Editorial illustration for the lesson on let it ride: pull the bets back, in the Mayfair Casino School.
Editorial illustration for the lesson on let it ride: pull the bets back, in the Mayfair Casino School.

Let It Ride: pull the bets back

Let It Ride lets you pull two of your three bets back before the hand resolves. Most players don't pull back often enough.

AC
Annabel Cavendish
Editor in Chief · Reviewed 14 May 2026
Annabel
0:000:00

Welcome to the lesson on Let It Ride.

I'm Annabel, and Let It Ride occupies an unusual position in the casino poker family.

It was invented by Shuffle Master in nineteen ninety-three and was, at the time, a significant innovation in carnival game design.

The pull-back mechanic, the ability to take money off the table mid-hand based on new information, was genuinely new.

It remains the game's defining structural feature, and understanding it correctly is the entire point of optimal play.

Here's the setup.

You place three equal bets in designated spots on the table layout, conventionally labelled one, two, and the dollar sign.

The dollar-sign bet is always in play; you cannot pull it back under any circumstances.

You then receive three hole cards face-down.

After seeing them, you decide whether to pull back bet one or let it ride.

The dealer then turns over the first of two community cards, giving you a partial view of the developing hand.

You decide whether to pull back bet two or let it ride.

The dealer turns the second community card.

All five cards are evaluated, your three hole cards plus the two community cards, and if your hand is tens or better, you're paid according to the pay table on any bets that remain in action.

The critical insight is this: letting a bet ride is correct in a minority of situations.

The decision to leave money on the table is a decision to invest in a hand that has genuine strength already, not a decision to hope for improvement on a weak holding.

With three cards, the conditions for letting bet one ride are specific.

Let it ride with any paying hand already in your three cards: a pair of tens or better, or three of a kind.

Let it ride with three to a royal flush, three suited cards that could make a royal.

Let it ride with three suited consecutive cards, with the exception of two-three-four or ace-two-three, where blocked outs make the combination too weak.

Let it ride with three to a straight flush with a spread of four or less and at least one high card of ten or better.

Let it ride with three to a straight flush with a spread of five and at least two high cards.

Pull back bet one on everything else.

That means most three-card combinations, including hands that feel promising such as three medium cards, unsuited connectors, or an ace with two mismatched low cards.

Use it without hesitation.

After the first community card, you have four cards.

The conditions for letting bet two ride are slightly more permissive.

Let it ride with any paying hand, any four cards to a flush draw, any four to an outside straight with at least one high card, any four to an outside straight with no high cards, or any four to an inside straight with four high cards.

Pull back bet two on inside straights with fewer than four high cards, and on anything that doesn't meet these conditions.

The inside straight trap is worth naming explicitly.

Holding five-six-eight-nine and hoping for a seven is a pull-back situation.

The gut feel that you're "close" is not reflected in the mathematics.

The combined effect of pulling back correctly is to reduce your effective hourly cost below the headline three point five one percent edge.

The element of risk, which is the edge expressed relative to all money actually wagered across all three bet positions, works out to approximately two point eight five percent when you follow the pull-back strategy correctly.

The standard pay table pays even money on tens or better, two to one on two pair, three to one on three of a kind, five to one on a straight, eight to one on a flush, eleven to one on a full house, fifty to one on four of a kind, and two hundred to one on a straight flush.

The high payouts for straights and better mean the game has elevated variance compared to its edge.

Sessions swing more widely than the edge figure alone suggests.

The standard deviation relative to a single bet is approximately five point one seven, higher than most other carnival games.

The pace is slow, roughly thirty to thirty-five hands per hour at a live table.

At Aspers Westfield Stratford, Let It Ride runs noticeably more slowly than Three Card Poker on adjacent tables.

The slower pace means your expected hourly cost at comparable stakes is lower in absolute terms, even if the edge percentage is similar.

Pull back early and often.

The dollar-sign bet is the one you're always defending.

Start with the mechanics, because the pull-back option is what distinguishes this game from every other carnival table.

Let It Ride was invented by Shuffle Master (now Light & Wonder) in 1993 and introduced to casino floors that year. The concept was deliberately designed around the pull-back mechanic: unlike every other carnival game where you post a bet and see what happens, Let It Ride lets you take money off the table mid-hand based on new information. It was a breakthrough in carnival game design and helped establish Shuffle Master as the dominant casino game licensor of the 1990s.

