Perfect Blackjack Strategy

"Perfect blackjack strategy" means two different things depending on who is asking. For most players it means the mathematically best decision for every hand, which is what basic blackjack strategy already gives you, plus a small set of composition-dependent exceptions worth a few hundredths of a per cent. For a smaller group it means Perfect Blackjack, the Playtech table game with Perfect Pairs and 21+3 side bets built in. This page covers both, in that order, and gives you the chart and the house edge figures for each.

One thing to be clear about first: no strategy, perfect or otherwise, guarantees a win. Perfect play gets the house edge down to around 0.3% to 0.5% depending on the rules. That edge is a long-run average, and the result of any one session will sit a long way from it.

What "perfect strategy" means

Basic strategy is the best play for your hand total against the dealer's up-card. Hard 16 against a 10 is one decision whether your 16 is 10-6, 9-7 or 5-4-7. Composition-dependent strategy goes one step further and looks at the exact cards that make up your total, because those cards are now out of the shoe and change the odds of the next card very slightly.

The gain from doing this is small. In a six-deck game, perfect composition-dependent play improves on total-dependent basic strategy by roughly 0.01% to 0.04% of the amount wagered. On £10 a hand at 60 hands an hour, that is somewhere between 6p and 24p an hour. For practical purposes, basic strategy and perfect strategy are the same thing.

The jump from casual play to basic strategy is where almost all the value sits. The jump from basic to perfect is a rounding error by comparison, which is why every section below builds on the standard chart rather than replacing it.

Where basic strategy is not quite perfect

The exceptions all follow the same logic: when your hand is made of several small cards, those small cards are gone from the shoe, so the next card is slightly more likely to be a high one. That tilts a close decision. The gap between options in these spots is tiny, so getting them "wrong" costs almost nothing, but they are the only places where basic strategy and perfect strategy disagree.

Three points to take from the table. First, the single-deck rows do not apply to the six-deck and eight-deck shoes used in almost every UK casino and online game; they are listed because single-deck composition effects are larger and people ask about them. Second, most of the multi-deck exceptions are variations on one theme: a hard 16 against a 10 made of lots of small cards stands rather than hits. Third, if you forget every row on this table and play straight basic strategy, you give up at most 0.04%.

The perfect strategy chart

For multi-deck games the chart below is perfect strategy for all but the exceptions above. It assumes 4 to 8 decks, dealer stands on soft 17 (S17), double on any two cards, double after split allowed (DAS), late surrender available, and a hole card game where the dealer checks for blackjack before you act. If your game differs, use the right chart from the blackjack strategy chart, which covers H17, no hole card, no surrender and single-deck rule sets.

Legend: H hit, S stand, D double (hit if not allowed), Ds double (stand if not allowed), P split, R surrender (hit if not allowed).

Hard totals

Soft totals

Pairs

5,5 is never split; it is played as hard 10. If the dealer hits soft 17, five cells change: 11 v A becomes D, A,7 v 2 becomes Ds, A,8 v 6 becomes Ds, 15 v A becomes R, and 8,8 v A becomes R (split if surrender is not offered). If you play a UK no hole card game, see the UK blackjack strategy guide, where 11 v 10, 11 v A, A,A v A, 8,8 v 10 and 8,8 v A all become hit.

House edge under perfect play

The house edge depends far more on the rules than on the difference between basic and perfect play. All figures below assume you follow the chart without error.

The lesson is in the bottom rows. A 6:5 payout adds about 1.4% to the house edge, which is 35 to 140 times the value of every composition-dependent exception combined. Choosing a 3:2 table over a 6:5 one matters more than any chart refinement.

Worked example: £10 a hand, 60 hands an hour, UK typical rules at about 0.5%. Expected loss is £10 x 60 x 0.005 = £3 an hour. Perfect composition-dependent play takes that to about £2.80 to £2.95. The actual result in any single hour will usually be tens of pounds either side of both numbers.

