Two tables can both say “Blackjack” and charge very different mathematical prices. The difference is the rule package.
A strong table might pay 3:2 on a natural, stand on soft 17, allow double after split, allow flexible doubling, offer late surrender, and permit generous resplitting. A weak table might pay 6:5, hit soft 17, restrict doubles, remove surrender, and limit splits.
This page is the rule-cost ledger: what the common switches are worth and how to compare a whole table without pretending blackjack has one universal house-edge number.
For the definition of house edge itself, use Blackjack House Edge. For a single-rule deep dive, follow the links from the table below.
Start with a benchmark, then adjust carefully
Rule-effect tables need a reference game. One widely used benchmark starts from an eight-deck game where the dealer stands on soft 17, the player may double on any first two cards, double after split is allowed, and resplitting to four hands is permitted.
Under that type of benchmark, common changes have approximate effects like these:
| Rule change | Approximate effect on player return | Equivalent direction for house edge |
|---|---|---|
| Blackjack pays 6:5 instead of 3:2 | -1.39% | House edge increases ~1.39 points |
| Dealer hits soft 17 instead of stands | -0.22% | House edge increases ~0.22 points |
| No double after split | -0.14% | House edge increases ~0.14 points |
| Double only on 10-11 | -0.18% | House edge increases ~0.18 points |
| Split limited to two hands | -0.10% | House edge increases ~0.10 points |
| Late surrender available | +0.08% | House edge decreases ~0.08 points |
| Resplit Aces available | +0.08% | House edge decreases ~0.08 points |
| Six decks instead of eight, equal rules | +0.02% | House edge decreases ~0.02 points |
| Single deck instead of eight, equal rules | +0.48% | House edge decreases ~0.48 points |
These are benchmark effects, not numbers to add blindly in every game. Rule interactions, strategy adjustments, no-hole-card procedures, and deck composition can make the exact combined edge differ from a simple sum.
But the scale is extremely useful. It tells you which signs on the table deserve attention first.
The payout is usually the first thing to read
A natural blackjack is one of the player’s most valuable outcomes.
At a $20 wager:
- 3:2 blackjack profit = $30;
- 6:5 blackjack profit = $24.
The $6 gap occurs every time the player is paid for a winning natural.
The benchmark penalty of roughly 1.39 percentage points is enormous in blackjack terms. It is larger than several smaller unfavorable rules combined.
That is why an attractive “single deck” sign can be misleading. A single-deck 6:5 game can be much more expensive than a six- or eight-deck 3:2 shoe with otherwise reasonable rules.
For the full payout arithmetic, see 3:2 vs 6:5 Blackjack.
H17 versus S17 is the next visible price signal
A dealer standing on soft 17 (S17) stops with hands such as A-6. A dealer hitting soft 17 (H17) draws again and sometimes improves.
The common benchmark effect is about 0.22 percentage points against the player when H17 replaces S17.
That number is meaningful but should be placed in perspective:
- it is much smaller than the 6:5 payout penalty;
- it is larger than the value of several minor split-rule changes;
- it also changes some basic-strategy decisions.
The dedicated H17 house-edge page and S17 house-edge page explain the two sides separately.
Double rules are a cluster, not one switch
“Doubling allowed” is incomplete. Check:
- any first two cards or only selected totals;
- whether doubling after split is allowed;
- whether doubling for less is permitted;
- whether special split-Ace restrictions apply.
Restricting doubles removes the option to increase the wager exactly when a strong player total and weak dealer upcard make the extra money valuable.
Two common benchmark penalties are approximately:
- no DAS: +0.14 percentage points to the house edge;
- doubling only 10-11: +0.18 points compared with doubling on any first two cards.
The effects are not identical because they remove different decisions.
For the rule mechanics, see Double After Split. For the isolated house-edge value, see House Edge with DAS.
Surrender is small but real
Late surrender lets the player replace selected very bad decisions with a fixed half-loss.
Its benchmark value is around 0.08 percentage points of player return under common assumptions. That is not enough to rescue a 6:5 table, but it can be the deciding feature between two otherwise similar 3:2 games.
The effect depends on correct strategy. Offering surrender does not help a player who never uses it when appropriate or who uses it on the wrong hands.
See Blackjack Surrender for the expected-value test and common strategy cells.
Resplitting rules matter at the margins
A table can limit total hands after a split, prohibit resplitting Aces, or restrict the number of cards to split Aces.
These rules affect relatively infrequent situations, so their house-edge effect is normally smaller than payout or soft-17 rules. But “small” is not “zero.” A benchmark value of roughly +0.08 percentage points of player return for resplitting Aces is enough to matter when comparing otherwise close tables.
Read Resplitting Aces for the procedure and exposure implications.
Deck count only makes sense after the other rules are matched
Fewer decks are generally favorable when the rest of the rules are equal because card-removal effects are stronger and naturals behave slightly differently in the composition.
But casino products rarely keep everything else equal. A single-deck game may pay 6:5, restrict doubling, or use H17. A six-deck game may offer 3:2, S17, DAS, surrender, and better split rules.
