Blackjack insurance

The Wizard often says in his Blackjack articles that insurance is a sucker bet, and you should never take it.

However, I've heard of card counters being able to make smart insurance bets under certain circumstances. Is that possible? Are there any circumstances where it becomes advantageous to take insurance? If the deck is particularly tens-heavy, and the chance of dealer BJ is high?

If so, what initial 2-card combos should it be taken on?


What is the formula for calculating this?

Suppose that you've counted the cards, and know exactly how many how many cards, and how many tens are left, thus you can give precisely the chance of drawing a ten, thus getting blackjack.


I've been thinking about it, and come up with the formula below. I'm not sure if it's right, though.

C10 = chance of drawing a ten

BSEV = the EV for the initial 2 cards, using the optimum move u8nder basic strategy, assuming the dealer has checked for BJ, and doesn't have it.


EV for taking insurance = ((2*C10) +(1-C10)*BSEV))-1
EV without insurance = (1-C10)*BSEV

Is my maths correct? And will there ever be any hands where it is worth taking insurance?

I will let the Wiz defend himself, if he so decides. As for questioning your math, I am computer illiterate. In the Wiz's defense he has stated in this forum and other that 30% of a counter's profit come from insurance bets. I believe he stated the original source of that statement was some blackjack novice by the name of Stanford Wong.

Yep. Insurance is a correct play for the competent card counter, depending on the count. It doesn't matter what 2-card hand you have.

The basic strategy player should NEVER take insurance.

" The basic strategy player should NEVER take insurance. "

But it's a winner. Everybody ,including the dealer and John Patrick, will laugh at me if I turn down a guaranteed winner and the dealer has BJ.

Buzz -

Dan was inferring that a 'Basic strategy player', i.e. a non-counter, should never take insurance.

Played properly and under certain count conditions, a card counter should deviate from basic strategy in other areas as well.

Just a little satire on my part. Always amazed at players who say " Insurance is a sucker bet : But they always take even money.
As I stated before Insurance bets are 30% of a counter's winning.

I am math challenged, but it makes sense to me. Deck is rich in tens, counter using 8 to 1 spread, maybe getting away with betting 2 spots with good counts, and dealer turns up an ACE. But instead of losing 16 units when dealer has BJ, counter gets instant 6 to 8 %
return on 8 units.

Average player has no idea about insurance. When I dealt, over 20 years ago, after giving a player change ,who wanted to insure
just one hand of his two hands, I gave him a 2.50 chip and place the other chip next to the hand in front of him. He got a panic look as he screamed " Not that hand . Not that one. I want to insure the other hand. "

Like it makes a difference. DUHHHHHH

Quote: DanMahowny

Yep. Insurance is a correct play for the competent card counter, depending on the count. It doesn't matter what 2-card hand you have.

The basic strategy player should NEVER take insurance.

This is correct. Stated more directly as a response to the OP - there is NO initial 2-card hand which makes insurance a profitable bet if you do not have information about the previous hands dealt (i.e., the count). If your hand is two non-ten cards, insurance is a slightly less-bad bet than if your hand is two ten cards. But without knowledge of the remaining cards (from card counting) or knowledge of the hole card (from...hole carding), you should never take insurance.

Quote: buzzpaff

When I dealt, over 20 years ago, after giving a player change ,who wanted to insure
just one hand of his two hands, I gave him a 2.50 chip and place the other chip next to the hand in front of him. He got a panic look as he screamed " Not that hand . Not that one. I want to insure the other hand. "

This is classic, nice work buzz.

Insurance has nothing to do with the total of your hand. It is merely a side bet on whether the dealer has a 10 in the hole. I'll insure any total if the count calls for it. Basic strategy players should listen to Buzz and never buy insurance.

My insurance correlation rides 80% correct the past three visits. Before that it was about 55-60%. Of course you must always allow for the law of large numbers so those in the wise know that those numbers fluctuate between a certain percent. One night I went 8 for 8. It was like those dealer blackjacks never happened. I was gitty as a school boy. Bad thing is that it was at one casino. :( I know right. Shame on me.

As others have stated, Insurance is simply at side bet, at 2 to 1 odds, on whether the dealer has a 10 in the hole. It does not have anything to do with the two cards in your hand, nor their total.

If you're keeping track of the number of 10s and non 10s left in the deck, the formula you are looking for is easy. Divide the number of non 10s by the number of 10s. If the answer is less than 2, taking Insurance is a good bet. If it's more than 2, it's a bad bet.

Example:

There are 23 Non 10s and 11 10s left in the deck. 23/11 = 2.09 Don't take Insurance.
There are 28 Non 10s and 16 10s left in the deck. 28/16 = 1.75 Take Insurance.