Deuces Wild Video Poker
A positive-expectation video poker variant, solved exactly and priced for a bad read.
This solves exact optimal-play returns for four Deuces Wild pay schedules and any schedule typed in, and prices what one wrong hold costs on top. It says nothing about how any individual session will go.
Deuces Wild is the one video poker family with a genuinely positive-expectation schedule in wide circulation, and it is also the one where optimal strategy is the least forgiving: every deuce dealt changes what a hand is, not just what it pays.
Each card reads off six rows, wild royal flush down to flush.
The schedule cited everywhere as returning over 100% at flawless play. Rare outside a handful of locals casinos and cruise ships.
A gold card is a deuce, standing in for whichever card the hand needs.
Exact optimal-play solution for Full pay Deuces Wild, computed here over every one of the 2,598,960 five-card deals, with every deuce resolved to whichever card makes the best hand rather than looked up.
What a hand actually returns
55%
of hands return nothing
Typical result: nothing on the middle hand
hands rarer than 1 in 1,000every other paying hand
| Hand | Chance | Returns | Share of return |
|---|---|---|---|
| Natural royal flush | 0.0022%1 in 45,282 | 800 for 1 | 1.8% |
| Four deuces | 0.020%1 in 4,909 | 200 for 1 | 4.0% |
| Wild royal flush | 0.18%1 in 557 | 25 for 1 | 4.5% |
| Five of a kind | 0.32%1 in 312 | 15 for 1 | 4.8% |
| Straight flush | 0.42%1 in 240 | 9 for 1 | 3.7% |
| Four of a kind | 6.49% | 5 for 1 | 32.2% |
| Full house | 2.12% | 3 for 1 | 6.3% |
| Straight | 5.61% | 2 for 1 | 11.1% |
| Flush | 1.68% | 2 for 1 | 3.3% |
| Three of a kind | 28.5% | stake back | 28.3% |
| Nothing | 54.7% | nothing | 0.0% |
The number on the glass, an average over every hand at flawless strategy. Every deuce dealt is resolved to whichever real hand it can complete that pays the most, which is a different and larger search than deciding which of five cards to keep on a non-wild game.
Exact enumeration, not a simulation and not an interpolation. Every deal is evaluated against all 32 holds and every possible draw, with each deuce priced as whichever card it is best resolved to, and the best hold is taken. See how this is calculated, below.
What this table costs to sit at
5 coins a hand, which is $1.25 at this denomination. Max coin is assumed throughout: below it the royal pays 250 rather than 800 and every return on this page is too high.
Machine concedes / hour
At flawless play, a ceiling of $5.71
- Through the machine / hr
- $750
- Over 3 hours
- At flawless play, a ceiling of $17.14
This is the best Deuces Wild schedule on the list (Full pay), so there is no better version of this game here to walk to. It still has a cost, and it is the figure above.
This schedule returns more than 100% at flawless play, so the figure above is what the machine concedes rather than what it takes. Read it as a ceiling, not as income: it assumes the best hold on every deal, and the "How much rides on picking correctly" section below shows how much a single wrong read can give back.
- House edge
- -0.76%
- Four deuces
- 1 in 4,909
- Wild royal flush
- 1 in 557
How much rides on picking correctly
Every deal has one best hold and one second-best, and this engine already finds both while it searches. Averaged over all 2,598,960 deals, taking the second-best hold instead of the best gives up 5.7c per coin staked. This is not an estimate of how often a player actually errs, and no real player takes the worst-of-two-best choice on every hand. It is the cost of that one specific policy, not a ceiling on every kind of misread (holding outside the top two costs more): at 25c a coin and 600 hands an hour, always taking the second-best hold would cost an extra $171 an hour beyond the table's own edge, which is the size of the gap this particular wild-card misread opens between a close-enough read and the right one.
Second-best hold on every deal
5.7c
given up per coin staked
Machine concedes / hour$5.71
Extra from the second-best hold / hour$171
Every figure here is an expected cost, not a forecast of your night, and it assumes perfect play on every hand. Full-pay Deuces Wild returns over 100% only at that flawless play, on a schedule that is rare outside a handful of casinos: the machine in front of you is more likely to be one of the other schedules on this list. Video poker is gambling: play only with money you can afford to lose, and only where it is legal for you to do so. 21+.
