Free Bet Blackjack: The Free Doubles Are Not Free

Updated

The casino puts up the money when you double a hard 9, 10 or 11, and again when you split a pair. You keep the winnings and never risk the extra chip. In exchange, the dealer no longer busts on 22 — it pushes. We computed all three rules separately, and the trade is not close.

The short version

What each rule is worth

Start from an ordinary six-deck game where the dealer hits soft 17. We compute that at 0.62%. Then we switched each Free Bet rule on by itself.

RuleEffectGood or bad for you
Dealer busting with 22 is a push+6.95%bad, and by a mile
Free double on hard 9, 10, 11−4.19%good
Free split on any pair but tens−1.94%good
Dealer pushes on 22 −6.95% Free doubles +4.19% Free splits +1.94%
The red bar is longer than the two green ones combined. That is the entire business model of the game.

Those three do not simply add up, and the reason is worth knowing: a pair of fives is a hand you would rather free-double than free-split, and a pair of aces is the reverse. The gifts overlap, so you cannot price the game by adding rule effects from a table — they have to be measured together. That is why our full-game figure is computed in one pass rather than assembled from the rows above.

Why they picked 22

Twenty-two is not an arbitrary choice. It is the single most common total a busted dealer holds, because it is what every stiff hand — 12 through 16 — turns into when the next card is a ten, and ten-value cards are nearly a third of the deck.

So the rule is precision-engineered. It does not soften the dealer's bust rate across the board; it removes the most frequent slice of it and leaves the rest. Our figures for how often the dealer breaks on each total are in how often the dealer breaks.

There is one thing the rule does not touch: a player blackjack still wins, even against a dealer 22. That is the only reason the game is playable at all.

The rule most write-ups leave out

Free Bet lets you re-split up to four hands, including aces — and every one of those extra hands is on the casino. In ordinary blackjack, going from one split to four hands is worth about five hundredths of a percent. Here it is worth eight times that, because you are being handed free bets rather than putting up your own money.

We know the size of it because we got it wrong first. Our model originally split a pair once and stopped there. Here is what each step was worth:

Split once: 1.43%. Re-split to four hands: 1.09%. Re-split aces as well: 1.02%. The published figure is 1.04%. Two rules that are worth five hundredths of a percent in ordinary blackjack are worth four tenths here, because the extra hands are not your money.

With re-splitting included we compute 1.0165% against the published 1.04%. The remaining twenty-four thousandths run the same way in every variant we model fully: published figures come from simulations dealt from a shoe with a cut card and played from a strategy chart, while our engine works out the exact composition and plays the best move available. The cut card alone is worth 0.020% on Wizard of Odds’ own measurement.

Is it a good game?

No. At around 1% it is roughly twice the house edge of a plain six-deck table with a 3:2 payout, and the free doubles do the psychological work of making that invisible: you remember the hand where the house paid for your double on eleven, and you do not remember the four times a dealer 22 quietly took your twenty away.

For a counter it is worse than the headline suggests. The push-22 rule caps what a rich shoe can do for you — a deck full of tens breaks the dealer more often, and this game has taken the most valuable part of that away. If you want a variant that rewards a good shoe, this is not it. Compare with what each ordinary rule costs before you sit down.

The live-dealer version is a different game

Everything above is the casino-floor game: six decks, dealer hits soft 17, and re-splitting to four hands including aces. Evolution runs a Free Bet table in its live studio, and it does not have those rules. Live tables allow one split only, as they do at every studio.

That is not a detail. By the decomposition above, going from one split to four free hands was worth about four tenths of a percent — the largest single item on this page after the push-22 rule. Take it away and the game gets meaningfully worse, and it was not a good game to start with.

No operator or database publishes a house edge for the live version. The main live-dealer rules database lists the table and leaves both the RTP and the house edge blank, because its model does not handle free bets. So we will not print a figure for it either: our 1.02% is the floor game, and applying it to the live table would understate what that table costs by an amount we have not measured.

Keep reading

Variants comparedall of them, side by side Blackjack Switchthe other push-22 game Double after splitthe ordinary version of this

By Martin Kosek · every number on this site is computed by us, not copied.

← Back to guides