I know mathematically it's the right thing to do, but when you take insurance and the dealer doesn't have BJ, what do you do with a stiff (12 - 16)? You have your max bet out against an Ace, and you've already lost 1/2 of that on the insurance bet. Hitting seems to be the only option, but the deck is still rich in high cards that can bust you out. Any insight into this situation? Thanks.
Jay
> I know mathematically it's the right thing to do, but
> when you take insurance and the dealer doesn't have
> BJ, what do you do with a stiff (12 - 16)? You have
> your max bet out against an Ace, and you've already
> lost 1/2 of that on the insurance bet.
You have not lost half "of that" on anything. The insurance bet is a side bet, completely separate from your main bet. You're not actually "insuring" anything at all. You are making a side bet on whether or not the dealer has a 10 value holecard. If you lose that bet, you then play out your hand as usual.
> Hitting seems
> to be the only option, but the deck is still rich in
> high cards that can bust you out. Any insight into
> this situation? Thanks.
The deck may be rich in high cards, but the dealer is unlikely to bust with an ace upcard. Remember, when you have a stiff vs. an ace, you will lose more often than not regardless of how you play it - which has nothing to do with the insurance bet.
> You have not lost half "of that" on
> anything. The insurance bet is a side bet, completely
> separate from your main bet. You're not actually
> "insuring" anything at all. You are making a
> side bet on whether or not the dealer has a 10 value
> holecard. If you lose that bet, you then play out your
> hand as usual.
When you say that "you're not actually 'insuring' anything at all," this isn't correct; even though a lot of gambling and blackjack books say the same thing. You ARE "insuring" your hand. When you buy insurance, you are, in a sense, taking out an insurance policy against a specific type of loss, i.e., the dealer having a 10-value card under his ace whereby you'd loose your initial wager. An analogy would be insuring you home for flood. If your home is lost to a flood, your loss (the value of your home would be the amount of your initial wager) is covered. If there is no flood you could still loose your home to a fire (which is analogous to losing your hand to the dealer's ace-up even though he didn't have a blackjack). Flood insurance (taking insurance at blackjack) has nothing to do with fire insurance (losing your initial wager to a dealer's ace-up non-blackjack).
The popular (mis)conception is that the insurance bet is related to your main hand. If you have a strong hand (such as a pair of faces), you might consider taking insurance, but never on a hand such as a hard 16, since you?re probably going to lose that hand even if the dealer does not have blackjack. This is wrong.
The insurance bet is actually a side bet on whether or not the dealer has a 10 hole card. We base our decision whether or not to ?insure? on the count, not on the composition of our hand.
The ?insurance? bet has nothing to do with the main bet. You will win or lose your main bet the same, regardless of whether or not you take insurance. The fact that winning the insurance bet and losing your main bet effectively turns the net result of the hand into a push is no more relevant that the fact that if you win a hand and lose the following hand (same bet out), the net result is a push.
As it turns out, there is a tenuous relationship between the insurance bet and the main bet having to do with the composition of the main hand. This is discussed in a post by Cacarulo in the archives of Don?s Domain. This is advanced stuff, and not a topic for the Beginner?s page.
Again, the insurance bet is a side bet which is unrelated to the primary hand/bet.
> The popular (mis)conception is that the insurance bet
> is related to your main hand. If you have a strong
> hand (such as a pair of faces), you might consider
> taking insurance, but never on a hand such as a hard
> 16, since you?re probably going to lose that hand even
> if the dealer does not have blackjack. This is wrong.
> The insurance bet is actually a side bet on whether or
> not the dealer has a 10 hole card. We base our
> decision whether or not to ?insure? on the count, not
> on the composition of our hand.
> The ?insurance? bet has nothing to do with the main
> bet. You will win or lose your main bet the same,
> regardless of whether or not you take insurance. The
> fact that winning the insurance bet and losing your
> main bet effectively turns the net result of the hand
> into a push is no more relevant that the fact that if
> you win a hand and lose the following hand (same bet
> out), the net result is a push.
