Blackjack Basic Strategy & House Edge
Basic strategy computed from the dealer's actual outcome odds, not copied from another chart. Adjust the rules and every cell is worked out again.
Change a rule and every cell is recomputed. The chart is not stored anywhere — it is worked out from the dealer's outcome distribution each time you touch a control.
What each rule is worth
Each line is the difference that one change makes to the house edge, everything else held as you have it. Priced against your table, not a standard one, so turning a rule on changes what the others are worth.
One hand at a time
—
Hard totals
Soft totals
Pairs
How often the dealer busts
Conditioned on the dealer not having blackjack, because if they did the hand would already be over. That is why an ace looks so much safer here than folklore suggests — the ten that would have busted it has been ruled out.
What this chart is and is not
- This chart is computed, not copied. The dealer outcome distribution is worked out by recursion and every cell is whichever action has the highest expected value, so the familiar rules — always split aces and eights, never split tens, stand on hard 17 — are results here rather than instructions.
- The dealer is assumed to peek for blackjack, which changes everything about an ace or a ten showing. Once you are playing, the hole card is known not to make 21, so the dealer bust figures below are conditioned on that. It is why an ace up is far worse for you than its raw bust percentage suggests.
- A dealer showing six busts about 42% of the time — not most of the time. The "bust card" is a hand you are still expected to lose more than half the time you play it out badly, which is why the chart tells you to stand on stiff totals against it rather than chase.
- Basic strategy does not make the game beatable. Played perfectly it loses slowly, and every deviation loses faster. Nothing here counts cards, and nothing here changes what the next card will be.
- The 6:5 blackjack payout costs more than every rule variation on this page combined. It is worth about 1.4% of every bet, against a house edge that is otherwise well under one — it roughly triples the cost of playing.
- An infinite shoe is assumed, so cards already dealt do not change the odds. Against a real six-deck game that is worth roughly a tenth of a percent and moves two or three marginal cells; against a single deck it is meaningfully wrong.
How to use
- Set the rules your table actually uses.
- Read the chart, or look up a single hand below it.
- Check what each rule is costing you in the house edge.
Frequently asked questions
Where does this chart come from?
It is computed rather than transcribed. The tool works out how the dealer’s hand finishes from every upcard by recursion, then the expected value of standing, hitting, doubling and splitting for each hand, and prints whichever is highest. The familiar rules — always split aces and eights, never split tens, stand on hard 17 — come out of that arithmetic rather than being typed in.
Should I hit 16 against a ten?
Yes, narrowly, and the numbers show why it feels wrong: both options lose. Standing wins only when the dealer busts, which they do barely a fifth of the time from a ten. Hitting loses slightly less often, so it is the better of two bad choices — and if the table offers late surrender, giving up half the bet is better than either.
How often does the dealer really bust?
Less than people think. A six — the famous bust card — busts about 42% of the time, so the dealer still makes a hand more often than not. A ten busts around 21% and an ace less still, because once the dealer has peeked for blackjack the ten that would have busted the ace has been ruled out.
What does the 6:5 payout actually cost?
About 1.4% of every bet, which is more than every other rule variation on this page put together. That figure is exact arithmetic: you make a natural roughly one hand in twenty-one, and each one now pays 0.3 of a bet less. Against a house edge otherwise well under one per cent, it roughly triples the cost of playing.
Why does hitting soft 17 matter so much?
Because it only ever helps the dealer. A dealer who must hit soft 17 busts more often, but also turns more seventeens into eighteens, nineteens and twenties — and the hands they improve on were the ones you were beating. It costs the player about 0.22%, which this tool measures rather than quotes.
Does basic strategy make blackjack beatable?
No. Played perfectly it loses slowly, and every deviation loses faster. Nothing here counts cards or predicts the next card; it only ensures you are not giving away more than the rules already take. Treat the house edge shown as the price of the entertainment, not a hurdle you can clear.
How accurate is the infinite-deck assumption?
Close, and stated openly rather than hidden. Assuming cards already dealt do not change the odds is worth roughly a tenth of a percent against a real six-deck shoe and moves two or three marginal cells. Against a single deck it is meaningfully wrong, so use a single-deck chart if that is what you are playing.
🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.