The correct order is:
- payout;
- H17/S17;
- doubling rules;
- surrender;
- split/resplit rules;
- deck count;
- no-hole-card / peek settlement rules;
- table speed, minimum, and side bets for total-dollar cost.
The House Edge by Deck Count page isolates the deck effect.
A worked comparison: why the rule package beats the label
Suppose Table A is advertised as “single deck” but has:
- 6:5 blackjack;
- H17;
- no DAS;
- no surrender.
Table B uses six decks but has:
- 3:2 blackjack;
- S17;
- DAS;
- late surrender.
A casual comparison may prefer Table A because “single deck is better.” The rule ledger points the other way.
Using benchmark effects only to understand scale:
- single deck may improve return roughly 0.48 points versus eight decks;
- 6:5 can cost about 1.39 points;
- H17 can cost about 0.22 points;
- no DAS can cost about 0.14 points;
- no surrender gives up roughly 0.08 points of available value.
The favorable deck count is overwhelmed by the payout and rule restrictions. Exact house edge still requires calculating the full ruleset, but the table-selection conclusion is already clear.
Convert rule differences into dollars of action
Once you know the house-edge difference, translate it into money.
If two tables differ by 0.50 percentage points and you expect $4,000 of original-bet action:
[ \text{Extra expected loss} = 4{,}000 \times 0.005 = $20 ]
A 1.00-point difference over the same action is:
[ 4{,}000 \times 0.01 = $40 ]
That is why “only half a percent” is not trivial for a long session or a high average wager.
The Expected Loss Calculator helps translate percentage differences into dollars without confusing theoretical loss with the actual session result.
Rule effects are not perfectly additive
It is tempting to take a base edge and add every line from a rule-variation table. That produces a useful rough comparison but not always the exact answer.
Why?
- Basic strategy changes when rules change.
- Some rules affect the same hand categories.
- Deck count changes composition-dependent effects.
- No-hole-card rules can change the risk on doubles and splits against a dealer natural.
- Surrender value depends on whether it is early or late and on the dealer procedure.
For an exact table estimate, use a calculator that accepts the whole ruleset rather than a homemade sum of rounded effects.
The House Edge Calculator is the site’s appropriate next step after the rule screen.
A rough adjustment model is useful for screening, not certification
For quick comparison, a player can think in terms of:
[ HE_{rough} \approx HE_{baseline} + \sum \Delta_i ]
where (\Delta_i) is the house-edge direction of each rule change.
Suppose a baseline table is estimated at 0.40% under its complete rules. A nearby table changes only two items:
- dealer hits soft 17: roughly +0.22 points to the house edge;
- DAS removed: roughly +0.14 points.
A rough screen gives:
[ 0.40% + 0.22% + 0.14% \approx 0.76% ]
That is useful for deciding that the second table is probably more expensive. It should not be published as an exact edge without running the full rule set through a strategy-aware calculator.
The distinction between screening estimate and certified calculation is important because rounded rule effects are not always perfectly additive.
Table minimum and house edge answer different cost questions
A $10 table with a 1.5% edge and a $25 table with a 0.5% edge create a real trade-off if the player will bet only the minimum.
At 70 hands per hour:
$10 table:
[ 10\times70\times0.015=$10.50\text{ expected loss/hour} ]
$25 table:
[ 25\times70\times0.005=$8.75\text{ expected loss/hour} ]
Despite the higher minimum, the stronger game is slightly cheaper in this example.
Change the strong table minimum to $50:
[ 50\times70\times0.005=$17.50 ]
Now the weaker low-minimum table has the lower expected dollar cost per hour for a strict minimum bettor, even though its percentage price is worse.
This is why table selection needs both edge per dollar and dollars wagered. “Always choose the lowest house edge” is incomplete if it forces a player to wager much more money than intended.
Speed changes cost without changing the per-hand edge
A continuous shuffler, electronic interface, heads-up table, or efficient dealer can increase hands per hour. If the rule set and strategy are unchanged, the house edge per unit wagered may be the same while hourly theoretical loss rises because more action is completed.
For a $25 player at 0.5%:
- 50 hands/hour: $6.25 expected loss/hour;
- 80 hands/hour: $10.00;
- 120 hands/hour: $15.00.
The table did not become mathematically “worse” per hand. The player simply bought more hands per hour.
That distinction belongs in any practical comparison of land-based and electronic blackjack.
Penetration is not a normal basic-strategy house-edge switch
Players often hear card counters discuss penetration—the fraction of the shoe dealt before the shuffle—and assume deeper penetration automatically lowers the house edge for everyone.
For a flat-betting basic-strategy player, penetration is mainly a pace and card-counting issue rather than a direct posted-rule adjustment like 6:5 or H17.
For a counter, deeper penetration can be extremely important because it creates more opportunities to reach informative high or low true counts before the shuffle.
The same feature therefore has different value depending on the player’s information model. Avoid adding a “penetration discount” to a basic-strategy house edge unless the calculation explicitly models it.
