Blackjack Odds and RTP: House Edge, Probabilities and Payouts

Blackjack odds start with one number, the RTP (return to player), which is usually between about 99.3% and 99.7% when you play perfect basic strategy at a table with good rules, which means a house edge of roughly 0.3% to 0.7%. A six-deck game where the dealer stands on soft 17, doubling after splits is allowed and late surrender is offered has a house edge of about 0.28% (RTP about 99.72%). Bad rules change that fast: a table that pays 6:5 for blackjack instead of 3:2 adds 1.39 percentage points to the house edge on its own, and a player who ignores basic strategy can easily face 2% or more.

In other words, bro, blackjack is the best-value table game in most casinos, but only if you pick the right table and make the right decisions. This guide sets out the real odds behind every part of the game: hand outcomes, the chance of a blackjack, dealer bust rates, what each rule costs, the RTP of popular online variants, and why most side bets are a trap.

Blackjack RTP and House Edge: What the Numbers Mean

RTP and house edge are two views of the same thing. RTP is the share of all money wagered that returns to players over the long run; house edge is the share the casino keeps. RTP = 100% − house edge. A 0.5% house edge means a 99.5% RTP, or an average cost of 50 cents per $100 wagered.

Three details trip people up:

• Blackjack RTP assumes perfect basic strategy. Unlike a slot, where the RTP is fixed by the game, blackjack’s published figure is what a player gets by making the mathematically best decision on every hand. Real players who guess, follow hunches or “play like the dealer” get a lower return.
• House edge is measured per initial bet. Doubles and splits put more money on the table, but the figure is quoted against your opening bet.
• House edge is not the casino’s “hold”. Casinos report how much of the chips bought at a table they keep, and that hold percentage is much higher than the house edge because players recirculate the same money many times.

Blackjack Odds at a Glance

Typical Blackjack House Edges by Rule Set

These figures combine the rule effects in the next section and the Wizard of Odds rule-variation tables. They are approximate because the effects interact slightly, but the order of magnitude is reliable: rules and decisions move blackjack’s cost from well under 1% to several percent.

What Each Rule Does to the House Edge

The Wizard of Odds measures every common rule against a baseline of eight decks, dealer stands on soft 17, double on any first two cards, double after split allowed and splitting to four hands. Positive numbers help the player; negative numbers help the casino.

The 6:5 trap, in dollars

A blackjack happens about once every 21 hands. At 3:2, a $10 blackjack pays $15. At 6:5 it pays $12. That $3 shortfall on roughly 4.75% of hands is why the rule costs 1.39 points. Worked over 100 hands at $10 a hand ($1,000 wagered), a 3:2 game with a 0.5% edge costs about $5 on average; the same table paying 6:5 costs about $19. Single-deck 6:5 games are often marketed as player-friendly because single deck sounds good. They are not: the +0.48 from one deck does not come close to covering the −1.39 from the payout.

Deck count

Fewer decks help mainly because blackjacks become slightly more likely and doubling hands improve. Beyond four decks the difference is small. The payout and dealer rules matter far more than the deck count.

How Often You Win, Lose and Push

Here is the surprise for most players: even with perfect basic strategy, you lose more hands than you win. In a six-deck game with S17, double after split and late surrender, the Wizard of Odds calculates:

So how is the house edge only 0.28%? Because the hands you win are worth more on average than the hands you lose:

• Blackjacks pay 3:2. You get 1.5 units on about 4.75% of hands, while the dealer’s blackjack only takes 1 unit from you.
• Doubling and splitting let you put more money out in good spots, such as 11 against a dealer 6. You win those extra bets more often than you lose them.
• Surrender lets you cut the loss on the worst hands to half a bet.
• Standing on low totals lets you avoid busting when the dealer is likely to bust.

The dealer’s only advantage is acting last: if you bust, you lose even if the dealer would also have busted. Everything in basic strategy is about clawing back that one structural edge.

The Odds of Getting a Blackjack

A natural blackjack is an ace plus any ten-value card (10, J, Q, K) as your first two cards. The probability depends slightly on the number of decks:

The calculation for six decks: 2 × (24 aces ÷ 312 cards) × (96 tens ÷ 311 remaining cards) = 4.75%. The dealer gets one equally often. When both of you have blackjack, it is a push.

