Editorial illustration for the lesson on basic strategy: the chart that beats casual play, in the Mayfair Casino School.
Editorial illustration for the lesson on basic strategy: the chart that beats casual play, in the Mayfair Casino School.

Basic strategy: the chart that beats casual play

The 1956 Baldwin paper gave the world the first mathematically correct blackjack strategy. Every chart in circulation since is a refinement of that work.

AC
Annabel Cavendish
Editor in Chief · Reviewed 14 May 2026
Annabel
0:000:00

Welcome to the lesson on basic strategy.

I'm Annabel, and this is the lesson where casual play ends and informed play begins.

Basic strategy is the set of mathematically optimal decisions for every blackjack hand, given only the cards in front of you and the dealer's up-card.

It doesn't require counting.

It doesn't require memory of previous hands.

It requires learning one lookup table until it's automatic, and then playing that table without deviation.

The origin of the chart matters, because it tells you what you're trusting.

In nineteen fifty-six, four researchers, Roger Baldwin, Wilbert Cantey, Herbert Maisel, and James McDermott, published a paper in the Journal of the American Statistical Association presenting the first mathematically rigorous analysis of blackjack strategy.

They spent two years working through the calculations on mechanical desk calculators.

Their published strategy reduced the house edge on the single-deck game to near zero, which was sufficiently alarming to casino management that the paper received limited attention at the time.

Every version of the basic strategy chart you'll encounter today is a descendant of that nineteen fifty-six computation.

Hard totals first.

A hard hand either contains no ace or contains an ace counted as one because counting it as eleven would bust you.

Hard eight and below: always hit.

Hard nine: double if the dealer shows three through six, otherwise hit.

Hard ten: double unless the dealer shows a ten or ace.

Hard eleven: double against almost everything except an ace in a six-deck game.

Hard twelve through sixteen: these are the difficult hands.

Stand against dealer two through six, hit against seven through ace.

The logic isn't that you have a good hand; it's that the dealer is likely to bust when showing a small card.

Soft totals next.

A soft hand contains an ace counted as eleven.

Because you can't bust on one more card, you play soft hands more aggressively.

Soft thirteen and fourteen: double against dealer five and six, otherwise hit.

Soft fifteen and sixteen: double against four through six, otherwise hit.

Soft seventeen: double against three through six, otherwise hit.

Soft eighteen is the one that catches most players out.

Double against three through six, stand against two, seven, and eight, hit against nine, ten, and ace.

It feels wrong to hit on eighteen.

The numbers say it's correct in those spots.

Pairs.

Always split aces and eights.

Never split fives or tens.

A pair of fives is a hard ten, one of the best doubling hands in the game; splitting it creates two weak hands instead of one strong one.

Twos and threes: split against dealer two through seven.

Sixes: split against dealer two through six.

Sevens: split against two through seven.

Nines: split against two through six and eight through nine, stand against seven, ten, and ace.

Surrender, where it's available.

In an H17 game, surrender hard seventeen against dealer ace.

These are the situations where your expected loss from playing the hand is worse than simply losing half your stake immediately.

Here is the part most players miss: the house edge for basic strategy depends on which game you're playing.

For a representative UK six-deck S17 game with three-to-two blackjack and double after split, the figure is approximately zero point five percent.

That number is rule-set dependent.

Add late surrender and it drops to around zero point four three percent.

Find a six-to-five payout and it climbs above one point eight percent.

The chart doesn't change between those two games.

The payout structure does all the damage.

That's the price of the entertainment at a well-structured table.

Compare it to the same calculation at a six-to-five game with an H17 rule: the edge climbs to around one point eight percent, and your expected hourly cost becomes twenty-two pounds fifty.

Same playing decisions.

Different rules.

Very different cost.

The only reliable path to making the chart automatic is repetition under pressure.

In practice, five or six rules cover most hands.

At Aspers Westfield Stratford, or the Hippodrome on Leicester Square, or any serious London floor, a player who runs correct basic strategy consistently is paying roughly half a percent for their entertainment.

That's the baseline worth establishing before you consider anything more advanced.

Learn the chart.

Make it automatic.

Everything else builds on top of it.

Start with the arithmetic, because the chart only makes sense once you understand what it optimises.

Basic strategy is the set of playing decisions that maximises expected value on every hand, given only the two cards in front of you and the dealer's up-card. It doesn't require counting. It doesn't require memory of previous hands. It requires knowing one lookup: what is the optimal action for this hand composition against this dealer card, under these specific rules? The 1956 paper by Roger Baldwin, Wilbert Cantey, Herbert Maisel, and James McDermott, published in the Journal of the American Statistical Association, was the first rigorous answer to that question. The four researchers spent two years working through the calculations with mechanical desk calculators. Their published strategy reduced the house edge on the single-deck game to near zero, which was sufficiently alarming to casino management that their paper was not widely publicised at the time.

Edward Thorp's 1962 book Beat the Dealer popularised and refined the strategy, adding the card counting systems that the Baldwin group had not included. Every version of the basic strategy chart you'll find today, whether on a laminated card sold in a Las Vegas gift shop or in our strategy calculator, is a descendant of Baldwin's original computation.

The Four Decision Families

The chart divides every hand you can hold into four families, and each family has its own logic.