Written and reviewed by Alexander Korsager · last reviewed October 7, 2026 · corrections
The idea
What this tool shows
Deuces Wild is famous for one fact and light on the honest version of it: at the full-pay schedule, the game returns 100.76% to a player who never misses a hold, which is one of the very few pay tables in a casino that pays back more than it takes. What gets repeated far less is that the schedule is rare outside a handful of locals casinos and cruise ships, that most machines carrying the name pay meaningfully less, and that the strategy needed to reach that number is genuinely unforgiving: every deuce dealt changes what a hand actually is, not just what it pays, so a hold that would be routine on Jacks or Better can be wrong here.
This tool solves four real Deuces Wild schedules exactly, the same way the four non-wild games on this site are solved: every one of the 2,598,960 possible five-card deals is evaluated against all 32 ways to hold or discard, with each held deuce resolved to whichever real card completes the best hand rather than counted as a plain two. Nothing here is looked up from a frequency table. Edit any row and the browser solves your own table from scratch, in the same couple of seconds the site's other video poker page takes.
Below the return sits a figure no other free Deuces Wild tool publishes: what it actually costs, in cents per coin and in dollars an hour, to take the second-best of the 32 holds instead of the best one, computed exactly rather than assumed. It is not an estimate of how a real player plays. It is a measure of how much is riding on getting the read right, on a game whose reputation for paying over 100% depends entirely on doing that every single time.
Video poker is gambling with a house edge on every schedule here except the rare full-pay one, and even that one only clears 100% at flawless play that essentially nobody sustains. Nothing on this page identifies a way to win; it identifies exactly what each schedule costs, and what a mistake on top of it costs too.
FAQ
Questions players ask
Is Deuces Wild really the only positive-expectation video poker game?
Among schedules in real circulation, full-pay Deuces Wild (royal 800, four deuces 200, wild royal 25, five of a kind 15, straight flush 9, four of a kind 5, full house 3, flush 2, straight 2, three of a kind 1) returns 100.76% at flawless play, which this tool's own enumeration confirms rather than assumes. It is rare: most machines carrying the Deuces Wild name pay one of the shorter schedules also on this page, several of which return under 99%. The other three video poker games this site solves (Jacks or Better, Bonus Poker, Double Bonus, Double Double Bonus) can also clear 100% on their own full-pay tables, so "the only" overstates it; what is true is that Deuces Wild's full-pay table is one of the better-known and more widely searched-for examples.
Why does a 100.76% return not mean the house is guaranteed to lose?
Because 100.76% is an average over every one of the 2,598,960 possible deals at flawless strategy, not a guarantee on any one session. A player can and regularly does lose money on a positive-expectation machine over a few hours; the figure describes what happens if the same schedule were played an enormous number of times with no mistakes. The section on this page pricing a single wrong hold is the other half of the honest picture: strategy errors that are common on other video poker games are the difference between the advertised 100.76% and what a real player actually sees.
What makes Deuces Wild strategy harder than Jacks or Better?
A deuce is not a card with a fixed value, it is a placeholder for whichever card makes the best hand, so holding one changes what every other card in the hand is worth. A hold that discards a deuce is almost never correct, but beyond that the right play depends on how many deuces are dealt and what they are dealt alongside, in a way that has no equivalent on a non-wild game: this tool's engine evaluates the four deuces case, wild royals, and wild five-of-a-kind as their own distinct outcomes because the strategy genuinely branches on them.
Does this tool cover every Deuces Wild pay table?
No. It solves four schedules that are documented and in real circulation, from full pay down to a common short-pay table, and it solves any schedule you type in yourself. Casinos and video lottery terminals run further variants beyond these four; typing the numbers off the machine in front of you into the editable table solves that exact schedule rather than approximating it from the nearest listed one.
How is this different from the other video poker tool on this site?
The other tool solves Jacks or Better, Bonus Poker, Double Bonus and Double Double Bonus, none of which involve a wild card, and it says so rather than approximating Deuces Wild from that engine. A wild card changes what a hand of five cards evaluates to, not only what a schedule pays for it, so it needed its own solver rather than a fifth entry on that page.
Methodology
How this is calculated
Every one of the 2,598,960 possible five-card deals is evaluated against all 32 ways to hold or discard, with each deuce in a held combination resolved to whichever real card completes the highest-paying category the hand can reach.
The best-paying hold for each deal is taken, and the return is the average payout over all 2,598,960 deals at those holds.
