Video poker: the 99.5% game almost nobody plays

Video poker looks like a slot machine and behaves like a solvable problem. You are dealt five cards, you choose which to keep, and the machine replaces the rest. Payouts follow a published table. What makes it unusual: • The odds are fully knowable. A fifty-two card deck has fixed probabilities, and the optimal play for every hand can be calculated exactly • Your decisions matter. Unlike slots, where you press a button, video poker gives you a real choice on every hand, and that choice moves the return substantially • The best versions return over 99% to a player using correct strategy — better than blackjack, better than baccarat, better than almost anything else in the building The full-pay Jacks or Better machine returns 99.54% with perfect play. That is a house edge under half a percent. The catch, and it is a serious one: that figure applies to one specific paytable. Machines that look identical, with the same name and the same graphics, can return several percentage points less. Almost nobody checks.

The entire difference between a great game and a mediocre one sits in two lines of the paytable: the payouts for a full house and a flush, per coin bet.

Two numbers, four and a half percentage points. That is a bigger difference than exists between most entirely different casino games, and it is invisible unless you look. How to check: open the paytable and find the row for a full house and the row for a flush, at the one-coin column. If they read 9 and 6, you have found the good machine. If they read 8 and 5, you are paying more than five times the house edge for the same game. Always bet maximum coins. The royal flush payout jumps disproportionately at maximum bet, and every return figure above assumes it. Playing fewer coins costs roughly one and a half percentage points. If the maximum bet is beyond your budget, drop to a lower denomination machine rather than betting fewer coins.

Perfect strategy is a large chart. This simplified version captures nearly all of the return for Jacks or Better. Work down the list and play the first line that applies to your hand. 1. Royal flush, straight flush or four of a kind — hold all five 2. Four cards to a royal flush — hold them, break anything to do it 3. Full house, flush or straight — hold all five 4. Three of a kind — hold the three 5. Four cards to a straight flush — hold them 6. Two pair — hold both 7. High pair (jacks or better) — hold the pair 8. Three cards to a royal flush — hold them 9. Four cards to a flush — hold them 10. Low pair (tens or lower) — hold the pair 11. Four cards to an open-ended straight — hold them 12. Two suited high cards, then any single high card — hold them If none applies, discard everything. The two counter-intuitive lines are the ones that cost casual players the most. Hold a low pair over three cards to a straight or a flush. And break a made flush or straight to chase four cards to a royal — the royal pays enough to justify it. Both feel wrong and both are correct.

The variants change the strategy completely. Deuces Wild, Bonus Poker, Double Double Bonus and the rest have different paytables and different optimal play. A full-pay Deuces Wild machine can theoretically return over 100%, but those paytables are rare and the strategy is considerably harder. Do not carry Jacks or Better strategy into another variant — it will cost you more than playing badly at a game you understand. On bonuses: video poker is very often excluded from wagering requirements or weighted at zero, for the obvious reason that a 99.5% game is nearly break-even. Check before accepting an offer if this is what you intend to play. Discipline notes: • Play slowly. The edge is small but it applies to every hand, and video poker can be played very fast • Do not deviate on hunches. The whole value of the game is that the correct play is known. Improvising gives that away • Keep the strategy card open while you play. There is no shame in it, and it costs nothing • Expect long stretches without a royal flush. It appears roughly once in forty thousand hands, and a meaningful share of the game's return is concentrated in it. Sessions will feel worse than the return figure suggests