Jacks or Better strategy means choosing which cards to hold so that the average payout from the draw is as high as possible. The correct decision depends on the full five-card deal and the machine’s paytable. A low pair can be worth keeping even though it does not pay immediately; a four-card royal draw can justify breaking an otherwise complete hand.
This guide focuses on conventional 9/6 Jacks or Better with a 52-card deck, no wild cards, and an 800-for-1 royal-flush return at the qualifying wager. It includes a practical strategy hierarchy, worked decisions, paytable comparisons, and the limitations behind quoted return percentages. It does not apply unchanged to Deuces Wild, Bonus Poker, or Double Double Bonus.
How Jacks or Better works
- Select the denomination and number of credits to wager.
- Receive five cards from the deck.
- Choose any cards to hold, from none to all five.
- Draw replacements for the discarded cards.
- Receive the displayed payout if the final hand qualifies.
The lowest-paying result is a pair of jacks, queens, kings, or aces. A pair of tens does not pay, nor does an ace-high hand without a qualifying combination. You are paid against a fixed schedule rather than competing with a dealer’s hand.
In the conventional game, discarded cards do not return as replacements during that hand. If you discard a king, there are only three kings left among the 47 unseen cards. This information matters to exact expected-value calculations.
The deal/draw button can move play forward immediately. Confirm every HOLD indicator before pressing it. Accidentally releasing one card can change a sound decision into a completely different wager.
Read the paytable before using a strategy chart
The name 9/6 refers to the full-house and flush payouts: nine and six credits respectively for each credit wagered. It does not mean nine decks, six cards, or a 96% return. You still need to inspect every other row.
| Final hand | Standard 9/6 return per credit at the royal-qualifying wager | Five-credit return |
|---|---|---|
| Royal flush | 800 | 4,000 |
| Straight flush | 50 | 250 |
| Four of a kind | 25 | 125 |
| Full house | 9 | 45 |
| Flush | 6 | 30 |
| Straight | 4 | 20 |
| Three of a kind | 3 | 15 |
| Two pair | 2 | 10 |
| Pair of jacks or better | 1 | 5 |
| Other hand | 0 | 0 |
These are total credits returned, not profit odds. A five-credit bet finishing with a pair of queens returns five credits and breaks even on that hand. A full house returns 45 credits, leaving 40 credits of profit after the original five-credit stake.
Some traditional machines pay only 250 credits per coin for a royal at one through four coins, then award 4,000 for five. Other implementations use proportional payouts at every available stake. Read your machine instead of assuming that a max-bet button is always necessary.
Why 9/6, 8/5, and 7/5 are different games mathematically
Reducing a full house or flush by one credit sounds small, but these outcomes recur far more often than a royal. With the conventional remaining payouts and an 800-for-1 royal, published optimal theoretical returns are approximately:
| Paytable | Optimal theoretical return | Corresponding house edge |
|---|---|---|
| 9/6 | 99.54% | 0.46% |
| 9/5 | 98.45% | 1.55% |
| 8/5 | 97.30% | 2.70% |
| 7/5 | 96.15% | 3.85% |
| 6/5 | 95.00% | 5.00% |
The figures come from the Wizard of Odds paytable calculations and assume the correct optimal strategy for each schedule. A game merely showing nine and six in two rows does not qualify if other payouts differ.
For perspective, a 0.46% edge corresponds to about $4.60 expected loss per $1,000 of turnover; 2.70% corresponds to about $27. Neither is a prediction of your next session. A rare large payout contributes to the mathematical average and can be absent from a long personal sequence.
Terms used in a Jacks or Better strategy chart
High cards are jacks, queens, kings, and aces. A ten can help form a royal or straight, but a pair of tens does not qualify for the minimum payout.
A royal draw contains suited cards drawn from ten, jack, queen, king, and ace. For example, ten-queen-king of hearts is three to a royal. Ten-jack-queen of mixed suits is not.
An outside straight draw has four distinct consecutive ranks that can finish at either end, such as six-seven-eight-nine. A five or ten completes it. An inside draw has only one missing rank, such as six-seven-nine-ten needing eight. Ace-low and ace-high end sequences also have only one completing rank.
A penalty card is a card outside your intended hold whose removal reduces certain future possibilities. Exact strategy sometimes changes because that card is no longer available to draw. A short beginner chart cannot describe every such exception.