What Starting Hands You Will Be Dealt

Your first two cards set the odds for the rest of the hand, and the distribution is not what most players imagine. Exact figures for a six-deck shoe:

More than a third of all starting hands are stiffs between 12 and 16. That is the core of the game’s difficulty and the reason the dealer wins more hands overall. Basic strategy cannot make those hands profitable; it only makes them lose less. The money is made back on the 11.9% of doubling hands, the blackjacks, and the occasional good split.

Spanish 21 and Double Exposure: Odds of Two Classic Variants

Two land-casino variants show how rule packages trade off against each other.

Spanish 21

Spanish 21 is dealt from six or eight Spanish decks, which have the four pip tens removed (jacks, queens and kings stay). Removing tens hurts the player, so the game adds generous rules: a player 21 always wins, late surrender, doubling on any number of cards, “double down rescue” (surrender after doubling), resplitting aces, and bonus payouts for five-card 21 (3:2), six-card 21 (2:1) and certain 6-7-8 and 7-7-7 combinations. The Wizard of Odds puts the house edge at 0.40% when the dealer stands on soft 17, 0.42% when the dealer hits soft 17 with redoubling allowed, and 0.76% when the dealer hits soft 17 without redoubling. That is competitive with good standard blackjack, but only with Spanish 21’s own strategy: regular basic strategy plays these hands wrongly.

Double Exposure

In Double Exposure both dealer cards are dealt face up, a huge informational advantage. To pay for it, blackjacks pay only even money and the dealer wins all ties except blackjack. The Wizard of Odds lists house edges from 0.26% to 1.47% depending on the casino’s exact rules. The game is now rare in land casinos, and when it does appear online the rules panel shows the RTP to compare.

The lesson from both: a rule that sounds good or bad in isolation tells you little. Compare the published RTP, or the full rule package, against a standard 3:2 game.

Dealer Bust Odds by Up Card

The dealer’s up card is the single most useful piece of information at the table. These figures are for six decks, dealer stands on soft 17, after the dealer has checked for blackjack (Wizard of Odds).

Two practical lessons. Against a 5 or 6 the dealer busts more than four times in ten, which is why basic strategy stands on stiff hands (12 to 16) and doubles aggressively there. Against a 7 the dealer most often finishes on exactly 17, so a player 17 against a 7 is usually a push or a narrow result, and a player 16 against a 7 should hit.

Your Bust Odds When You Hit

These are close approximations treating the shoe as endless, so each rank appears 1 time in 13 and ten-value cards 4 times in 13.

A 61.5% chance of busting on 16 sounds like a reason never to hit it. But against a dealer 10 you lose most of the time anyway if you stand, because the dealer only busts about one time in four with a ten showing. Bust odds alone never decide a hand. The decision depends on the expected value of each option, which is what basic strategy summarises.

Expected Value of Key Decisions

Here are exact expected values from the Wizard of Odds’ six-deck, dealer-stands-on-soft-17 tables, per unit of your original bet. A value of −0.54 means you lose 54 cents per $1 on average from that spot.

Notice how small some of the gaps are. Hitting 16 against a 10 beats standing by less than a cent per dollar, which is why players argue about it. Other gaps are large: doubling 11 against a 6 roughly doubles your expected profit from the hand, and splitting tens against a 6 throws away about 26 cents per dollar. Basic strategy simply picks the highest number in each row for every possible hand.

Blackjack Payouts at a Glance

Insurance and Even Money: The Odds Say No

When the dealer shows an ace, you can take insurance: a side bet of up to half your stake that the dealer has a ten underneath. It pays 2 to 1. For it to break even, the dealer would need a blackjack one time in three (33.3%). In a six-deck shoe, with only the dealer’s ace known, 96 of the 311 unseen cards are tens, which is 30.9%.

Expected value of insurance = 3 × 30.9% − 1 = −7.4%. So:

That is more than 20 times the house edge of the main game. “Even money”, offered when you hold blackjack against an ace, is exactly the same bet in disguise: taking it locks in a 1:1 win but gives up more on average than letting the 3:2 hand ride. Only a card counter who knows the remaining shoe is rich in tens should ever insure.

