Use this Blackjack House Edge Calculator to compare table rules and estimate their long-run cost. Select decks, S17 or H17, blackjack payout, double after split, surrender, resplit aces and dealer blackjack settlement. Enter your original bet and hands per hour to convert the estimate into dollars.
The result is a calibrated approximation for a freshly shuffled shoe and correct strategy. It combines reference benchmarks with rule adjustments; it does not solve every possible casino rule combination exactly. A positive house edge means an expected player loss. Individual sessions can finish above or below that average.
Blackjack House Edge
Compare table rules and estimated cost at your original bet.
Basis: correct strategy, a freshly shuffled shoe, up to four split hands, one card per split ace, and ordinary 21 after splitting. Cut-card effects, counting and side bets are not included. Surrender with no-peek rules needs a separate analysis.
How the estimate is built
| Component | Contribution (pp) |
|---|---|
| Decks + S17/H17 + DAS benchmark | |
| Natural blackjack payout | |
| Doubling restriction | |
| Resplit aces | |
| Surrender | |
| Dealer blackjack settlement | |
| Estimated total |
pp means percentage points. Positive adjustments increase the house edge; negative adjustments reduce it. Components use more decimals than the headline so you can check their sum.
For the next move, use the 4–8-deck basic strategy chart. To compare actions on a particular hand, use the decision EV calculator under its infinite-deck assumptions.
What went wrong?
Default example: six decks, S17, a 3:2 natural payout, DAS, no surrender, no resplit aces, dealer peek and doubling on any first two cards. The estimate is 0.4062%, shown as about 0.41%. At $25 and 80 hands per hour, expected net is −$8.12 per hour and −$32.50 over four hours. Dollar calculations use the unrounded estimate.
How to use the blackjack house edge calculator
- Read the actual blackjack rules. Use the table placard or the game’s help page. Deck count alone is insufficient; two tables with the same number of decks can have different expected costs.
- Choose the dealer rule and DAS. S17 stands on soft 17; H17 hits soft 17. Both stand on hard 17. Double after split determines whether a newly split hand may be doubled.
- Set the natural blackjack payout. A 3:2, 6:5 or 1:1 payout applies to an original two-card blackjack. Ordinary wins still pay 1:1, and two natural blackjacks push.
- Match surrender precisely. Late surrender follows a completed dealer blackjack check. Early surrender against 10 only is a separate, less precisely calibrated option; it does not include surrender against Ace.
- Check what happens to added bets. For no-peek rules, determine whether a dealer blackjack takes only one original bet or all split and double wagers. The calculator separates those outcomes.
- Set resplit aces and doubling restrictions. The model assumes up to four total split hands. Each split ace receives one card; drawing further cards to split aces is outside the model.
- Enter the original bet and playing volume. The bet is the amount before a double or split. Hands per hour counts original rounds, not each hand created by splitting. The four-hour result is an illustration using the same bet and pace throughout.
- Inspect the breakdown. Positive percentage-point contributions increase the house edge; negative contributions reduce it. Reset restores the initial profile. Changing an input clears the old result until recalculation.
For a recommended action under conventional 4–8-deck rules, use the basic strategy chart with rule settings. To compare the expected return of actions on a particular starting hand, use the hand decision EV tool, which uses an infinite-deck model. Those tools answer different questions from the average cost of a rule set.
Blackjack house edge by decks and dealer rules
These baseline figures assume 3:2, dealer peek, no surrender, no resplit aces, doubling allowed on every initial two-card hand and splitting to four hands. They use total-dependent basic-strategy benchmarks on a fresh-shoe basis, derived from the Wizard of Odds house-edge reference. Surrender and other optional adjustments have not been applied.
| Decks | S17 + DAS | S17, no DAS | H17 + DAS | H17, no DAS |
|---|---|---|---|---|
| 1 | −0.1442% | −0.0029% | 0.0464% | 0.1905% |
| 2 | 0.1923% | 0.3350% | 0.3939% | 0.5393% |
| 4 | 0.3530% | 0.4951% | 0.5627% | 0.7076% |
| 6 | 0.4062% | 0.5480% | 0.6187% | 0.7633% |
| 8 | 0.4329% | 0.5744% | 0.6468% | 0.7911% |
The differences are specific to the stated conditions. The calculator selects a benchmark for the combination of decks, dealer rule and DAS, rather than adding the same deck or H17 correction to every game. Negative entries represent the favorable theoretical conditions listed in the caption; they do not establish an advantage at an unverified casino table.
How surrender, resplitting and doubling enter the estimate
Late surrender, resplit aces, restricted doubling and ENHC are measured against the matching baseline. Their adjustments vary with decks, S17/H17 and DAS. For example, under six-deck S17/DAS, late surrender reduces the estimate by 0.072603 percentage points and resplit aces by 0.068542 percentage points.
