Calculate the expected value of taking insurance in blackjack. Learn when insurance is profitable for card counters and why basic strategy players should always decline.
EV = P(BJ) × Win - P(No BJ) × LossInsurance EV
-$4
-7.40% of bet
Dealer BJ Probability
30.87%
Break-even: 33.3%
Insurance bet: $50
Affects baseline probability
Standard: 96 for 6 deck(s)
Standard: 312 for 6 deck(s)
Without card counting knowledge, insurance has a 7.40% house edge in a 6-deck game. This is one of the worst bets in the casino.
Insurance in blackjack is almost always a bad bet. With no card counting knowledge, insurance has a house edge of 7.4% (6-deck). On a $100 main bet, insurance costs $50 with EV of -$3.7. Only take insurance when the true count is +3 or higher in Hi-Lo counting.
No, for basic strategy players insurance is always a bad bet. It has a 7.4% house edge in a 6-deck game. On a $100 bet, you bet $50 on insurance and expect to lose $3.70. The only exception is card counters who know when the deck is rich in tens.
Insurance pays 2:1 if the dealer has blackjack, but the probability is only about 30.8% (6-deck). You need 33.3% probability to break even. EV = (0.308 × $100) - (0.692 × $50) = $30.80 - $34.60 = -$3.80 per $50 insurance bet.
In Hi-Lo counting, take insurance when the true count is +3 or higher. At TC +3, there are enough tens remaining that the probability of dealer blackjack exceeds the 33.3% break-even point. Some systems use +2.5 as the index.
"Even money" when you have blackjack vs dealer Ace is mathematically identical to insurance. You're giving up the potential for a 3:2 payout ($150 on $100) for a guaranteed 1:1 ($100). Expected value of not taking even money: 0.692 × $150 = $103.80 - always better than $100 guaranteed.
Quick-start with common scenarios
Insurance EV
-$4
-7.40% of bet
Dealer BJ Probability
30.87%
Break-even: 33.3%