The moment the dealer offers insurance
You have a decent hand, maybe a nineteen or a twenty. The dealer's upcard is an ace. Before anyone plays another card, the dealer pauses and asks whether you want insurance. The word itself does a lot of work in that moment: it sounds like a safeguard for the hand you already built. It is not. Insurance is a fresh, separate wager that the dealer's face-down card is a ten-value card, giving the dealer a blackjack. Your original hand does not get protected, preserved, or improved by it. You are simply placing a new bet on a specific event, and that new bet carries its own mathematics.
People do take it. Some take it every time the dealer shows an ace, some take it only when they hold a strong hand, and some take it because a tablemate insists it is the smart move. The question worth asking is not whether anyone takes it, but what the bet pays, what it costs, and what the long-run arithmetic says about taking it repeatedly.
What the insurance wager is
When the dealer shows an ace, the house offers you the chance to put up to half your original wager on insurance. If the dealer's hole card is a ten, jack, queen, or king, the dealer has blackjack and the insurance bet pays at 2 to 1. If the hole card is anything else, the insurance bet is collected and the hand continues normally. Your original wager is settled separately, on its own terms, regardless of what happens to the insurance bet.
That separation is the part most players miss. Insurance does not cancel your original bet, does not reduce its loss, and does not turn a losing hand into a push. It is a side bet with its own outcome. You can win the insurance bet and still lose the hand, or lose the insurance bet and still win the hand. The two bets are connected only by the timing of when they are offered and settled.
Because the payout is 2 to 1, the bet feels generous. But a payout is only half of a wager's mathematics. The other half is how often the event you are betting on occurs.
Counting the ten-value cards
A standard deck holds 52 cards. Sixteen of them are ten-value cards: four tens, four jacks, four queens, and four kings. The other 36 are not. That ratio matters because the dealer's hole card is drawn from whatever cards remain after the initial deal.
Suppose, as an example, you are playing a single-deck game and you hold two cards that are not ten-value cards, while the dealer shows an ace. From your position, 49 cards are unknown. Sixteen of them are ten-value, and 33 are not. The chance the hole card is a ten-value card is therefore 16 in 49, a little under one in three. The insurance bet pays 2 to 1, which would be a fair price if the event happened exactly one time in three. It happens slightly less often than that in this example, and the gap between the true chance and the payout is where the house edge on insurance lives.
In a multi-deck game the arithmetic shifts slightly because the proportion of ten-value cards to non-ten-value cards changes as cards are removed from the shoe. The principle stays the same: the payout is fixed at 2 to 1, while the chance of a ten-value hole card depends on the composition of the remaining cards. Most of the time, for a player who is not tracking cards, the chance sits below the one-in-three threshold that the payout implies.
Expected value on a side bet
Expected value describes what a wager returns on average over a very large number of repetitions. It is not a prediction of the next hand. It is the arithmetic of the bet itself.
Build the insurance bet from hypothetical inputs. Suppose you wager 10 units on insurance. If the dealer has blackjack, the bet pays 2 to 1, so you receive 20 units in winnings and keep your 10-unit stake, for a net gain of 20 units. If the dealer does not have blackjack, you lose the 10-unit stake. Now suppose the true chance of the dealer having blackjack is 16 in 49, as in the single-deck example above. The expected value of the bet is:
(16/49 × 20) + (33/49 × −10)
That works out to (320/49) − (330/49), which is −10/49, or about −0.20 units per 10-unit insurance wager. The negative sign means the bet loses money on average. The house edge on this example is roughly 2 percent of the insurance stake. Different deck counts and different visible cards move that number, but the direction stays the same for a player who is not tracking the composition of the remaining cards.
This is the same arithmetic that governs every casino wager. The house edge and the return to player sum to 100 percent for the same bet under the same rule set. If a bet carries a house edge of about 2 percent, its RTP is about 98 percent over the long run. That does not mean you will get 98 units back from a 100-unit session. It means the average across an enormous number of repetitions converges toward that figure, while any individual session can land far above or far below it.
Insurance versus the original hand
One reason insurance persists is that players compare it to the alternative of losing the original bet. If the dealer has blackjack and you did not take insurance, your original wager loses, unless you also have a blackjack. If you did take insurance, the insurance payout offsets part or all of that loss, depending on the size of your insurance stake. That feels like a good trade in the moment.
The comparison is misleading because it only looks at the cases where the dealer has blackjack. The insurance bet is also collected in all the cases where the dealer does not have blackjack, and those cases are more numerous. Over many repetitions, the small losses on non-blackjack outcomes outweigh the occasional 2-to-1 payouts. The bet is not evaluated against the original hand; it is evaluated against its own probability and payout, and on those terms it is a negative-expectation wager for most players.