The engine that does this is shared with the site's non-wild video poker tool; only the hand classifier and the category set differ, so the two pages solve from the same Mobius-transform machinery rather than two independent implementations.
Common misreadings3
- A return over 100% is not a guarantee of winning. It is an average at flawless play over millions of deals; any single session can lose money on a positive-expectation machine, and most do at least some of the time.
- The full-pay schedule most people mean when they say "Deuces Wild returns over 100%" is uncommon. Checking the actual pay table on the machine in front of you against this page's presets, or typing it in directly, is the only way to know which game you are actually sitting at.
- A low listed house edge does not describe how a single hour will go. A positive or near-breakeven return still carries real variance, and the second-best-hold figure on this page shows how much a strategy slip can give back even before variance is considered.
Assumptions5
- Max coin (5 coins) throughout, which is what every figure on this page assumes. Below max coin the royal pays 250 for 1 rather than 800 and every return here is too high.
- The cost-per-hour and mistake-cost-per-hour figures assume 600 hands an hour, the same pace the site's video poker page and house-edge comparison use for the format (
video-poker/session.ts'sDEFAULT_HANDS_PER_HOUR, shared by both pages rather than reimplemented here). It is an editable input, not a fixed constant, and both dollar figures scale linearly with it. - Optimal strategy on every hand: the best of the 32 holds by expected value, found by exhaustive search rather than a memorized strategy chart. A real player's return sits below the figures here by however much their own play deviates from it, which is what the second-best-hold figure on this page prices.
- The category hierarchy (natural royal, four deuces, wild royal, five of a kind, straight flush, four of a kind, full house, flush, straight, three of a kind, nothing) is fixed and does not depend on the pay schedule: a wild deuce is always resolved toward the highest category reachable. This holds for every schedule this tool solves; a schedule engineered so that a lower category paid more than a nominally higher one could in principle invert part of this order, and none of the four presets or any schedule we tested does.
- Four documented schedules ship as presets: full pay (100.76%), NSUD (99.73%), Illinois (98.91%) and a fourth short-pay table (98.94%), all sourced to Wizard of Odds' Deuces Wild pay table appendix, retrieved 2026-09-12. Any other schedule is solved from what is typed in, not interpolated from these four.
Limits3
- This prices hands, not sessions. It says nothing about how a bankroll behaves over a number of hands, which is the Bankroll Calculator and Variance Simulator's job.
- The second-best-hold figure is the cost of one specific policy, always taking the second-best hold, not a ceiling on every kind of misread: a hold outside the top two costs more. It is computed exactly from the same enumeration as the return, it is not a model of how often real players make that specific kind of error, and it should not be read as an estimate of typical player loss beyond the return figure itself.
- Only the ten paying categories and "nothing" are modeled. Any casino rule beyond the pay schedule itself (bonus jackpots on a progressive machine, multi-hand variants) is out of scope.
Evidence
The shared engine is checked in engine.test.ts, and its Deuces Wild classifier (classify.ts) is checked in classify.test.ts: the classifier is pinned against 20+ hand-constructed deals covering every category (natural royal, four deuces at 1/2/3 wild counts, wild royal, five of a kind, wild straight flush including the wheel, wild four of a kind and full house, and the boundary where a single wild cannot reach three of a kind), the full 2,598,960-deal enumeration is checked against textbook counts (4 natural royals, 48 four-deuces hands), and all four documented schedules are solved and checked against Wizard of Odds' own published returns, to two decimal places (full pay 100.76%, NSUD 99.73%, Illinois 98.91%, cut wild royal 98.94%), retrieved 2026-09-12.
solved.test.ts recomputes every preset from the engine on every run and fails on any disagreement with the committed figures, the same gate the site's non-wild video poker tool runs on its own solved.test.ts.
Gambling can cause serious harm. The National Problem Gambling Helpline is free, confidential and open 24/7, call or text 1-800-MY-RESET (1-800-697-3738), or chat at ncpgambling.org/chat.
Related tools
Try next
- Video Poker Return CalculatorType in any pay table and get the exact optimal-play return, solved over all 2,598,960 deals.
- House Edge ComparisonThe house edge on every common casino bet, side by side, and what it costs you per hour.
- Card Counter TrainerPractice the Hi-Lo count
- Poker Hand OddsRun a Hold'em matchup