> As it turns out, there is a tenuous relationship
> between the insurance bet and the main bet having to
> do with the composition of the main hand. This is
> discussed in a post by Cacarulo in the archives of
> Don?s Domain. This is advanced stuff, and not a topic
> for the Beginner?s page.
> Again, the insurance bet is a side bet which is
> unrelated to the primary hand/bet.
If you take insurance you are insuring against a dealer's blackjack--another way of saying you are making a side bet that the dealer has a ten under. They are the same thing. If the dealer does have a blackjack then the hand IS over and the player has "insured" against the loss of his original bet successfully. If the dealer does not have a 10 under (a blackjack), then the "premium" (the insurance bet) is lost and the play resumes. You can still loose the hand to another "peril" (busting or being beat).
Yes, it makes no difference that the player is insuring a 20 or a 16; the player is really insuring against the loss of his original wager against the "peril" of a dealer blackjack. One can state that one is really making a side bet as to wheather the dealer has a 10 under, and it is important to realize this; especially when considering that the relationship of 10s to non-tens remaining in the deck/s is what should guide one's decision as to wheather or not to take insurance. But the fact that you are "insuring" is a correct analogy.
> The fact that you are "insuring" is a correct analogy.
>snip: John, think of these questions. If you had no bet on any hand at all, would you ever want to make an Insurance bet if they'd let you? What would be your criteria? What would you be insuring if you did?
It might be best to call Insurance a "hedge bet". Like most hedge bets, it produces a separate result that typically runs opposite to some current risk. And -- like other hedge bets it's a bad bet unless it carries its own positive expectation.
> It might be best to call Insurance a "hedge
> bet". Like most hedge bets, it produces a
> separate result that typically runs opposite to some
> current risk. And -- like other hedge bets it's a bad
> bet unless it carries its own positive expectation.
I agree that falling it a ?hedge bet? would be more accurate. However, I don?t think that we?ll see ?Hedge Bet Pays 2:1? stenciled on blackjack table felts any time soon. :-)
> It might be best to call Insurance a "hedge
> bet". Like most hedge bets, it produces a
> separate result that typically runs opposite to some
> current risk. And -- like other hedge bets it's a bad
> bet unless it carries its own positive expectation.
It's good to know that the insurance bet can and does carry its own positive expectation, at least on a long-term basis, if made when the ratio of 10s to non-tens remaining in the deck/s is greater than 33.33%. One could say that the insurance bet doesn't insure anything because a player will win or lose his original bet anyway. But this is exactly what happens with insurance. If you have a flood insurance policy and you lose your house (your original wager) to a flood, the insurance proceeds restore your home to its original condition. That the insurance bet is called "insurance" adds flavor to the game. The dealer has to utter only one word at the proper time to entice the players to insure against the peril of a dealer blackjack; and the insurance wager hooks in the casual hunch-playing ploppy with an approximately 6-8% house vig. Of course, if you reduce the insurance bet to its elemental propostion you are making a side-bet as to whether the dealer has a 10-card under his ace. But if you tell the players that, they might do the math.
Greasy John
It's good to know that the insurance bet can and does carry its own positive expectation, at least on a long-term basis, if made when the ratio of 10s to non-tens remaining in the deck/s is greater than 33.33%.
Greasy John,
Actually, the advantage occurs when the ratio of 10's to all remaining cards exceeds 33.33%. This corresponds to a 10's to non-10's ratio larger than 50%.
Dog Hand
> It's good to know that the insurance bet can and does
> carry its own positive expectation, at least on a
> long-term basis, if made when the ratio of 10s to
> non-tens remaining in the deck/s is greater than
> 33.33%. Greasy John,
> Actually, the advantage occurs when the ratio of 10's
> to all remaining cards exceeds 33.33%. This
> corresponds to a 10's to non-10's ratio larger than
> 50%.
> Dog Hand
and I'll bet few readers caught that. Look at my post a few down and you'll find a mistake that yells. $9,999?
