When should you use Insurance in Blackjack?

Let's talk about a major point of contention in the game of blackjack. As you're aware, there is an option called "insurance" which allows you to cover yourself with a hand, whether it is good or bad, by, in effect, betting that the dealer has a blackjack when he or she has an Ace showing. The insurance bet is half of the wager you have out on the table, and pays off at 2-to-1 odds, so you have the possibility of getting your bet back if you have guessed right and the dealer turns over a ten-value card to complete a two-card natural.

What is the recommendation for the Basic Blackjack Strategy player when it comes to insuring a hand? Well, to simplify it, don't do it, because it doesn't make mathematical sense.

But, you may ask, what about when the player has a blackjack, a hand that pays off at 3-to-2 odds? Won't the player always wind up with a profit? The answer to that is yes, there will be a profit in that situation.

Yet you should still, as a Basic Strategy player, never insure, and we will illustrate it.

Let's grab a 52-card deck. Its contents include sixteen (16) ten-value cards and 36 non-tens, which constitutes a ratio of 2.25-to-one. We're going to look at this in its most elementary form. When you are faced with the situation in which you have a two-card blackjack and the dealer has an Ace showing, there are now 15 tens and 24 non-tens remaining. That's a ratio of 2.27-to-one. Remember that insurance pays only two-to-one, so on a percentage basis you're getting the worst of it on that basis alone.

Let's imagine you have a $10 bet and each of the following situations:

1) You take insurance and the dealer has blackjack

2) You take insurance and the dealer does not have blackjack;

3) You do not take insurance and the dealer has blackjack;

4) You do not take insurance and the dealer does not have blackjack.


In (1) and (2), which as stated are plays we do not recommend, the return to you is $10. If you bet $10 and insure for $5, you're going to get a "push" on the primary bet and win $10 on the insurance bet. If the dealer doesn't have blackjack, you're going to win $15 from the primary bet, and lose the $5 insurance bet, which will leave you with a $10 profit.

In (3), you have not thrown down an insurance bet, and tied the dealer with your blackjack. You will retain the $10 bet, producing net gain of zero. In (4), you beat the dealer with your blackjack, with a gain of $15 on the hand.

If you notice, when you take insurance, you are guaranteed a return on that betting scenario, while in one of these four cases, you get nothing.

Now let's talk about probabilities: remember, you've got 15 tens and 34 non-tens left, which makes a theoretical total of 49 cards remaining. Since with an Ace showing the dealer has to have a ten in the hole to complete a two-card blackjack, what are the chances of that happening? Well, just divide 15 by 49, and you come up with a 30.6 percent chance of the dealer having blackjack, which means it's a 69.4 percent chance of the dealer NOT having blackjack. Scenarios #2 and #4 are going to happen a lot more than the other two.

Here is your expected gain per 100 situations for each of the four scenarios:

1) You take insurance and the dealer has blackjack = $10 (30.6%)

2) You take insurance and the dealer does not have blackjack = $10 (69.4%)

3) You do not take insurance and the dealer has blackjack = $0 (30.6%)

4) You do not take insurance and the dealer does not have blackjack = $15 (69.4%)


From there you can just multiply. Scenario (1) gives you an expected return of $306 for every 100 situations; Scenario (2) yields $694 for that same sampling. That's an even $1000.

Scenario (3), in which there is no insurance, produces an expected return of $0, since, after all, there was no protection for the blackjack. But in (4), where the dealer doesn't have a blackjack, you're going to win $15 per play, or $1041 over the course of 100 of these situations ($15 x .694 x 100).

The gain reflected by the difference between the results when you take insurance (Scenarios #1 and #2 added) and don't take it (Scenarios #3 and #4 added) is $41 for every 100 situations (or $1000 wagered).

Your most profitable situation is the one which occurs the vast majority of the time. As this example illustrates, your net gain is 4.1% for every $1000 wagered by not insuring the hand. Unless you're counting cards and can make a more precise evaluation as to the probability of a ten-value card in the dealer's hole, you are better off ignoring the insurance option in your blackjack game.