Perfect Blackjack, the Playtech game

Perfect Blackjack is a Playtech RNG table game, and the name refers to its two built-in side bets rather than to any strategy. Rules can vary by casino, so check the in-game rules tab before you rely on the figures below. The typical configuration is:

Strategy adjustments for Perfect Blackjack

The main change is the double restriction. Double only on 10 and 11 adds about 0.18% to the house edge on its own, and 9-to-11 doubling sits between that and the full "any two cards" rule. Every soft double on the chart (A,2 to A,7 against 3 to 6) is not available, so you hit instead. A,7 against 3 to 6 stands, because Ds means stand when doubling is not allowed. Hard 9 doubles are still permitted, so 9 v 3 to 6 doubles as normal.

Under those rules the house edge on the main bet is roughly 0.6% to 0.7% with correct play, depending on the hole card and DAS settings.

The side bets

Perfect Pairs pays when your first two cards are a pair (typically 5:1 mixed, 10:1 coloured, 30:1 perfect), and 21+3 pays when your two cards plus the dealer's up-card form a three-card poker hand. Both carry a house edge of several per cent, typically in the 3% to 7% range depending on the paytable, against roughly 0.6% on the main bet. They are not part of perfect strategy. A £1 side bet at 5% costs five times as much per pound as the main hand at 1%, and there is no play decision that improves it.

Can software play perfectly?

Yes, and this is the simplest way to check the composition-dependent exceptions. A hand calculator such as the one at blackjack strategy calculator takes your exact cards, the dealer up-card and the rule set and returns the expected value of each option. If you enter 5-4-7 against a 10 in a six-deck S17 game, it will show stand and hit within a fraction of a per cent of each other, with stand slightly ahead.

What calculators assume matters. Most assume a freshly shuffled shoe with only the visible cards removed, which is correct for an RNG game that reshuffles every hand and a fair approximation for a live shoe early in the deal. They do not know what has already been dealt in a live shoe, which is the job of card counting, not perfect strategy.

A blackjack strategy trainer is the tool for making the chart automatic. It deals hands, scores your decision against the chart and flags errors. Aim for 100 consecutive hands without a mistake before you worry about any exception on this page; one basic strategy error per 20 hands costs more than every composition-dependent play combined.

Frequently asked questions

Is perfect strategy the same as basic strategy?

For practical purposes, yes. Basic strategy decides by hand total; perfect (composition-dependent) strategy decides by the exact cards. In six decks the difference is worth 0.01% to 0.04% of the amount wagered, and the two agree on every hand except a handful of close calls such as a multi-card 16 against a 10.

Does perfect strategy beat the casino?

No. Perfect strategy reduces the house edge to around 0.3% to 0.5% depending on the rules, but the edge stays with the house. Only card counting in a favourable live game moves the edge to the player, and that is a separate skill covered on the card counting page.

What is the house edge with perfect strategy?

About 0.33% in a six-deck S17 DAS game with late surrender and a hole card, about 0.4% without surrender, and about 0.5% under typical UK no hole card rules. Composition-dependent play shaves 0.01% to 0.04% off each of those. Dealer hitting soft 17 adds about 0.2% and a 6:5 payout adds about 1.4%.

Is there one perfect strategy for every game?

No. The chart above is for 4 to 8 decks, S17, DAS, late surrender and a hole card. H17 changes five cells, no hole card changes five more, single deck changes several, and variants such as Free Bet or Power Blackjack need their own charts. The blackjack strategy charts has one for each rule set.

Is Perfect Blackjack a good game?

The main bet is reasonable but not the best available: the 9-to-11 double restriction pushes the house edge to roughly 0.6% to 0.7%, against about 0.4% to 0.5% on a standard six-deck table. The Perfect Pairs and 21+3 side bets carry several per cent and should be skipped. Check the in-game rules, as casinos can configure the game differently.

How long does it take to learn perfect strategy?

Basic strategy takes most people a few hours of practice with a trainer to reach error-free play, and a week or two of regular sessions to make it automatic. The composition-dependent exceptions take about ten minutes to read and are worth so little that there is no need to memorise them; if you play a lot, the multi-card 16 v 10 stand is the only one worth remembering.

Back to Blackjack strategy: basic strategy explained, with the full chart