Greasy John
Yours is an excellent post, as is Fred's, but the fact that insurance is, overall, a negative-expectation play doesn't mean that, for some people, under some circumstances, it isn't the "right" thing to do. Defining "right" sometimes goes beyond sheer e.v. values and involves the more subtle concept of utility theory.
Suppose, for example, that you're a tourist who plays BJ once a year. On this trip, you've got your heart set on trying to win $1,000 to buy a new fancy TV. that's all you really want or hope for. You've played them even for four days, and now, as your plane is about to leave, you realize that you have your original stake, but no winnings. So, you decide to make one "hail-Mary" bet for the TV. You plunk down $1,000 and, lo and behold, you get a snapper. Jubilant, you look at the dealer's upcard: the dreaded ace. What to do?? Under these circumstances, e.v. be damned, do you understand why it wouldn't be "irrational" or "wrong" to take even money?
Do you think the above applies only to the "other guy," and not to you sophisticated, intelligent advantage players? Well, then, answer this -- as honestly as you possibly can:
You've just won a promotional contest where you get to play one hand of BJ for ... a million dollars! (And yes, insuring a natural, or calling "even money" is allowed.) Single-deck, dealt right off the top. You get a natural, and the dealer shows an ace. Now, I ask ALL of you: Are you going to tell me, to a man (or woman!), that not one of you would take the million?! :-) You don't expect me to believe that, do you? :-)
Food for thought, and the concept of utility vs. e.v.
Don
Are you going to tell me that not one of you would take
> the million?!
>snip: Don, your example is why I give those bookmakers(actuaries) at the insurance company their house edge and insure my home against a complete loss -- negative EV that it is. Yet, I do not carry collision insurance on either of my cars, since there, I could take a complete loss in stride if it should come to that. As you say, at some point EV becomes secondary to the gravity of the risk.
In the case of the $1000 TV? If the man had $1000 to bet, I suspect his prospect of losing the TV would be outweighed by his nearly 70% chance to buy another $500 worth of stereo equipment along with the TV when he decines even money.
Several vivid illustrations of this same principle were on television a few years ago when "Who Wants to be a Millionaire?" was so popular.
All are examples of why we bet within our means at blackjack, so as not to be forced into worsening our overall result with negative EV propositions.
> Yours is an excellent post, as is Fred's, but the fact
> that insurance is, overall, a negative-expectation
> play doesn't mean that, for some people, under some
> circumstances, it isn't the "right" thing to
> do. Defining "right" sometimes goes beyond
> sheer e.v. values and involves the more subtle concept
> of utility theory.
> Suppose, for example, that you're a tourist who plays
> BJ once a year. On this trip, you've got your heart
> set on trying to win $1,000 to buy a new fancy TV.
> that's all you really want or hope for. You've played
> them even for four days, and now, as your plane is
> about to leave, you realize that you have your
> original stake, but no winnings. So, you decide to
> make one "hail-Mary" bet for the TV. You
> plunk down $1,000 and, lo and behold, you get a
> snapper. Jubilant, you look at the dealer's upcard:
> the dreaded ace. What to do?? Under these
> circumstances, e.v. be damned, do you understand why
> it wouldn't be "irrational" or
> "wrong" to take even money?
> Do you think the above applies only to the "other
> guy," and not to you sophisticated, intelligent
> advantage players? Well, then, answer this -- as
> honestly as you possibly can:
> You've just won a promotional contest where you get to
> play one hand of BJ for ... a million dollars! (And
> yes, insuring a natural, or calling "even
> money" is allowed.) Single-deck, dealt right off
> the top. You get a natural, and the dealer shows an
> ace. Now, I ask ALL of you: Are you going to tell me,
> to a man (or woman!), that not one of you would take
> the million?! :-) You don't expect me to believe that,
> do you? :-)
> Food for thought, and the concept of utility vs. e.v.
> Don
Don,
It's a small world, for I too have often thought of the exact situation you express. And If I "had" to make a $1,000,000 bet at blackjack, I'd take even money, even though at SD one would win $9,999 more per play by NOT taking even money. (I think I did the math right.)
Greasy John
Posting Permissions
-*-
Bookmarks