Blackjack Basic Strategy Chart

Checked against the textbook basic-strategy chart: 5,5 doubles, 8,8 splits, 9,9 stands against a 7, hard 12 hits against 2 and 3; and Wizard of Odds’ rule-variations table, every rule’s effect on the house edge within 0.05 points. How

The best play for every hand you can hold against every card the dealer can show. Pick the table’s rules, find your hand down the side and the dealer’s card along the top: the letter tells you whether to hit, stand, double, split or surrender. Switch to the odds to see what that play is worth.

Dealer Stands on Soft 17, DAS, pays 3:2

SStandHHitDDoublePSplit

Hard totals

Hand2345678910A
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

Soft totals

Hand2345678910A
13
14
15
16
17
18
19
20

Pairs

Hand2345678910A
A,A
2,2
3,3
4,4
5,5
6,6
7,7
8,8
9,9
10,10

Two cells differ from a printed chart. soft 15 against a 4 and soft 13 against a 5 read hit here and double on most printed six-deck charts. This grid is worked out against an infinite shoe, where hit wins by a hair; in a real six-deck shoe, with your cards and the dealer’s gone, the two plays swap. Either is defensible, and getting it “wrong” costs nothing measurable.

Reading the grid: rows are your hand (hard totals, soft totals with an ace, and pairs), columns are the dealer’s up-card. A hard 12–16 is a “stiff” hand — the chart is where it earns that name, since whether to hit or stand flips by dealer card. Pairs of 8s split against almost everything, even a dealer 10, because two hands starting at 8 beat one hand stuck at a hard 16. Soft hands (an ace counted as 11) double more aggressively than the equivalent hard total, since busting isn’t possible on the next card.

How the chart was worked out

Each cell is the play with the highest expected value, found by working through every way the hand can go rather than by simulating hands. Like virtually every published basic strategy chart, it assumes an infinite shoe, where every card is equally likely whatever has already been dealt.

Why there is no deck-count control: a basic-strategy chart is total-dependent and very nearly deck-agnostic — the same reason published charts do not ship one per shoe size. These cells are computed against an infinite shoe, which is the standard convention and what makes them comparable to any chart you already own. Deck count moves the house edge rather than the plays, and that is where this site varies it: rule impact quotes every edge against a real six-deck shoe dealt without replacement.

Never take insurance

When the dealer shows an ace, they’ll offer insurance — a side bet, up to half your wager, that pays 2:1 if their hole card is a ten. It sounds like protection; it’s actually a house edge of about 7.7% on its own, worse than almost anything else at the table. Only 4 of the 13 unseen card ranks are worth ten, so the true odds (4/13 ≈ 30.8%) are worse than the 1-in-3 the 2:1 payout would need to break even — the bet loses about 1/13 of a unit on average. That holds regardless of what your own hand is worth, including the tempting-sounding “even money” offer on your own blackjack, which is the identical bad bet under a friendlier name. The one exception is a player counting cards: once tens make up more than a third of what is left, insurance turns profitable, at a Hi-Lo true count of about +3, which the card counting page works through.

Or answer one hand at a time

Three taps instead of parsing the full grid.

What’s my play right now?

Tap your two cards and the dealer’s up-card for an instant answer.

Your first card
Your second card
Dealer shows
Split
8+8 vs dealer 10
Why? See the math →

Try next

Split EV uses the standard infinite-deck approximation (each resulting hand treated independently); resplitting further hands isn’t modeled, which can very slightly understate splitting a few borderline pairs (e.g. 8s against a strong dealer card).

By Gabe · Updated · How every figure is checked

Cookies for analytics and ads; decline and both stop. Privacy