The table below separates what insurance does from what it does not do.
| What insurance is | What insurance is not |
|---|---|
| A separate side bet on the dealer's hole card | Protection for your original hand |
| Paying 2 to 1 when the dealer has blackjack | A refund of your original wager |
| Settled independently of the main hand | A way to cancel or reduce the main bet |
| Carrying its own house edge | Neutral or free protection |
| Offered only when the dealer shows an ace | Available on every hand |
Why short sessions hide the arithmetic
None of this means you will lose the insurance bet every time you take it. You might take it once and win. You might take it three times and win twice. A 2-to-1 payout on a bet that hits a little under a third of the time produces streaks in both directions, and a short session is far too small a sample to reveal the underlying edge.
This is the gap between variance and expected value. Variance describes how outcomes scatter in the short run. Expected value describes where the average settles over the long run. A negative-expectation bet can win for an evening, a weekend, or a month. That does not mean the edge has disappeared. It means the sample is small enough that randomness dominates the result.
The same dynamic applies to the main blackjack hand. A player who wins a session after taking insurance on every dealer ace has not found a flaw in the bet. They have experienced the short-run scatter that all negative-expectation wagers produce. The arithmetic of the bet is unchanged by the result of any single session.
Decisions per hour and total exposure
Blackjack hands move quickly. A full table might deal 60 to 80 hands per hour; a heads-up game can move considerably faster, depending on the dealer and the number of players. Every dealer ace creates a new insurance decision, and every insurance decision is a new wager with its own house edge.
The cost of a side bet is not just its edge. It is the edge multiplied by how often you make the bet and how much you stake on it. A player who takes insurance on every dealer ace is adding a steady stream of negative-expectation wagers to their session. A player who never takes insurance avoids that stream entirely. The difference in expected loss between the two players is not a matter of skill at reading the dealer; it is a matter of how many times each one voluntarily places a bet that loses money on average.
This is also why the size of the insurance stake matters. The house allows up to half the original wager, but the player chooses the amount. A smaller insurance stake reduces the total exposure to the side bet's edge, while a larger stake increases it. The choice is not between good and bad protection. It is between more and less exposure to a negative-expectation wager.
When the arithmetic changes
There is one situation where insurance can become a positive-expectation bet: when the composition of the remaining cards is known well enough to conclude that ten-value cards make up more than one-third of the unseen cards. That requires tracking the cards that have already been played, and it is not something a casual player can do reliably while also following the flow of the game. Card counting is a separate discipline with its own methods, and most players at a blackjack table are not doing it.
For everyone else, the insurance bet is a wager whose payout does not quite cover its true probability. The gap is small on any single bet, but it is real, and it compounds across every repetition. The dealer's offer sounds protective because the word insurance carries a meaning from everyday life. In blackjack, the word is attached to a side bet with its own arithmetic, and that arithmetic does not favour the player who takes it routinely.
If you want to know the exact house edge on insurance at the table where you are playing, the relevant inputs are the number of decks in use and the cards you can see. The game's help screen or the table rules will state the deck count. From there, the arithmetic is a matter of counting ten-value cards among the unseen cards and comparing that proportion to the 2-to-1 payout. The bet itself does not advertise its edge; the player has to build it from the composition of the shoe.
FAQ
Is insurance ever a good bet for a casual player?
For a player who is not tracking the composition of the remaining cards, insurance is a negative-expectation wager. The payout is fixed at 2 to 1, while the chance of a ten-value hole card is usually below one in three. Casual players are generally better off declining the bet and accepting that the dealer will sometimes have blackjack.
What is the difference between insurance and even money?
When a player has a blackjack and the dealer shows an ace, the dealer may offer even money, which pays the original bet at 1 to 1 immediately instead of risking a push. Even money is mathematically the same as taking insurance on a blackjack hand. The player is trading the chance of a 3-to-2 payout for a guaranteed smaller payout, and the same arithmetic applies.
Does taking insurance reduce the house edge on the main hand?
No. The main hand and the insurance bet are separate wagers with separate outcomes. Insurance neither improves the main hand's chance of winning nor reduces its expected loss. It adds a new bet with its own house edge, so taking it increases total expected loss rather than reducing it.
Gambling involves risk. Please only gamble with funds that you can comfortably afford to lose. While we do our best to provide good quality information, we cannot be held responsible for any loss you may incur from gambling. We do not bet on our own games, and we do not provide gambling advice. If you or someone you know is struggling with gambling, help is available: call or text the National Problem Gambling Helpline at the helpline printed in the footer below.
Play responsibly
Gambling should be treated as paid entertainment, never as a way to earn income or recover losses. 1-800-MY-RESET (1-800-697-3738)
18+ or 21+ depending on where you are; follow the minimum age that applies to you.
This article is general information about how these mechanics work. It is not legal advice and not a recommendation to gamble or to use any particular operator. Availability and legality differ by jurisdiction — check the rules that apply where you are.