Side Bet Odds: Where the RTP Collapses

Side bets pay big multiples, which is exactly why they cost so much. Figures from the Wizard of Odds:

In expected-loss terms, a small side bet can cost more per hand than a much larger main bet. On the main bet at 0.5% you lose about 12.5 cents per hand; on a $5 Perfect Pairs bet at 6.1% you lose about 30.5 cents. If RTP matters to you, skip the side bets.

RTP of Popular Online and Live Blackjack Games

Online blackjack games print their RTP in the help or rules panel, always assuming optimal strategy. Figures for well-known live games from Evolution, the largest live-casino supplier, as published for its standard rule sets:

The RTP printed in the panel is a maximum. It assumes the game’s own optimal strategy, which differs between variants: Free Bet Blackjack, for example, rewards taking every free double and split, and the Wizard of Odds calculates a 1.04% house edge for its standard six-deck, dealer-hits-soft-17 version. Power Blackjack’s missing 9s and 10s change which hands you should double. Using standard basic strategy on these games returns less than the headline figure.

RNG versus live dealer odds

RNG (software) blackjack usually reshuffles a virtual shoe before every hand, so earlier cards never matter. Live tables deal from a physical shoe, typically eight decks, and shuffle well before the end. The rules, not the format, decide the RTP: check the panel for deck count, soft 17, the blackjack payout, surrender and double rules before you sit down.

Volatility: Why Short Sessions Swing So Much

The house edge is tiny compared with the swings. In the six-deck game above, the standard deviation is about 1.142 units per hand. For $10 hands:

Even after 10,000 hands, around four players in ten are still ahead. Compare that with European roulette, where almost nobody betting even money is ahead after 10,000 spins. Low house edge plus moderate volatility is why blackjack is the classic “best odds” table game. It is also why short-term results tell you almost nothing about whether you are playing well. Our roulette strategy guide shows the same calculation for roulette.

What a session costs

Expected cost per hour = bet × hands per hour × house edge. At $25 a hand and 60 hands an hour:

• Good rules, perfect strategy (0.5%): $7.50 per hour.
• 6:5 table with the same strategy (about 1.9%): $28.50 per hour.
• Good rules but adding a $5 Perfect Pairs side bet (6.1% on the side bet): $7.50 + $18.30 = $25.80 per hour.
• Good rules but playing like the dealer (about 5.5%): $82.50 per hour.

RNG blackjack can deal several hundred hands an hour if you click fast, which multiplies the same edge by the higher pace.

Losing streaks are normal

Because you lose about 49% of hands, streaks come thick and fast. Treating each hand as independent (a close approximation in a shoe game), the chance of losing a run of hands in a row from any given starting point is:

In a 100-hand session, a run of five straight losses is more likely than not, and six in a row is a coin flip. That is exactly why doubling-up systems such as the Martingale fail at blackjack as they do at roulette: the long streak that breaks your bankroll or the table limit is not rare, it is routine. Size your bets so that a ten-hand losing run is survivable without changing your plan.

Bankroll guide for blackjack

These are practical rules of thumb, not guarantees: with a standard deviation of about 1.14 bets per hand, a 100-hand session typically swings around 11 bets either way, and sometimes much more.

How the House Edge Is Built: A Worked Example

To see how 42% wins and 49% losses become a sub-1% edge, take 1,000 hands at $10 each in the six-deck game (0.28% edge). Approximately:

• About 47 or 48 hands are player blackjacks. Most pay $15 each, a few push against a dealer blackjack.
• Roughly one hand in ten is a double, where $20 is at risk rather than $10, and the player wins most of them.
• A few percent of hands are splits, turning one bet into two, mostly in favourable spots (aces, 8s against weak cards, 9s).
• Surrender recovers $5 on the handful of worst hands, such as 16 against a 10 or 15 against a 10.
• The remaining hands are ordinary 1:1 wins, losses and pushes, with the dealer slightly ahead because the player busts first.