These are separate adjustments. Applying several at once can leave a small interaction error because the best play under one rule can depend on another. Early surrender against 10 uses the broader −0.24 percentage-point reference adjustment. The −0.39 figure sometimes quoted for early surrender concerns Ace, not 10. See the rule-variation reference and the surrender definitions and strategy.
Peek, original-bet-only and ENHC are different settings
Peek before play: dealer blackjack is checked before the player commits extra split or double bets. Original bet only: if the dealer later has blackjack, exactly one original unit is lost and all additional wagers are returned. ENHC, all bets lost: a dealer blackjack also takes the added split and double wagers.
The first two share a baseline in this estimate under that strict original-bet-only definition. Rules that retain busted wagers as well are a different settlement and are not included. With the default six-deck rules, changing to all-bets-lost ENHC raises the estimate from 0.4062% to 0.5173%. The calculator does not estimate surrender combined with either no-peek option because the exact surrender and settlement terms need separate treatment.
What 6:5 changes in dollars
On a $25 original bet, a winning natural blackjack earns $37.50 at 3:2 and $30 at 6:5: a $7.50 difference. To estimate the average cost, multiply that difference by the probability of a winning natural. A player blackjack tied by a dealer blackjack earns neither payout, so counting every player natural overstates the penalty.
For a fresh shoe with N = 52D cards, A = 4D aces and T = 16D ten-value cards, where D is the number of decks:
P(winning natural) = [2AT / (N(N − 1))] × [1 − 2(A − 1)(T − 1) / ((N − 2)(N − 3))].
The house-edge increase in percentage points is 100 × (1.5 − payout) × P(winning natural), where payout is 1.2 for 6:5 or 1 for even money. This payout component is calculated from card counts; the combined house-edge estimate still has the limitations described below.
| Decks | Winning natural | 6:5 edge increase | Extra cost/hour |
|---|---|---|---|
| 1 | 4.6492% | 1.3948 pp | $27.90 |
| 2 | 4.5783% | 1.3735 pp | $27.47 |
| 4 | 4.5437% | 1.3631 pp | $27.26 |
| 6 | 4.5323% | 1.3597 pp | $27.19 |
| 8 | 4.5266% | 1.3580 pp | $27.16 |
The rule-variation summary lists about 1.39 percentage points for the cost of 6:5. Here the penalty is calculated from the explicit fresh-shoe probabilities above, excluding tied naturals, and ranges from 1.3580 to 1.3948 percentage points by deck count. Use the value for your selected deck count rather than treating 1.39 as a universal constant.
With six decks, switching only the payout from 3:2 to 6:5 adds about 1.3597 percentage points, or $27.19 per hour at $25 and 80 hands. Over four hours, the additional expected cost is $108.78. This is an average difference, not an amount necessarily lost during a particular visit.
Real-World Examples
These are illustrative rule sets, all on the fresh-shoe basis above. The labels describe the scenarios; a room’s minimum bet or atmosphere does not determine its mathematical quality.
Example 1: The “High Limit” Game (Great Odds)
Six decks, S17, 3:2, DAS, late surrender and resplit aces. Also assume dealer peek, double on any first two cards and a four-hand split limit.
Estimated house edge: 0.2651%. Using a $25 original bet for 80 rounds each hour, expected net is −$5.30 per hour and −$21.21 over four hours. The surrender and resplit adjustments improve this estimate, but they do not make the modeled expectation positive for the player.
Example 2: The “Party Pit” Trap (Terrible Odds)
Eight decks, H17, 6:5, DAS, no surrender and no resplit aces. Assume dealer peek and double on any first two cards, with the same split limit.
Estimated house edge: 2.0048%. At the same bet and pace, expected net is −$40.10 per hour and −$160.38 over four hours. Most of the increase relative to a comparable 3:2 game comes from the natural payout.
Example 3: The “Single Deck Trap”
One deck, H17, 6:5 and no DAS, with no surrender or resplit aces. Assume dealer peek, double on any first two cards and splitting non-Ace pairs to four hands.
Estimated house edge: 1.5852%. Expected net is −$31.70 per hour and −$126.82 over four hours. For comparison, the default six-deck S17/3:2/DAS profile estimates 0.4062%. Fewer decks alone do not offset every unfavorable rule.
Understanding the Output
Estimated house edge: in the blackjack house edge calculator, this is the expected casino profit as a percentage of each original bet. A 0.50% edge corresponds to an average player result of −$0.125 on a $25 original bet before multiplying by the number of rounds. This is not the probability that the casino wins an individual hand.