A practical 9/6 Jacks or Better strategy hierarchy
Evaluate the available holds from the top down and use the first applicable row. This compact hierarchy follows Michael Shackleford’s simplified 9/6 strategy, with an approximate 99.46% theoretical return under its stated paytable. It is deliberately simpler than fully optimal play.
| Priority | Cards to preserve |
|---|---|
| 1 | A completed royal, straight flush, or four matching ranks |
| 2 | Four suited royal cards |
| 3 | A full house, flush, straight, or three matching ranks |
| 4 | Four cards toward a straight flush |
| 5 | Both pairs |
| 6 | A paying pair |
| 7 | Three suited royal cards |
| 8 | Four cards sharing one suit |
| 9 | A nonpaying pair |
| 10 | Four cards in an outside straight draw |
| 11 | Two high cards sharing a suit |
| 12 | Three cards toward a straight flush |
| 13 | Two high cards of different suits; use the lowest two if more are present |
| 14 | A suited ten with a jack, queen, or king |
| 15 | One high card |
| 16 | Nothing: replace all five cards |
A straight-flush draw must be capable of fitting inside a five-card straight of one suit. Three clubs such as two, eight, and queen do not qualify simply because they share a suit. Aces can be low or high, but sequences do not wrap through king-ace-two.
For four of a kind, hold the four matching cards; the fifth cannot improve the payout in standard Jacks or Better. Holding it as well produces the same award, but this equivalence should not be assumed in variants that pay for specific kickers.
Worked examples: what to hold and why
Example 1: keep a low pair ahead of unrelated high cards
Deal: 6♣, 6♦, A♠, Q♥, 3♣. Hold the two sixes. They do not pay yet, but drawing three cards gives opportunities to make two pair, three of a kind, a full house, or four of a kind.
The ace and queen can tempt a beginner because they form paying pairs. However, keeping those disconnected high cards throws away the existing pair’s improvement paths. Strategy compares all possible payouts, not just the chance of a single high pair.
Example 2: keep a high pair ahead of four to a flush
Deal: J♥, J♣, 8♥, 5♥, 2♥. Hold both jacks under the simplified hierarchy. The guaranteed break-even pair also retains improvement possibilities. Breaking it to chase four hearts sacrifices that existing value.
This example shows why “always hold four of one suit” is incomplete. The competing hold matters. The same flush draw can be preferred when it competes with a low pair instead.
Example 3: four to a flush ahead of a low pair
Deal: 7♥, 7♣, K♥, 9♥, 2♥. Hold the four hearts and discard 7♣. A replacement heart completes the flush, and some non-heart draws can still produce a paying pair with the king.
There are nine unseen hearts among the 47 possible replacement cards because four hearts are already in your hand. The flush completion probability is therefore 9/47, about 19.15%. That calculation alone does not establish which hold is best; it is one component of the expected-value comparison.
Example 4: do not break two pair to chase four of a kind
Deal: K♣, K♦, 4♥, 4♠, 9♣. Keep both pairs. Drawing one card guarantees at least the two-pair payout and gives four possible cards that complete a full house.
Keeping only the kings creates a chance of four kings but discards a guaranteed two-credit return per credit wagered. A larger possible prize does not automatically make a draw more valuable.
Example 5: a four-card royal can outrank a made straight
Deal: 9♣, 10♥, J♥, Q♥, K♥. Hold the four hearts and discard the nine. One unseen ace of hearts produces the royal, while other cards can make a flush, straight, or paying pair.
This decision relies on the qualifying royal payout. The royal possibility alone contributes 800/47, about 17.02 returned credits per credit bet, before adding the other winning outcomes. That exceeds the four-credit return for keeping the straight. Do not generalize this logic to a three-card royal draw, where two exact replacement cards are needed.
Example 6: sometimes discard everything
Deal: 2♣, 5♦, 8♥, 10♠, 3♦. There is no pair, high card, four-card straight draw, or suitable suited combination in the hierarchy. Replace all five cards.
Keeping a ten because it feels close to a jack is not justified. It cannot make the minimum paying pair, and this particular deal offers no qualifying suited ten combination.
