Blackjack House Edge Calculator
Compare two tables by their rules and stakes. See estimated house edge and long-run expected cost with correct basic strategy. Actual session results can be very different.
Comparison updated September 27, 2026 · Assumptions and calibration
Rounded estimates, not exact game analysis. Both tables assume double on any first two cards, up to four split hands, no resplitting aces and one card to each split ace, and a US peek game. Different rules need a different calculation.
Practice the decisions behind the estimate
These estimates assume basic strategy. Play free practice hands with coaching, and set the supported table rules in Settings.
For the return on one winning bet, use the payout calculator. For hit, stand, double or split on a specific hand, use the hand calculator. House edge describes a long-run average, not your chance of losing the next hand.
How these estimates are calculated
We add rounded rule effects to a 0.62% six-deck H17, DAS, 3:2, no-surrender baseline. That baseline approximates basic strategy with a continuous shuffler or reshuffle after each hand. Cut-card games can have a higher edge. The model does not recompute all rule interactions.
We checked these three six-deck examples against the Wizard of Odds house-edge calculator on September 27, 2026, using its “basic strategy with continuous shuffler” result and the same split, double and surrender rules stated above.
| Rules | Our estimate | Reference | Difference |
|---|---|---|---|
| H17, 3:2 | 0.62% | 0.61873% | +0.00127 pp |
| S17, 3:2 | 0.41% | 0.40622% | +0.00378 pp |
| H17, 6:5 | 2.01% | 1.97842% | +0.03158 pp |
These examples are calibration checks, not an error bound for every combination. A small estimated advantage, particularly near zero, is not evidence of a beatable game. The hand-engine benchmark tests a separate model and does not certify these house-edge estimates.
Reading the dollar comparison
Expected loss = initial bet × hands per hour × house edge / 100. A lower percentage can still produce a larger dollar loss at a higher bet or faster table. This is a long-run average, not a prediction or limit for one session. Splits and doubles are already incorporated into the edge per original wager; do not add them again to the initial bet.
What each rule is worth
| Rule change | Approximate change (percentage points) |
|---|---|
| Blackjack pays 6:5 instead of 3:2 | +1.39% |
| 8 decks → single deck | −0.48% |
| Dealer stands on soft 17 (S17) | −0.21% |
| No double after split | +0.14% |
| Late surrender offered | −0.08% |
| Resplit aces allowed | −0.08% |
| Double on 10–11 only | +0.18% |
The published rule-effect table uses an eight-deck S17 reference. Our 0.21-point soft-17 adjustment is calibrated to the six-deck example above; surrender remains a rounded 0.08-point estimate. Effects are not exact across all combinations.
Two takeaways: the payout line dwarfs everything else, and "single deck!" advertising means nothing if it comes with 6:5. Compare both the edge and the amount wagered; lower stakes can matter more than a small rule difference.
Frequently asked questions
Where do these numbers come from?
They are rounded additive estimates. The calibration table above shows actual differences from a published calculator for three matching six-deck configurations. Other rule combinations have not been certified; consult a full rules calculator when precision matters.
Is a 0.6% house edge actually beatable?
Not by bet sizing — no betting system changes it. Card counters beat it by betting more only when the remaining shoe favors them; everyone else should treat ~0.6% as the price of the entertainment.
Why does the number of decks change the edge?
Fewer decks mean more blackjacks for you, more favorable doubles, and bigger swings from card removal. The effect shrinks fast: 1→2 decks costs ~0.29%, but 6→8 decks costs just ~0.02%. Full breakdown: how many decks matter.