100% − edge: an RTP-style return equivalent using that same initial-bet denominator. The expectation includes the effect of extra split and double bets. It is not a direct measure of total cash returned divided by all money wagered, including those additions; that denominator requires separate accounting.
Expected net per hour = −original bet × original hands per hour × edge / 100. Negative output means an expected loss, positive output an expected profit within the estimate. The four-hour metric multiplies the hourly figure by four. Neither describes the likely size of a short-term swing.
Original bets per hour is the original bet multiplied by the number of rounds. At $25 and 80 rounds, it is $2,000. Actual money committed can be higher after splits and doubles; do not use this figure as a maximum cash requirement.
For the separate question of short-term variation, use the variance and standard-deviation calculator or the session distribution simulator. House edge supplies an average, while session outcomes also depend on variance and the stopping conditions.
How accurate is this estimate?
The blackjack house edge calculator compares rules using calibrated reference values. It is not an independent finite-shoe strategy solver. Its base includes total-dependent strategy on a fresh-shoe basis; optional rules enter as separate increments measured from that base.
In checks of 360 compatible rule profiles, the combined estimate differed from the corresponding full reference combination by at most about 0.006 percentage points before payout rounding. That measures agreement with the chosen reference, not a guaranteed error bound for casino play. Early surrender against 10, unusual settlement rules and cut-card effects are outside that comparison.
A shoe dealt to a cut card can have a different expectation from the fresh-shoe benchmark. The cut-card analysis explains that distinction. This calculator does not ask for penetration and does not simulate a changing shoe. Counting, strategy mistakes, special bonuses, side bets, draw-to-split-aces rules and altered decks also require separate models.
Use the estimate to understand the direction and approximate size of rule changes. For a close comparison or a negative edge, confirm the complete rules and the applicable strategy. The extra decimals in the breakdown let you check the arithmetic; they do not certify a casino’s return to that precision.
Related Blackjack Tools
- Basic Strategy Calculator — conventional 4–8-deck starting-hand decisions, including H17, DAS and surrender.
- Decision EV Calculator — infinite-deck action values and the expected cost of choosing an alternative.
- Card Counting Calculator — running and true count calculations.
- Variance Calculator — standard deviation and long-run sample-size questions.
- Session Simulator — modeled session outcomes under its specified assumptions.
- Bankroll Calculator — bankroll requirements for a specified edge and risk model.
- Penetration Calculator — deck estimation and cut-card depth.
- True Count Calculator — converting a running count using estimated decks remaining.
Frequently Asked Questions (FAQ)
What is a good house edge in blackjack?
Compare complete rules under the same model and playing strategy. A lower positive edge means a lower expected cost at the same bet and pace. The default six-deck S17/3:2/DAS example estimates about 0.41% on a fresh-shoe basis; this is a benchmark, not a universal grade for a casino.
Why is 6:5 blackjack payout so bad?
It pays less for a winning natural. A $25 blackjack earns $37.50 at 3:2 and $30 at 6:5. In the six-deck fresh-shoe model, that adds about 1.3597 percentage points to the house edge, or $27.19 per hour at 80 hands. Tied naturals are excluded from the payout difference.
What is the difference between S17 and H17?
S17 means the dealer stands on soft 17; H17 means the dealer hits soft 17. Both stand on hard 17. The calculator uses different baseline values for each setting at the selected deck count and DAS rule. Basic-strategy decisions can change as well.
Does the number of decks matter?
Yes. Deck count changes natural-blackjack probabilities and other aspects of the game. Its effect depends on the rest of the rules and the dealing model. This calculator uses separate benchmarks for 1, 2, 4, 6 and 8 decks; it does not apply one universal deck correction.
What is RTP in blackjack?
RTP is often presented as 100% minus house edge. Here that is a return equivalent per original bet, with split and double outcomes already reflected in the expectation. It is not a measured ratio of cash returned to all stakes committed, and it does not predict a session result.
How do I calculate my expected loss per hour?
Multiply the original bet by original hands per hour and by house edge divided by 100. At $25, 80 hands and 0.50%, expected loss is $10 per hour. The widget displays the player’s expected net, so that loss appears as a negative amount.
Is this house-edge result exact?
No. The calculator combines reference benchmarks with calibrated rule adjustments. It does not independently solve a finite shoe for every combined rule set. Unusual rules, cut-card effects and early surrender can require a more detailed analysis.
Why can the estimate be negative?
Some favorable theoretical rule sets can produce a negative house edge. The calculator preserves that sign rather than clipping the result. Check the full rules, dealing method and appropriate strategy before interpreting a negative estimate as an available player advantage.
Set spending and time limits before playing. Mathematical analysis does not make gambling risk-free. If gambling is causing harm, see the warning signs and confidential help resources.