Blackjack Basic Strategy Chart

The complete basic strategy chart for 4–8 deck blackjack, dealer stands on soft 17. Hard totals, soft totals and pairs, with the reasoning behind the plays people get wrong.

Updated September 2026 · how these numbers are calculated

This is the complete chart for the most common shoe game in North America: 4 to 8 decks, dealer stands on soft 17, doubling allowed on any two cards, double after split allowed, blackjack pays 3:2. Played perfectly against those rules, the house edge is about 0.43%.

If your table differs — particularly if the dealer hits soft 17 — see house edge by rule for what changes.

Hard totals

A hard total is any hand without an ace counted as 11.

The block of stands from 12 through 16 against a dealer 2–6 is the part new players resist most, because you are standing on a hand that loses to almost everything. You are not standing because 15 is good. You are standing because the dealer showing a 5 busts about 42% of the time, and letting them do it beats drawing to a hand that busts more often than it improves.

Soft totals

A soft total contains an ace counted as 11, so the hand cannot bust on the next card.

A,7 is the hand that costs people the most. A soft 18 feels finished. It is not. Against a dealer 9, 10 or ace, standing on soft 18 is a losing play — the dealer’s likely 19 or 20 beats it — and hitting is correct because you cannot bust. Against 3 through 6 you should be doubling, because the dealer is weak and you have a free draw.

Note that soft 18 has three different correct plays depending on the upcard: stand against 2, 7 and 8; double against 3–6; hit against 9, 10 and A. There is no simple rule that covers it, which is exactly why it is worth memorising.

Pairs

Split decisions come before hard-total decisions: if you have a pair and the pair chart says split, split.

Four rows carry almost all the value here.

Always split aces. Two aces played as one hand is a hard 12 or a soft 12 — a terrible hand. Split, and you have two hands each starting with the best card in the deck. You get one card each and cannot resplit in most games, and it is still overwhelmingly correct.

Always split eights. Sixteen is the worst total in blackjack. Two hands starting with 8 are both bad, but they are both better than one hand of 16 — including against a dealer ten, where the split loses money but loses less money than standing or hitting.

Never split tens. Twenty is the second-best hand in the game and wins about 85% of the time. Breaking it into two hands that each average worse is giving away a made hand for the feeling of action.

Never split fives. A pair of fives is a hard 10, which is a doubling hand against everything except a ten or an ace. Splitting it turns one excellent starting hand into two mediocre ones.

The plays people get wrong

Ranked by how much they cost, per occurrence:

What the chart cannot do

Basic strategy assumes you know nothing about the cards still in the shoe. That assumption is what makes it “basic”. It produces the best possible play averaged over all shoe compositions, which means it is occasionally the wrong play for the specific shoe in front of you — but you have no way to know which occasions those are without counting.

The remaining 0.43% is the price of that ignorance. Card counting is the only method that recovers it, by tracking the ratio of high to low cards and betting more when the shoe favours the player. It works, it is not illegal, and casinos are permitted to bar players for it. It also requires a bankroll in the thousands, hundreds of hours of practice, and game conditions — deep penetration, hand-shuffled or single-deck games, generous table limits — that most casinos have specifically engineered away.

For everyone else, the chart above is the whole of the strategy, and it is worth exactly what it costs to learn: roughly 1.5% of everything you will ever wager at a blackjack table.