Video poker is a single-player draw poker game against a paytable rather than
against other players. There is no bluffing, no reading opponents and no dealer to beat. You are
paid for making a qualifying hand, full stop.

The game descends from five-card draw. You are dealt five cards, you choose which to keep, the
discards are replaced once, and the final hand is compared against a fixed schedule of payouts. That
schedule is the paytable, and it is displayed on screen before you commit a coin.

Why it is not a slot

What separates it from slots is determinism you can act on. A slot’s return is
locked in software you cannot inspect. In video poker the deck is a standard 52 cards, the payouts
are published, and the mathematically best hold for any dealt hand can be calculated exactly. Two
players on identical machines will see different returns purely because one plays better holds.

It is also the reason video poker attracts the most disciplined players in the casino. The best
paytables sit close to break-even, and a few historic variants crossed it.

A round of video poker takes four steps and about five seconds. The complexity is entirely in
step three.

1. Choose your denomination and coin count. Denomination sets the value of one
   coin. Always play the maximum five coins, for reasons covered below.
2. Deal. Five cards appear face up from a freshly shuffled 52-card deck.
3. Hold. Select which cards to keep. Everything not held is discarded. This is the
   only decision in the game and it is where your return is won or lost.
4. Draw. Discards are replaced from the remaining 47 cards. The final hand is paid
   against the paytable, or it is not.

There is no second draw and no way to improve a finished hand. Every round is independent: the
deck is reshuffled, and nothing carries over. Card counting has no meaning here.

Turn autohold off

One practical note. Most machines let you hold cards by clicking them directly or by pressing the
number keys, and most have an autohold feature that pre-selects a hold for you. Autohold is usually
correct for obvious hands and reliably wrong on the marginal ones, which are exactly the hands where
your edge lives. Turn it off once you know the priority order.

Video poker uses standard poker hand rankings, but the money is distributed very differently from
table poker. A royal flush is the jackpot; a pair of jacks is the floor. In Jacks or Better, anything
below a pair of jacks pays nothing at all.

How often each hand actually lands, playing correct strategy on 9/6 Jacks or Better:

Reading the frequency table

Read that table next to the paytable and the structure becomes obvious. Roughly one hand in five
merely returns your stake. About one in eight is a genuine small win. The variance comes almost
entirely from four of a kind and above, which is why a session can drift downward for an hour and
then recover in a single hand.

The hold decision is a comparison of expected values, not a hunch. Every possible hold from a
dealt hand has a computable expected return, and correct strategy simply means choosing the highest
one every time. You do not need to do the arithmetic at the table. You need to know the order.

Two machines can run the same game, look identical, use the same graphics, and return returns
two full percentage points apart. The difference is printed on the paytable, and it is the single
most valuable thing to check before you play.

Here is full-pay 9/6 Jacks or Better, the benchmark. The name comes from the
full house paying 9 and the flush paying 6 per coin:

Now change two rows. Drop the full house to 8 and the flush to 5, and you have
8/5 Jacks or Better. Everything else is unchanged. The return falls from 99.54% to
roughly 97.30%. Those are the two rows casinos quietly shave, because they are the
rows players do not check and they hit often enough to matter.

How to check a paytable in ten seconds

Look at the full house and flush rows on the one-coin column. In Jacks or Better you want 9 and
6. If you see 8 and 5, or 7 and 5, the machine is worse and you should find another one. The rest of
the schedule is usually left alone, so those two numbers tell you almost everything.

The same principle applies to every variant, only the diagnostic rows change. Bonus Poker turns
on the full house and flush too. Double Bonus keys on the flush and the four-of-a-kind tiers. Deuces
Wild lives or dies on the four of a kind, straight flush and five of a kind rows. Learn the two or
three rows that matter for the variant you play and you can grade any machine at a glance.

Look again at the royal flush row. One to four coins pays a flat 250 per coin: 250, 500, 750,
1,000. Then the fifth coin pays 4,000, not 1,250. That single step is worth about
1.4% of total return, and it only exists on the maximum bet.

Play four coins on a 99.54% machine and you are playing a 98.1% machine. You have thrown away
more than the entire difference between a good paytable and a bad one, and you have done it
voluntarily.

