Blackjack strategy and the basic strategy chart

What basic strategy actually is

Blackjack is the one casino card game whose correct play has been solved completely. Every hand is a closed maths problem: you know your own cards, you know the dealer's upcard, and you know the rules the dealer must follow. From those three things you can calculate which decision loses the least money over the long run. Somebody has already done that calculation for all of the combinations, and the answers are the chart below.

Two things follow from that, and both matter. The first is that blackjack basic strategy is not advice or a playing style; it is arithmetic, and a hunch that disagrees with it is simply wrong. The second is that it has a ceiling. Perfect play gets the house edge down to around 0.5% in a decent six-deck blackjack game, and there it stops, because the dealer's structural advantage never goes away.

The chart only holds for the rules it was built for. This one assumes six decks, the dealer stands on soft 17, doubling on any two cards, doubling after splitting allowed, the dealer peeks for blackjack, and a 3 to 2 payout.

If you are not sure what any of those terms mean, the rules page walks through every one of them, and it is worth five minutes before you memorise anything here.

Hard totals

A hard hand has no ace, or has an ace that must count as 1. Read across from your total to the dealer's upcard.

H hit. S stand. D double if the table lets you, otherwise hit.

Hitting 16 against a ten feels terrible and is still correct: you lose either way most of the time, and you lose slightly less by drawing. If your table offers late surrender, give up 16 against a 9, 10 or ace, and 15 against a 10. Under a dealer-hits-soft-17 game, 11 against an ace becomes a double rather than a hit.

Soft hands

A soft hand holds an ace counted as 11, so one more card can never bust it. This is where most money is thrown away, usually by standing on soft 17 or 18 because the number looks respectable.

Ds means double if you are allowed to, otherwise stand. Soft 18 is the hand to study: it is strong enough to stand on against a 7 or 8, worth doubling against the dealer's weakest cards, and genuinely bad against a 9, 10 or ace, where you should hit it. In a game where the dealer hits soft 17, soft 19 becomes a double against a 6 and soft 18 becomes a double against a 2.

Pairs

Splitting turns one hand into two and puts a second bet out, so it is only right when the two new hands are clearly better than the one you started with.

P split. P* split only if the table allows doubling after a split; otherwise hit. Note that 5,5 is never split. It is a hard 10 and should be doubled like one.

Two rules survive without the chart. Always split aces and eights: aces because two hands starting from 11 are worth far more than a single soft 12, and eights because 16 is the worst total in the game and two hands of 8 are an improvement even against a ten. Never split tens; you are breaking up a 20 to chase something better, and there is nothing better.

How to memorise it without a printout

Nobody starts by knowing 300 cells. Learn basic strategy in layers, because the top layer covers most of the hands you will ever be dealt.

• Dealer showing 2 to 6 is weak, so stand on 12 through 16 and let them draw. Those are the upcards the dealer busts with most often, and a hand you do not draw to cannot bust at all. The exception is 12 against a 2 or 3, which you hit.
• Dealer showing 7 to ace is strong, so hit 12 through 16 and take the risk. A pat 17 is not a winning hand against a ten; it is just a slower loss.
• Always double 11 against anything except an ace, and double 10 against anything except a 10 or an ace.
• Always split aces and eights, never split fives or tens.
• Never stand on a soft 17. There is no downside to taking one more card.

Then drill the rest a few hands at a time. The same grid is sold in wallet form as a blackjack cheat sheet, and a laminated cheat sheet is legal in most casinos as long as you do not slow the game down; online you can simply keep it open beside you. Using it is not cheating, it is just reading; the house edge already assumes you are imperfect.

The mistakes that cost the most

These come up over and over, and they are expensive in that order.

1. Taking insurance or even money. It is a separate losing bet, covered below.
2. Playing a 6:5 table. No blackjack strategy recovers what that payout takes away. Check the felt before you sit.
3. Standing on 16 against a ten because hitting feels reckless. Both options are bad; hitting is less bad.
4. Not splitting eights against a strong upcard. Two hands from 16 beat one hand of 16, even against a ten.
5. Standing on soft 18 against a 9 or 10. It looks like a good hand and loses more often than it wins.
6. Mimicking the dealer. Drawing to 16 and standing on 17 with no doubling and no splitting throws away the only advantages you have.
7. Side bets every hand. Perfect Pairs and 21+3 pay well when they hit and cost far more than the main game over a session.

Insurance, even money and the maths of saying no

Insurance is offered when the dealer shows an ace, costs half your bet, and pays 2 to 1 if the hole card is worth ten. To break even you need better than one in three of the remaining cards to be ten-value, and only four of the thirteen ranks are. In a six-deck shoe roughly 30.9% of the cards are ten-value, so the bet gives up about 7% of every ringgit wagered on it.

Even money is the same offer aimed at a player holding blackjack. Accept it and you take a certain 1 to 1 instead of a 3 to 2 that arrives most of the time. It feels safe, which is exactly why it is offered so warmly.

The one exception is a player who already knows the shoe is loaded with tens. That is not a feeling, it is a tracked number, and the method is set out on the card counting page. Without a count, decline every time.

Why betting systems do not work

Martingale asks you to double your bet after every loss so that one win recovers everything. The maths is honest about what happens next. Six losses in a row from RM10 means your next bet is RM640 to win back RM10, and that run happens far more often than most players expect. Table limits then stop you exactly when you need one more double, and the bankroll you need to survive is enormous next to the money you are trying to win.

Every other progression, Paroli, Fibonacci, D'Alembert, 1-3-2-6, has the same property. Changing the size of your bets changes the shape of your session: more small wins and rarer big losses, or the other way round. It cannot change the expected value, because the house edge applies to each bet separately and adding up biased bets never produces an unbiased total.

Behind most systems sits the gambler's fallacy, the belief that a run of bad hands makes a good one due. Cards have no memory of who won. There is one honest exception, and it is narrow: in a shoe game the cards already dealt genuinely change what is left, which is why counting works at all. That is information about composition, not about luck balancing out.

Choosing the table beats playing the hand

Your blackjack odds are set by the rules printed on the felt, before your first decision and whatever you do afterwards. Run this checklist while you are still standing up.

Continuous shuffling machines deserve a note of their own. They return played cards to the shoe after every round, so the deck composition never drifts. That barely changes correct play, but it removes any chance of tracking the shoe and it speeds the game up considerably, which means more hands per hour and more expected loss per hour at the same edge.

Online the checklist starts one level up, because you choose an operator before you choose a table and what is on offer follows from that. Our page on how to judge an online casino covers the checks worth making first. The casino pages we maintain are indexed in one place.

When to deviate from the chart

There is exactly one legitimate reason to make a play the chart does not list: you have information the chart does not assume. The chart is calculated for an average, freshly shuffled shoe. If you have been tracking which cards are gone, some borderline decisions flip, insurance being the clearest of them.

That is advantage play, it takes real work, and it is covered honestly, including its limits, on the card counting page. Everything else that gets called a deviation, playing differently after three losses, changing because the dealer looks lucky, standing because the table groaned, is superstition with a costume on.

Play the chart, pick your table well, and treat the difference between a good session and a bad one as variance rather than skill. Set a budget before you start, keep it to money you would be comfortable losing, and stop when it is gone. 18+ only.