Add it all up and the expected net result is about −$28 on $10,000 wagered (the opening bets), the 0.28% house edge. The ordinary hands lose a few percent; blackjack bonuses, doubles, splits and surrender pull almost all of it back. Remove any one of those tools, by playing 6:5, refusing to double, or never splitting, and the edge balloons. The frequencies above are rounded for illustration; the edge itself comes from the Wizard of Odds’ exact calculation.

Blackjack Odds Myths

• “Bad players at the table ruin your odds.” Other players’ decisions change which cards you get in a given hand, but they are as likely to help as to hurt. Over time they do not change your expectation.
• “The dealer is on a hot streak.” Dealers follow fixed rules with no choices. Streaks are variance.
• “Always assume the dealer has a ten in the hole.” A useful shortcut for learning, but only about 30.8% of unseen cards are ten-value. Basic strategy already accounts for every possible hole card.
• “Sitting at third base matters.” Seat position does not change the odds of a correctly played hand.
• “Even money on blackjack is a guaranteed win, so take it.” It is insurance by another name and costs about 7.4% of the insurance portion on average.
• “Online blackjack is rigged against basic strategy.” Licensed games use tested random number generators or real shoes; their published RTP assumes basic strategy. Unlicensed sites give fewer assurances, so play at licensed casinos.

The Cost of Common Mistakes

The insurance figure is an estimate: insurance costs 7.4% of the insurance bet (half a stake), and the dealer shows an ace on about one hand in 13.

Card Counting and the Odds

Blackjack is the only common casino game where past cards change future odds, because cards dealt from a shoe are not replaced until the shuffle. Edward Thorp showed this in his 1962 book Beat the Dealer. The most popular modern method, Hi-Lo, counts 2 to 6 as +1, 7 to 9 as 0, and tens and aces as −1. A high positive “true count” (running count divided by decks remaining) means the shoe is rich in tens and aces, which favours the player through more blackjacks and better doubles.

Which cards help the player

Card counting works because individual cards affect the odds in different directions. Tens and aces left in the shoe help the player: they make blackjacks more likely (which pay 3:2 to you but only 1:1 to the dealer), they make doubles on 10 and 11 stronger, and they make the dealer, who must hit stiff totals, bust more often. Low cards from 2 to 6 help the dealer, who uses them to turn stiff hands into made hands without the option of standing. The 5 is the card whose removal helps the player most, and the 7, 8 and 9 are close to neutral. That is why Hi-Lo counts low cards as +1 when they leave the shoe and high cards as −1.

A widely used rule of thumb is that each +1 of true count moves the player’s expectation by about half a percentage point. In practice:

• Online RNG games reshuffle every hand, so counting is useless.
• Live online tables deal from eight decks and shuffle well before the end of the shoe, which limits any edge, and casinos can restrict accounts.
• Land casinos may ask suspected counters to leave or restrict them to flat betting in many places. In New Jersey, the state Supreme Court ruled in Uston v. Resorts International (1982) that Atlantic City casinos could not bar players simply for counting, though casinos there use other countermeasures such as shuffling more often.

Counting is legal in most places as long as you use only your head; devices that track cards are illegal in many jurisdictions. For nearly every recreational player, the realistic goal is the basic strategy edge of under 0.5%, not a counting advantage.

Rules Explained: Why Small Print Moves the Odds

Dealer hits soft 17 (H17)

A soft 17 is a 17 that contains an ace counted as 11, such as A-6. When the dealer must hit it, the dealer improves to 18 to 21 fairly often and only sometimes busts, so the dealer’s final totals get stronger. The Wizard of Odds measures the cost at 0.22 percentage points. On the felt, look for “Dealer must hit soft 17” (bad) or “Dealer stands on all 17s” (good). Online, the rule is in the help panel. H17 also changes a few basic strategy plays: double 11 against an ace, double soft 18 (A-7) against a 2, double soft 19 (A-8) against a 6, and surrender 15 and 17 against an ace where surrender is offered.