Drop the denomination, not the coin count

The correct response is not to bet more than you are comfortable with. It is to
drop the denomination and keep five coins. Five coins at $0.05 is $0.25 a hand and
gets the full royal flush bonus. One coin at $0.25 is the same $0.25 a hand and does not. Same
money, two percent apart in return.

This is the most common and most expensive mistake in the game, and it takes one setting to
fix.

Return to player is the long-run percentage a game pays back with correct strategy. House edge is
what is left. Volatility describes how violently the actual results swing around that average.

On 9/6 Jacks or Better the RTP is 99.54% and the house edge is
0.46%, better than almost anything else in the casino and comparable to good
blackjack. But that figure assumes two things people forget: the full five-coin bet, and correct
holds on every single hand. Play the same machine with casual holds and the real return lands
somewhere in the mid-98s. The published number is a ceiling, not a promise.

Volatility is the part that surprises people

Volatility is the part that surprises people. A 99.54% game still delivers long losing stretches,
because a meaningful slice of that return sits in hands you will hit once in 420 rounds or once in
40,000. Expect to be down over a short session more often than not. The royal flush contribution
alone is about 2% of total return, which means roughly two points of your expected payback arrive in
a hand most players never see.

Two implications. Bankroll for the variance, not the average. And do not read a losing hour as
evidence the machine is tight, or a lucky hour as evidence you have found a loose one.

Every variant is a trade. Wild cards and bonus four-of-a-kind payouts make the game more exciting
and are paid for elsewhere in the schedule, usually by gutting the low-end hands. Here is what the
best available paytable returns for each of the main games, with correct play.

Two of those returns exceed 100%, which is not a typo. Full-pay Deuces Wild and 10/7 Double
Bonus were genuinely positive-expectation games, which is exactly why they have almost entirely
disappeared. If you find one online, verify every row of the paytable before believing it. The
overwhelming likelihood is a reduced schedule wearing the same name.

Which variant should you actually play

If you are learning, 9/6 Jacks or Better. The strategy is the smallest to learn,
the variance is the mildest, and the return is excellent. Every other variant is a modification of
this baseline, so the effort transfers.

If you want bigger hits and accept deeper swings, Double Double Bonus is the
popular choice, but understand you are accepting roughly half a point of return and a much rougher
ride for the chance at kicker payouts.

Deuces Wild demands a completely separate strategy. Four wild cards change every
hold decision, and Jacks or Better instincts will actively cost you money. Do not switch between
them casually.

Correct video poker strategy is a ranked list. For any dealt hand, you scan the list from the top
and play the first hold that applies. That is the entire method, and for 9/6 Jacks or Better it
captures nearly all of the available return.

Simplified hold priority for 9/6 Jacks or Better, highest first:

1. Made royal flush, straight flush or four of a kind. Hold all five.
2. Four cards to a royal flush. Break a made flush or straight to draw for it.
3. Made full house, flush or straight.
4. Three of a kind, or four cards to a straight flush.
5. Two pair.
6. High pair (jacks or better).
7. Three cards to a royal flush.
8. Four cards to a flush.
9. Low pair. Yes, ahead of drawing to high cards.
10. Four cards to an open-ended straight.
11. Two suited high cards.
12. Three cards to a straight flush.
13. Two unsuited high cards, lowest two if you hold three.
14. One high card.
15. Nothing. Discard all five.

The two counter-intuitive rules

A low pair beats three high cards. It feels wrong to keep a pair of fours over
three face cards, but the pair can become three of a kind, a full house or quads, and the high cards
mostly become nothing.

Break a made hand for four to the royal. Discarding a paying flush to chase a
royal looks reckless and is mathematically correct, because the royal is worth 4,000 on a five-coin
bet. It will feel bad most of the time. Do it anyway.

Never keep a kicker

Holding a high card alongside a pair or three of a kind “just in case” is always wrong in Jacks
or Better. It removes a draw that could improve the hand and adds nothing. Keep exactly the cards
that form your hold and discard the rest.

Five errors account for nearly all of the gap between video poker’s advertised return and what
players actually get. None of them are subtle.

Betting fewer than five coins

Costs about 1.4% instantly by forfeiting the royal flush bonus. Drop the denomination instead.

Not checking the paytable

Playing 8/5 when 9/6 is available on the next machine costs 2.2%. Two rows, ten seconds.