Expected value versus the chance of winning something
Expected value adds each possible return multiplied by its probability. Suppose a hypothetical choice returns one credit 60% of the time and zero otherwise. Its average return is 0.60 credits. Another hypothetical choice returns three credits 25% of the time and zero otherwise, averaging 0.75 credits.
The first choice pays more frequently; the second has higher expected value. Neither hypothetical option is profitable against a one-credit stake. This distinction explains why an optimal video poker hold need not maximize hit frequency or guarantee that a session feels smooth.
For a real hand, a solver compares up to 32 possible subsets of five cards. For each subset it enumerates the legal draws, evaluates final hands, and applies the exact paytable. The best subset is the one with the largest weighted return.
This is also why a strategy error cannot be identified solely from the result. A correct hold can finish empty. An incorrect hold can make a royal. Judge the decision using information available before the draw, not the cards that happened to appear afterward.
Simple strategy versus optimal strategy
A short chart groups together hands that are nearly, but not always exactly, equivalent. Fully optimal 9/6 play uses more detailed distinctions involving ranks, gaps, suited combinations, and discarded-card effects.
For example, two deals can contain the same suited jack and queen but differ in the three cards that would be discarded. Those discarded ranks and suits affect which replacement combinations remain. A solver can identify a narrow exception that a broad “hold two suited high cards” rule misses.
The optimal strategy reference explains those exceptions. Treat the short hierarchy here as a learning tool, not as a claim of error-free 99.54% play. The optimal theoretical percentage also assumes the qualifying royal payout and all the listed standard awards.
Credit size, maximum coins, and total exposure
Denomination is the cash value of one credit. A five-credit hand at a $0.05 denomination costs $0.25. The same five-credit setting at $1 costs $5. Changing denomination can therefore matter far more to your budget than the number of credits shown.
If five credits are required for the enhanced royal, choosing a lower denomination can preserve that payout structure at a smaller total stake. That is a conditional comparison, not a reason to exceed your budget. A game with unfavorable affordable options can simply be skipped.
Multi-hand play multiplies the wager again. Ten hands at five credits of $0.25 cost $12.50 per deal. The hands commonly share the initial cards and your hold selection, then draw replacements separately. Their outcomes are therefore not equivalent to ten fully independent initial deals.
Optional multipliers and extra-credit features can alter both the stake and the game’s mathematics. A strategy or return figure for the plain base game should not be carried across without checking the feature-specific rules.
How to practice without learning the wrong lesson
- Choose a trainer that identifies the exact paytable.
- Confirm whether it teaches a simple chart or optimal play.
- Decide your hold before requesting feedback.
- Record the reason for an error, such as overlooking a four-card flush.
- Review repeated mistake categories instead of celebrating practice winnings.
- Recheck the paytable if you change games or denominations.
A useful practice session asks whether you recognized the correct competing holds. A less useful one asks whether the balance happened to rise. Focus on decision accuracy and expected-value loss from mistakes, because a random positive result can reinforce a bad habit.
Auto-hold features are not automatically perfect. Their design and settings may differ, and a highlighted hold should still be checked against the applicable strategy. Likewise, a feature labeled double or gamble creates another wager after a payout; it is not part of the base-game return calculation unless specifically included.
Frequently asked questions
Should you always keep a pair in Jacks or Better?
No. A four-card royal outranks a paying pair in the stated strategy, while several draws outrank a low pair. Identify all competing holds and apply the hierarchy rather than using one unconditional rule.
Should you keep an ace kicker with a pair?
Usually not in standard Jacks or Better. With an ordinary pair hold, retaining an unrelated ace reduces the number of replacement cards and changes improvement chances. Kicker rules in some bonus-poker variants are a separate matter.
Does 99.54% RTP mean I will recover 99.54% of my deposit?
No. It is an expected return on total qualifying wagers under optimal play, not a promise about a deposit or withdrawal. Repeatedly betting returned credits creates more turnover. Our RTP explanation develops the distinction between theoretical return and a personal result.
Is a royal flush due after a long drought?
No. Under the conventional independently shuffled-hand model, previous failures do not make the next initial deal more favorable. Average frequency is not a countdown, and playing longer to force a royal adds more wagering exposure.
Can skill remove the house edge?
Correct decisions reduce avoidable losses, but the conventional 9/6 schedule described here still has a house advantage even with optimal play. Promotions or unusual schedules require separate analysis and should not be assumed to make a game profitable.