Surrender

Late surrender lets you give up half your bet after the dealer checks for blackjack. It is worth 0.07 points, because a few hands are so bad that losing exactly half is better than playing them out. Hard 16 against a dealer 10 is the classic case: standing is worth −0.541 per unit and hitting −0.535, while surrender is −0.500. In a six-deck S17 game, basic strategy surrenders hard 16 (but not a pair of 8s) against 9, 10 and ace, and hard 15 against a 10. Early surrender, which lets you surrender before the dealer checks, is rare and worth much more: 0.39 points against an ace alone.

No hole card

In the European no-hole-card format the dealer takes the second card only after the players act. If the dealer then makes blackjack, you lose your doubles and splits as well as your original bet, which costs 0.11 points. Some tables use an “original bets only” rule that returns the extra money from doubles and splits in that case, which removes most of the penalty. Basic strategy adjusts by avoiding doubles and splits against a 10 or ace, for example hitting 11 against a 10 and not splitting 8s against an ace.

Doubling restrictions

Doubling is the player’s main weapon. Restricting it to 10 and 11 costs 0.18 points, to 9 to 11 costs 0.09, and banning doubles after splits costs 0.14. A table that allows doubling on any two cards and after splits is worth looking for.

Choosing Between Two Tables: A Worked Example

Imagine two live tables side by side, both at $25 minimum.

• Table A: eight decks, dealer stands on soft 17, double after split, no surrender, blackjack pays 3:2.
• Table B: six decks, dealer hits soft 17, double after split, late surrender, blackjack pays 6:5, plus a $5 “lucky” side-bet prompt on every hand.

Table B looks friendlier on paper: fewer decks, surrender offered. Run the numbers from the rule table. Table A sits at about 0.4% to 0.5%. Table B starts from the same base, gains 0.02 for six decks and 0.07 for surrender, then loses 0.22 for H17 and 1.39 for 6:5, landing near 2%. At 60 hands an hour and $25 a hand, Table A costs roughly $6 to $7.50 an hour and Table B about $30, before any side bet. If you also put $5 on a side bet with a 6% edge every hand, Table B climbs to around $48 an hour. Same game name, same minimum, roughly seven times the price.

The quick method: start at about 0.5%, add 1.4 for 6:5, add 0.2 for H17, subtract a little for surrender and fewer decks. It takes ten seconds and is accurate enough to pick a table.

How Online Blackjack RTP Is Verified

For RNG blackjack, licensed casinos use games whose random number generators are tested by independent laboratories before release, and regulators such as the UK Gambling Commission and the Malta Gaming Authority require game rules and return information to be available to players. Live dealer games are run from studios that deal physical cards from a shoe. Either way, the RTP shown in the rules panel is the theoretical return with optimal play, not a promise about any session. Unlicensed casinos offer no comparable assurance, which is one more reason to check the licence before you deposit.

Blackjack Odds Compared With Other Casino Games

Our roulette payouts and odds chart breaks down every roulette bet, and the keno guide shows why lottery-style games sit at the other end of the scale. For a look at how much blackjack is actually played worldwide, see how popular blackjack is.

Blackjack and Bonus Wagering

Because blackjack’s RTP is so high, casinos protect their bonuses from it. Common terms include blackjack counting at 10% or less toward wagering requirements, some variants counting 0%, and live dealer games excluded entirely. Bovada and Ignition, for example, count standard blackjack at 10%, double deck at 2% and single deck or Classic at 0%, with live dealer games at 0%. If a $100 bonus needs $3,000 of wagering and blackjack counts at 10%, you would need to wager $30,000 on blackjack to clear it. Read the game-weighting table before using bonus money at the tables; our guide to picking casino bonuses and the casino bonus hub explain the maths.

How to Find the Best Blackjack Odds: A Checklist

1. Blackjack pays 3:2. If the layout or rules say 6:5, find another table. This is the single biggest factor.
2. Dealer stands on soft 17. Look for “Dealer must stand on all 17s” printed on the felt or in the rules.
3. Double after split allowed, and doubling on any two cards.
4. Surrender offered, ideally late surrender.
5. Fewer decks are a small bonus, never a reason to accept 6:5.
6. Check the RTP in an online game’s rules panel: around 99.3% to 99.6% is a good standard game.
7. Skip insurance and side bets if you want the RTP you came for.
8. Learn basic strategy. Our casino blackjack strategy guide includes the full chart and how to adjust it for each rule set.

