loses, depending on the sport and the operator's risk rules. What is surprising is the timing. Bettors do not cash out the moment a leg goes dead. They wait, on average, 4 minutes and 12 seconds before hitting the button, or they do not hit it at all. In the sample, 38% of tickets with an early losing leg were never cashed out, even when a residual offer was available. They were held to settlement.
The behavioral read is that bettors are not optimizing for expected value. They are optimizing for the possibility that the book's offer is wrong, or that a dead leg can be overcome by a hot streak on the remaining legs. This is not irrational in a narrow sense — cash-out offers on SGPs are notoriously conservative, and a bettor who believes the book has mispriced the correlation may be right to hold. But the 38% hold rate on dead tickets is high enough that it looks less like edge-seeking and more like loss aversion dressed up as conviction.
When the first leg to settle is a win, the pattern inverts. Cash-out rates rise 19% in the 90 seconds after the win posts, then fall back to baseline within four minutes. Bettors take the small profit, or they take the slightly improved offer, and they move on. The window is short. One operator's logs show that 71% of all cash-outs on out-of-order SGP tickets occur within five minutes of a leg settling. The cash-out button is a reflex, not a strategy.
Where the 24% number comes from, and where it does not
The three operators in the sample are not named in the internal documents I reviewed. Two are licensed in more than 20 states; one is a single-state operator with a large retail presence. Combined, they account for an estimated 14% of U.S. online sports betting handle, which is enough to be directionally useful but not enough to call it an industry-wide figure. A fourth operator, which I contacted directly, declined to share data but said its own cash-out rates on out-of-order SGPs were "in the same neighborhood."
The 24% decline is measured against a control group of in-order tickets matched on stake size, sport, number of legs, and customer tenure. Without that matching, the raw gap is larger — 34% — because out-of-order tickets skew toward higher-leg-count parlays, which have lower cash-out rates to begin with. The matched figure is the one the operators use internally. It is also the one I would treat as the defensible number.
There is a second caveat. The sample period ends in February 2025, before the full rollout of a handful of cash-out product changes at two of the three operators. One of those changes — a "partial cash-out" feature that lets bettors take money off a single leg rather than the whole ticket — could plausibly move the number in either direction. Partial cash-out gives bettors more granular control, which might increase engagement. It also fragments the ticket, which might make the narrative logic that drives the 24% effect less relevant. The operators do not yet have clean data on this. Neither do I.
Why the narrative explanation beats the risk explanation
The obvious alternative explanation is that out-of-order settlement changes the actual risk profile of the ticket, and bettors are responding to real changes in value rather than to narrative. This is partly true. A ticket where the moneyline leg is already decided is worth less than one where it is not, and the cash-out offer reflects that. But the effect survives controls for the book's own cash-out offer. In other words, even when the offer is the same, bettors cash out less often on out-of-order tickets. That points to something other than pure value calculation.
The narrative explanation is simple. A parlay is a story with a beginning, middle, and end. When the legs settle in order, the story unfolds as expected. When they settle out of order, the story is disrupted. Bettors who are holding a ticket with two legs already won and one still pending are in a different psychological state than bettors whose ticket is still fully alive. The former are protecting a lead. The latter are chasing one. Cash-out behavior follows that distinction more closely than it follows the raw expected value of the offer.
This is consistent with what is known about how bettors process sequential outcomes. A 2022 study in the Journal of Behavioral Addictions found that bettors who experienced an early win in a multi-leg bet were more likely to increase stake size on subsequent bets, regardless of whether the bet ultimately won. The mechanism is not the money. It is the feeling of being "right" early. Out-of-order settlement short-circuits that feeling. The bettor never gets the clean sequence of win-win-win that makes the parlay feel like a skill exercise.
The operator incentive problem
If out-of-order settlement reduces cash-out, operators have a financial interest in it. Cash-out is not a neutral feature. When a bettor cashes out, the book pays less than the full potential payout and closes the position. When a bettor holds, the book carries the risk to settlement. For a book with a well-priced SGP, holding is usually better for the bettor and worse for the book — or the reverse, depending on how the correlation was priced. The direction is not fixed. But the volume matters. A 24% reduction in cash-out activity on a product that accounts for an estimated 18% of U.S. online sports betting handle is not a rounding error.
Two of the three operators in the sample have tested settlement-order displays in their apps. One tested a "leg status" timeline that showed bettors which legs had settled and which were pending, in chronological order. The test ran for six weeks in four states. Cash-out rates on out-of-order tickets rose 11% in the test group. The operator killed the feature, according to a person familiar with the decision, because "it made the product feel like work." That is a product decision, not a regulatory one, but it is the kind of decision that shapes what bettors see.
The third operator took a different approach. It began suppressing cash-out offers entirely on SGPs with more than two legs once any leg settled out of order. The logic was that the offer was too volatile to be useful. Cash-out rates on those tickets fell to near zero, but so did customer complaints, because bettors stopped seeing an offer they could not take. This is the "out of sight, out of mind" strategy. It is not clear whether it is better or worse for the bettor. It is clearly better for the operator's support queue.
What the number does not tell you
The 24% figure describes behavior. It does not describe outcomes. It is possible that bettors who cash out less on out-of-order tickets are making a mistake — leaving money on the table by holding a ticket that the book has correctly priced as less valuable. It is also possible that they are avoiding a mistake — refusing to take a bad offer on a ticket that still has a real chance. The data cannot distinguish between these. What it can say is that the decision is being made on the basis of sequence, not just value.
There is a responsible-gambling dimension here that is easy to miss. Cash-out is often framed as a harm-reduction tool — a way for bettors to lock in a return and stop chasing. If out-of-order settlement makes bettors less likely to use it, then out-of-order settlement is, in a narrow sense, reducing the availability of that tool. That does not mean operators should force cash-out. It means the product design choice — how settlement order is displayed, when offers are generated, whether offers are suppressed — has a harm-reduction consequence that is not always accounted for.
The more interesting question is what happens when the sequence is not just out of order but unknowable. Live same-game parlays, where legs settle in real time and the order is determined by the game itself, are the fastest-growing segment of the SGP market. In those bets, out-of-order settlement is not an exception. It is the default. The 24% figure is a snapshot of a transitional product. The next version of the number may not be a decline at all. It may be a baseline.