The structure: you place three equal bets in designated spots on the table layout, conventionally labelled 1, 2, and $. The $ bet is always in play and cannot be returned. You then receive three hole cards face down. After seeing them, you decide whether to pull back bet 1 or let it ride. The dealer then turns over the first community card (of two), which combines with your three hole cards to give you a partial view of the developing hand. You decide whether to pull back bet 2 or let it ride. The dealer turns the second community card. All five cards are evaluated: your three hole cards plus the two community cards. If your hand is tens or better, you're paid according to the pay table on all remaining bets.

The Pull-Back Strategy: Three Cards

The critical insight is that letting a bet ride is correct only in a minority of situations. With three cards, we confirm you should let bet 1 ride only when you have one of the following: any paying hand already (a pair of tens or better, three of a kind); any three to a royal flush (three suited cards that could make a royal); three suited consecutive cards except 2-3-4 or ace-2-3 (those combinations have too many blocked outs); three to a straight flush with a spread of four or less and at least one high card (ten or better); or three to a straight flush with a spread of five and at least two high cards.

Everything else: pull back bet 1. That means most three-card combinations, including things that feel promising like three medium cards, unsuited connectors, or an ace with two rags, should have their bet retrieved. The felt for pulled-back chips is right in front of you. Use it without hesitation when the conditions don't apply.

The reason the threshold is high is that two community cards still need to come, but the probability of turning a non-qualifying three-card holding into a paying five-card hand is low enough that the expected value of keeping the bet in action is negative in those cases. The maths on letting a marginal three-card combination ride is structurally similar to calling a raise on a weak hand in poker: you're paying for the chance of improvement in a situation where improvement is unlikely and non-improvement is costly.

The Pull-Back Strategy: Four Cards

After the first community card, you now have four cards. The second pull-back decision uses a different and slightly more permissive set of conditions. Let bet 2 ride with: any paying hand; any four cards to the same suit (a flush draw); any four to an outside straight with at least one high card; any four to an outside straight with no high cards (this is essentially a zero house-edge situation); any four to an inside straight with four high cards.

Fold (pull back bet 2) on: any four to an inside straight with fewer than four high cards; any hand that doesn't meet the above criteria. Inside straights with limited high cards are the classic trap. Holding 5-6-8-9 and hoping for a 7 is a weak play: the single missing card and the absence of high-card value makes this a pull-back situation.

The combined effect of pulling back correctly is to make Let It Ride's effective cost lower than the headline 3.51% suggests. When you pull both bets back and play only the third ($) bet, your money at risk is one unit instead of three. The element of risk across the full session works out to 2.85%, which means Let It Ride at correct strategy is running closer to Casino Hold'em territory in terms of actual expected cost relative to the full amount wagered.

The Pay Table and What It Means for Variance

Let It Ride pays on pairs of tens or better. The standard pay table is: pair (tens or better) 1:1, two pair 2:1, three of a kind 3:1, straight 5:1, flush 8:1, full house 11:1, four of a kind 50:1, straight flush 200:1, royal flush 1,000:1. The high payouts for straights and better produce significant variance: sessions at this table swing more widely than the house edge figure suggests. The standard deviation, in our analysis, is 5.17 relative to a single bet, higher than most carnival games.

This variance profile makes Let It Ride a game where the session experience can feel very different from the expected value. It's possible to pull back most of your bets correctly, hit a flush draw, and win 8x on all three bet units still in action. It's also possible to follow perfect strategy all night and experience a dry run of pulling back nearly everything. The pull-back mechanic gives you the impression of active control while the underlying pay table creates binary sessions: either the flush and straight hits land and you're ahead, or they don't and you're steadily down on retained third bets.

At Aspers Westfield Stratford, where the carnival pit runs Three Card Poker alongside Let It Ride at some periods, the pace of Let It Ride's 30-35 hands per hour is noticeably slower than Three Card Poker's 45-50. The slower pace isn't an argument for or against the game, but it does mean the same expected hourly loss at similar stakes actually involves more clock time. Use the casino poker trainers to internalise the pull-back conditions for both the three-card and four-card stages before your first live session.

Key numbers

MetricValue
House edge (initial minimum bet)3.51%
Element of risk (all money in action)2.85%
Standard deviation per bet5.17
Hands per hour (live table)30-35
Expected loss per hour (£10 per bet, 3 bets, 30 hands)~£10.53
Pair (tens or better) payout1:1
Royal flush payout1,000:1

Sources: our Let It Ride analysis, Hippodrome Casino table games, Casinos.org.uk Aspers Stratford review.

Resume
Next lesson