Dealer no-hole-card rules can change added-wager exposure
In an American peek game, a dealer Ace or ten-value upcard may be checked for blackjack before the player invests additional money through doubles and splits.
In some no-hole-card games, the player may act before the dealer’s second card is resolved. The financial effect depends on what happens to additional double/split wagers if the dealer later makes blackjack.
A rule where only the original wager is lost is different from a rule where all added wagers are lost.
This is why “European no hole card” should not be reduced to a label. Read the actual settlement rule.
The No Peek Rule page handles that procedure in detail.
Side bets do not inherit the main game’s low edge
A blackjack table can advertise favorable 3:2 rules and still place high-edge side bets beside the main betting circle.
The main-game house edge and a side-bet house edge are separate expected-value streams. If a player wagers $25 on the main game and $5 on a side bet every hand, a blended theoretical cost is:
[ EL = A_m h_m + A_s h_s ]
where:
- (A_m) = main-game action;
- (h_m) = main-game house edge;
- (A_s) = side-bet action;
- (h_s) = side-bet house edge.
A small side wager with a much larger edge can erase much of the benefit gained by choosing a strong main table.
See Blackjack Side Bets for paytable-by-paytable evaluation.
Comps reduce net cost; they do not rewrite the rule edge
If the casino returns part of theoretical loss through points, food, rooms, or discretionary offers, the player may receive economic value. That does not change the probability and payout structure of the game itself.
A 0.6% blackjack game remains a 0.6% blackjack game even if the player receives a meal. The more useful separate calculation is:
[ \text{Net expected entertainment cost} \approx \text{theoretical loss} - \text{usable reward value} ]
The Blackjack Comps Value page explains why retail value, personal value, and casino reinvestment are different numbers.
Rule effects can change the correct strategy chart
House edge is not calculated by changing rules while holding player decisions frozen.
If H17 replaces S17, some soft-hand, double, and surrender decisions can change. Removing DAS changes pair strategy. Changing deck count can alter borderline doubles and splits. Adding surrender creates new first-decision options.
A valid house-edge figure therefore assumes the player uses the correct strategy for the new rules.
This is also why using one laminated “universal blackjack chart” across every table can create a mismatch. The rules set the price and the rules also set the best response.
The rule sheet is part of the economic contract
From an operating perspective, table rules serve several functions at once:
- define settlement and dealer procedure;
- set the mathematical return of the product;
- differentiate low-limit and premium games;
- control training and surveillance expectations;
- determine which strategy chart actually applies.
A floor sign saying “Blackjack pays 3 to 2” is not decoration. It is part of the payout contract. So are H17/S17, surrender availability, double restrictions, and special split conditions.
The player who checks those terms before buying in is doing the casino-game equivalent of reading the price before ordering.
Official rules show why “blackjack” is not one fixed contract
Gaming regulations commonly define the core actions but allow multiple approved configurations. Colorado’s blackjack rules, for example, define insurance, surrender, split and double procedures and permit approved variations within the regulated framework. See the Colorado blackjack rules.
The operational lesson is simple: the game name tells you the family. The posted rules tell you the product you are actually buying.
Mathematical source for the benchmark effects
The Wizard of Odds blackjack rule-variation table publishes the player-return effects used as benchmarks on this page, including H17, DAS, surrender, deck count, split limits, and 6:5 payout.
Those figures assume specific reference rules and proper strategy adjustments. Treat them as controlled comparisons, not universal constants detached from assumptions.
A table-selection checklist that takes less than a minute
Before buying in, read the felt or placard for:
- Blackjack payout: 3:2 preferred; 6:5 is a major warning.
- Dealer soft 17: S17 preferred over H17.
- Doubling: any first two cards preferred over narrow restrictions.
- DAS: allowed is better.
- Surrender: late surrender adds value when used correctly.
- Resplits / RSA: more flexibility generally helps.
- Deck count: compare only after the larger rules are known.
- Dealer blackjack procedure: know whether added double/split wagers are exposed under no-hole-card rules.
- Side bets: price them separately; their house edge is not the main game’s edge.
If a table does not make its payout or rules clear, ask before the first wager. The cheapest information in blackjack is the information available before you sit down.
Responsible play
Choosing a lower-edge ruleset reduces mathematical cost. It does not make a session predictable, safe, or profitable.
A 0.4% game can still produce a large losing night. A 1.5% game can still produce a lucky winning night. House edge is about repeated expectation, not a promise about tonight’s cards.
Use better rules to pay less for the same entertainment—not as a reason to increase the bet or extend the session.
Author / Editorial Note
This page is written from a land-based casino operations perspective. The goal is not to sell blackjack as beatable entertainment. The goal is to show how table rules change the real mathematical cost, so players and readers can judge blackjack with less superstition and more structure.
Bottom line
Blackjack house edge is built one rule at a time. Check the payout first, then H17/S17, doubling and DAS, surrender, split rules, deck count, and dealer-blackjack procedure.
The strongest single habit is to compare complete rule packages rather than labels. “Single deck,” “low minimum,” and “blackjack” are not prices. The rule set is the price.