Using one strategy for every variant

Deuces Wild and Jacks or Better require genuinely different holds. Applying the wrong chart can
cost several percent, which dwarfs every other decision at the machine.

Trusting autohold on marginal hands

Autohold handles obvious hands correctly and errs on the close ones, which are precisely the
hands where correct play earns its keep.

Chasing the royal

The royal flush is already priced into the 99.54%. Playing longer, higher or looser to “get it
back” does not change its frequency and does increase your exposure. There is no due machine.

Progressive video poker adds a rising jackpot, usually on the royal flush, funded by a small
deduction from the base paytable. That deduction means the game is worse than its non-progressive
equivalent for most of the jackpot’s life, and better only once the jackpot passes a break-even
point. For 9/6 Jacks or Better on a five-coin quarter machine, that point sits somewhere near the
low thousands, well above where most progressives sit day to day.

Multi-hand video poker deals one hold across three, five, ten or a hundred hands at once. The
mathematics per hand are unchanged. What changes is speed and exposure: a hundred-hand machine puts
a hundred times the money through the same 0.46% edge, so the house edge arrives a hundred times
faster in absolute terms. Volatility smooths out, the expected loss per hour does not.

Neither format is a trap, but neither is an edge either. Play them because you enjoy the format,
not because you think the maths improved.

Licensed online video poker uses a certified random number generator that shuffles a genuine
52-card deck for every hand. Independent test labs verify both the RNG’s randomness and that the
game pays according to its published schedule. The deck is not weighted against your held cards, and
there is no mechanism that watches what you hold and adjusts the draw.

What varies between operators is not fairness, it is the paytable. The same licensed game can be
offered at 9/6 or 8/5 at the operator’s choice, and both are entirely legitimate. This is why “is it
rigged” is almost always the wrong question. The machine is not cheating you; it is publishing worse
terms and relying on you not to read them.

The practical check is unchanged. Confirm the operator is licensed by a real regulator, then read
the full house and flush rows before you bet. A licensed 8/5 game will take your money entirely
honestly.

Against slots, video poker wins on transparency and return. Slot RTP is typically 94 to 96% and
is not displayed; a full-pay video poker machine is 99.54% and prints its schedule on the glass.
Slots require no decisions, which is their appeal and their cost.

Against blackjack, the two are close. Good blackjack with correct basic strategy runs around
99.5%, essentially matching 9/6 Jacks or Better. Blackjack is slower, more social and more
demanding at the table; video poker is solitary, fast and lets you take as long as you like over a
decision without anyone sighing at you.

Against table poker, they share only the hand rankings. Poker is a game against other players
where skill compounds without limit and a strong player can be profitable indefinitely. Video poker
is a game against a fixed schedule where perfect play has a ceiling, and on almost every available
paytable that ceiling is still below 100%.

• Paytable The schedule of payouts per hand per coin. The product itself.
• Full-pay The best published paytable for a given variant, e.g. 9/6 Jacks or
  Better.
• Short-pay Any reduced schedule, e.g. 8/5. Same game, worse terms.
• 9/6 Shorthand for the full house and flush payouts per coin.
• Hold The cards you keep before the draw. The only decision in the game.
• Kicker An extra card kept alongside a made hand. Always wrong in Jacks or
  Better.
• Max coins The five-coin bet that unlocks the royal flush bonus.
• Wild card A card substituting for any other, as in Deuces Wild or Joker
  Poker.
• RTP Return to player, the long-run payback with correct strategy.
• Volatility How widely real results swing around that average.

Video poker is worth playing if you are willing to do two things: find a good paytable and learn
a hold order. Do both and you are playing at a house edge under half a percent, with your outcome
genuinely influenced by your decisions. Do neither and you are playing an expensive slot machine
that happens to use cards.

What it is not

Be clear about what it is not. It is not a living. On the paytables actually available online the
ceiling is below break-even, so the expected result over a long enough horizon is a slow loss. The
handful of positive-expectation schedules are historical curiosities, and where they survive they
come with variance that requires a bankroll most players do not have.

What it offers instead is the best combination of low edge, published terms and real decisions on
the casino floor. For a player who enjoys getting something exactly right, that is a good deal.