Hard totals. A hard hand is any hand that doesn't contain an ace counted as 11, or any hand where counting the ace as 11 would bust you. Hard 8 and below: hit. Hard 9: double if the dealer shows 3-6, otherwise hit. Hard 10: double unless the dealer shows a 10 or ace. Hard 11: double against almost everything except an ace (where you hit in some six-deck games). Hard 12: stand against dealer 4-6, otherwise hit (yes, you stand on 12 against a 4: the dealer busts frequently enough to make standing correct). Hard 13-16: stand against dealer 2-6, hit against 7 through ace. Hard 17 and above: always stand. The logic on 12-16 is not that you have a good hand; it's that the dealer is likely to bust when showing a small card, so you survive by not busting yourself.

Soft totals. A soft hand contains an ace counted as 11: soft 13 is ace-2, soft 17 is ace-6, soft 18 is ace-7. Because you can't bust on one more card (the worst that happens is the ace reverts to 1), you play soft hands more aggressively. Soft 13-14: double against dealer 5-6, otherwise hit. Soft 15-16: double against 4-6, otherwise hit. Soft 17: double against 3-6, otherwise hit. Soft 18: double against 3-6, stand against 2, 7, 8, and hit against 9, 10, ace. Soft 19-20: always stand. The soft 18 rule catches most players out; it feels wrong to hit or double on 18, but the numbers say it's correct in those spots.

Pairs. Always split aces and eights. Never split fives or tens. Split twos and threes against dealer 2-7. Split sixes against 2-6. Split sevens against 2-7. Split nines against 2-6 and 8-9 (stand against 7, because 18 beats a likely dealer 17; stand against 10 and ace because you're in trouble either way). The never-split-fives rule is because two fives form hard 10, one of the best doubling hands in the game; splitting them turns a strong position into two weak ones.

Surrender. Where late surrender is available, the rule is cleaner than most players expect. Surrender hard 16 (except 8-8, which you split) against dealer 9, 10, or ace. Surrender hard 15 against dealer 10. Surrender hard 17 against dealer ace in an H17 game. These are the situations where your expected loss from continuing is worse than losing half your stake immediately. correct surrender usage is worth at roughly 0.07% off the house edge.

Where the 0.42% Edge Figure Comes From

The commonly cited house edge for basic strategy blackjack varies slightly across sources because the answer is rule-set dependent. Our edge calculator lets you input any combination of rule variations and returns the precise house edge for basic strategy play.

For a representative UK six-deck game with S17, double on any two cards, DAS, no late surrender, and 3:2 blackjack, the figure is approximately 0.50%. Add late surrender and it drops to around 0.43%. The 0.42% figure you'll see in some textbooks (and in the lesson title) is typically derived from a very slightly more favourable rule set, or from a slightly different computation methodology. The range 0.40-0.50% covers most UK shoe games with decent rules; anything above 0.60% suggests an unfavourable rule variation you should identify before playing.

At £25 per hand and 50 hands per hour, the expected cost of correct basic strategy play is: £25 x 50 x 0.0050 = £6.25 per hour. That is the price of the entertainment, not a disaster. Compare it to the same calculation with a 1.80% edge on a 6:5 game: £25 x 50 x 0.0180 = £22.50 per hour. The chart doesn't change between those two games. The payout rule does all the work.

Memorising the Chart in Practice

There are roughly 270 distinct hand-dealer-card combinations in the standard six-deck chart. That sounds like a lot. In practice, most hands are resolved by five or six rules that cover the vast majority of situations you'll face. Hard totals 8 and below hit; 17 and above stand; the middle range follows the dealer's up-card. Pairs of aces and eights always split; pairs of fives and tens never do. Soft totals below 18 are mostly hit or double; 19 and above always stand.

The cases that trip players up are the ones where intuition disagrees with the maths: hitting soft 18 against a 9 or 10 (you're an underdog either way, but hitting gives a better average outcome), standing on 12 against a dealer 4, 5, or 6 (the dealer busts roughly 40% of the time with these up-cards, and you don't need to improve to win), and splitting nines against an 8 or 9 (18 is worse than it looks when the dealer shows a card that leads to 18 or 19).

The only reliable way to make the chart automatic is repetition under conditions that simulate the pressure of a real table. The blackjack trainer at tools/blackjack-trainer/ does exactly that: it deals hands, clocks your decision speed, and flags deviations from correct basic strategy. Annabel's strong recommendation is to reach a reliable 95% correct rate before sitting at Aspers Westfield Stratford or the Hippodrome with real chips. Below that threshold, you're paying for the educational experience at casino prices, which is a costly way to learn.

Key numbers

Decision typeKey rule (six-deck S17)Most common mistake
Hard 12Stand vs 4-6; hit vs 2, 3, 7+Standing vs 2 and 3 (should hit)
Hard 16Stand vs 2-6; hit vs 7+; surrender vs 9, 10, AHitting vs 6 (should stand)
Soft 18Double vs 3-6; stand vs 2, 7, 8; hit vs 9, 10, AStanding vs 9 or 10 (should hit)
Pair of 9sSplit vs 2-6 and 8-9; stand vs 7, 10, ASplitting vs 7 or 10
Pair of 8sAlways splitStanding on 16 instead of splitting
Hard 11Double vs 2-10; hit vs ace (S17 game)Not doubling vs 10
InsuranceNever (without count)Taking insurance on blackjack

Sources: Baldwin et al. 1956, JASA, our calculation, our calculation, our calculation.

Resume
Next lesson