Australian readers can find the local legal position, rule notes for Australian tables and the casinos that accept Australians in our Australian online blackjack guide, plus a deeper chart walkthrough in blackjack basic strategy for Australians. More game maths is in the Casino Bros blog.

Who Blackjack Suits

• Players who want the best value per dollar wagered and are willing to learn a strategy chart.
• Players who enjoy decisions rather than pure chance.
• Not ideal for bonus clearing, because of low or zero game weighting.
• Not ideal for players who want big multiplier wins: the main game rarely pays more than 1.5 times the stake, unless you double or split.

If you are new to the game, the fastest route to the odds on this page is simple: find a 3:2 table where the dealer stands on soft 17, keep a strategy chart beside you (most online casinos allow it, and many land casinos tolerate printed cards at the table), and skip insurance and side bets. Those three habits alone take you from a typical beginner’s 2% to 5% cost down to around half a percent. Learn the doubling and splitting rules next, because those are the plays that turn the game’s 42% win rate into a near-even contest.

Play It Safe

A low house edge is still a house edge, and no strategy guarantees a profit. Decide your budget before you play, stop when it is spent, and never chase losses. Free, confidential help is available if gambling stops being fun: 1-800-MY-RESET in the US, the National Gambling Helpline on 0808 8020 133 in the UK, the Gambling Helpline on 0800 654 655 in New Zealand, and the National Responsible Gambling Programme on 0800 006 008 in South Africa. Our responsible gaming page explains limits and self-exclusion.

Frequently Asked Questions

What is the RTP of blackjack?

With perfect basic strategy at a table with good rules, blackjack’s RTP is about 99.3% to 99.7%. A six-deck game with dealer standing on soft 17, double after split and late surrender returns about 99.72%. Evolution’s standard live blackjack lists 99.28%.

What is the house edge in blackjack?

Roughly 0.3% to 0.7% with basic strategy and 3:2 blackjack. It rises to around 2% at 6:5 tables and about 5.5% for players who simply copy the dealer’s rules.

What are the odds of getting a blackjack?

About 4.75% in a six- or eight-deck game, or once every 21 hands. In a single-deck game it is 4.83%.

What are the odds of winning a hand of blackjack?

With basic strategy in a six-deck game, you win about 42.4% of hands, push 8.5% and lose 49.1%. Blackjack’s 3:2 payout, doubles, splits and surrender make up most of the difference.

Which dealer up card busts most often?

The 6 (about 42.3% in a six-deck S17 game), followed by the 5 (41.8%) and 4 (39.6%). The ace busts least often, about 16.7% of the time once the dealer has checked for blackjack.

Is insurance ever a good bet?

Not for a regular player. Insurance pays 2 to 1 but the dealer has blackjack only about 30.9% of the time with an ace showing in a six-deck game, giving a 7.4% house edge on the bet.

Does 6:5 blackjack really make a big difference?

Yes. It adds 1.39 percentage points to the house edge, roughly tripling or quadrupling the cost of a good 3:2 game. A $10 blackjack pays $12 instead of $15.

Does the number of decks change blackjack odds?

Slightly. Compared with eight decks, a single deck improves the player’s return by 0.48 points, two decks by 0.19 and six decks by only 0.02. Payout and dealer rules matter much more.

Which blackjack game has the best RTP online?

Check each game’s rules panel. Among Evolution’s live games, Lightning Blackjack lists 99.56% and Infinite Blackjack 99.47%, against 99.28% for standard live blackjack, but each figure assumes that game’s own optimal strategy.

Can you beat blackjack in the long run?

Not with basic strategy alone: it reduces the house edge to well under 1% but does not remove it. Only advantage techniques such as card counting in shoe games can produce a player edge, and casinos counter them with frequent shuffles, continuous shufflers, RNG games that reshuffle every hand and, in many places, by restricting players.

Is online blackjack RTP the same as in a land casino?

It depends on the rules, not the format. Online and live games list their RTP in the help panel; standard eight-deck games usually show around 99.3% to 99.5%, similar to good land-based tables.