Deck count: single-deck vs multi-deck blackjack

Fewer decks lower the house edge when every other rule is held constant. Wizard of Odds measures a single deck as 0.48 percentage points better for the player than eight decks, double deck 0.19 points better and six decks 0.02 points better. Casinos know this, which is why single and double-deck games so often arrive with 6:5 payouts, H17 and tighter doubling rules that take back more than the deck count gives.

House edge by deck count

Two published tables cover the common rule sets. Wizard of Odds’ deck-count analysis assumes dealer hits soft 17, 3:2 payout, double any two cards, double after split, no surrender, resplit any pair including aces up to three times, dealt from a continuous shuffler. BlackjackInfo’s figures assume dealer stands on soft 17, double any two, double after split.

The columns differ by about 0.2 points throughout, which is the cost of H17 discussed in H17 vs S17. Within each column the pattern is the same: the big step is from one deck to two, the next from two to four, and from four decks up the differences shrink to hundredths of a percent. Six versus eight decks is almost irrelevant.

Both tables describe liberal rules that are rarely offered with one or two decks. The single-deck S17 DAS game in BlackjackInfo’s table gives the basic-strategy player a small edge, and no casino deals it.

Why fewer decks help

Every card dealt from a single deck shifts the remaining composition by almost 2%; from an eight-deck shoe, by about 0.24%. The player’s options are worth more when the next card is more predictable. Wizard of Odds breaks the 0.563-point difference between one and eight decks (under its H17 rules) into five parts:

The intuition for the largest item: suppose you hold 10-6 against a dealer 10 in a single deck. The three cards you can see are all ten-value cards, which leaves 13 tens among 49 unseen cards (26.5%) rather than the 16 in 52 you started with (30.8%). Your hit is safer, but so is the dealer’s; the net effect of composition on each decision is what the table measures, and it favours the single-deck player overall.

The naturals component is small, which surprises people. The arithmetic: in a single deck the probability of a two-card natural is 2 × (4/52) × (16/51) = 4.827%; in eight decks it is 2 × (32/416) × (128/415) = 4.745%. The difference is under a tenth of a percent of hands, and because dealer naturals move the same way it is worth only 0.06 points.

How casinos compensate

Because deck count is visible from across the room, it sells tables. Casinos then recover the value through rules that are harder to see. The common package in Las Vegas is “Single Deck, Blackjack Pays 6 to 5”, and BlackjackInfo describes 6:5 as “overwhelmingly common” on single-deck tables there.

Put together, a single-deck game with 6:5, H17 and double on 10–11 only sits about 1.3 points worse than a plain eight-deck 3:2 S17 DAS game. Wizard of Odds’ correspondence records single-deck games whose edge rose from 0.05% to 1.44% on the payout change alone. 6:5 blackjack vs 3:2 shows the full arithmetic. A double-deck game paying 3:2 with S17 and DAS, at about 0.20%, is the best commonly available pitch game; the same game under H17 is about 0.40%, per BlackjackInfo.

The practical rule is to rank tables by payout first, soft-17 rule second, doubling and splitting rules third, and deck count last. Choosing a table applies that order.

Penetration and the shuffle

Deck count also governs how the game is dealt. A single or double deck is hand-held and pitched face down; six or eight decks come from a shoe and are dealt face up. The shoe game guide covers the procedure.

Penetration is the proportion of the cards dealt before the shuffle. It does not change the basic-strategy house edge, but it determines how much information a card counter can act on. Because a count moves fastest in a single or double deck, casinos counter by cutting off a large share of the pack: BlackjackInfo notes that shallow penetration in double-deck games means “the best counts rarely arrive”. A single-deck game shuffled after every round, as online RNG games do, has no penetration at all and cannot be counted.

The cut-card effect is a subtler consequence of dealing to a fixed point. Wizard of Odds’ simulations show a hand-shuffled game with a cut card carries a slightly higher basic-strategy edge than a game shuffled after every hand, because rounds that use up more cards (small-card rounds) leave room for extra hands before the cut card. The effect is 0.113 points in single deck, 0.063 in double deck, 0.034 in four decks, 0.020 in six and 0.014 in eight. A continuous shuffling machine removes it, so a CSM is marginally better for a basic-strategy player per hand; it is worse per hour because it deals faster, and it ends counting entirely.

Wizard of Odds’ deck-count figures above are measured with a continuous shuffler; add the cut-card effect to approximate a hand-shuffled pitch game, and the single-deck advantage shrinks by a further tenth of a point.

What this means at the table

Single and double-deck games with 3:2 payouts and S17 are excellent and increasingly rare. Single-deck games with 6:5 are among the worst games on any floor. The rest of the deck-count question, four versus six versus eight, is worth a few hundredths of a percent and should never decide where you sit. Basic strategy itself changes little with deck count: Wizard of Odds publishes separate single-deck and double-deck charts, but the differences are a handful of borderline cells, and the six-deck S17 chart on this site is a sound default for any shoe game.

Blackjack World Tour is a free iOS blackjack game played with virtual chips; it is a social casino with in-app purchases and no real-money gambling. In a live casino the deck count sign is advertising; the payout line is the price.

Related reading

• 6:5 blackjack vs 3:2
• How rule variations change the house edge
• Card counting
• How a shoe game is dealt