front page
blackjack
poker
secrets
archive
video
gm store
resources
advertising
contact us

Back to The Blackjack Page
The Blackjack Page Archive
The Insurance Bet

 

To a non-counter, the insurance bet is a "sucker" bet, but to a counter, it's the most important Basic Strategy variation. How can a side-bet that has nothing whatsoever to do with your Blackjack hand attain this status? And, while I'm asking questions, why does a counter have to be careful with this bet in order to keep the welcome mat out at his or her favorite casino? Now that I've posed the questions, I'll provide the answers. (Kind of like a lawyer: Never ask a question you don't already know the answer to.)

Even the dealer and the "pit critters" (as we lovingly call the casino game supervisors) will tell you that you shouldn't take insurance, except in one situation that I'll discuss in a bit. Of course, they think they're giving this sage advice to the average gambler but, if you're a counter, it really doesn't apply to you although you should nod knowingly when they bring it up in conversation, especially when you're placing an insurance bet they believe you shouldn't be making. At that point, the PC will think you're either really stupid or simply cannot hear him. Or, they may think you know something they don't know, especially if the bet pays off. And now the "cat and mouse" game begins.

That's the problem with the insurance bet: it either makes you look very dumb or very smart and, as a counter, we much prefer the former. But, not everyone counts cards, so let's look at this both ways.

Insurance and the Non-Counter

Just about every book you'll ever read about Blackjack - be it a good book or a bad book - will tell you to never take insurance. For the most part I agree with them. But, as I've written in the past, there's one situation that will occasionally occur where taking insurance may be worthwhile. (Actually, you're not "taking" insurance, but are making an insurance bet. I treat both as the same; making an insurance bet is taking insurance.) If you have received a "natural" (a 'blackjack') and the dealer is showing an Ace, you'll have the opportunity to make an insurance bet, which can be any amount up to half of your original bet. So, if you have a $10 bet placed, you could make a $5 insurance bet and one of two things can happen: The dealer will also have a natural, so you'll "push" the hand, but win $10 for your insurance bet, but if the dealer does not have a natural, you'll lose the $5 insurance bet and win $15 on your $10 natural. It's easy to see that in either case your net win is $10, which is "even-money" on your original bet. In actuality, most dealers will ask you if you'd like even-money so that you don't have to place the insurance bet and go through all the rigamarole I've described.

This so-called even-money proposition, most casino personnel will tell you, is the only "sure thing" in the casino; you can't lose. What they won't tell you is what you're giving up in return for this sure thing. The odds or probabilities of insurance bets are easy to calculate because all that matters is whether or not the dealer has a 10 or face card "in the hole." Because 16 cards in a single deck are 10s and you already have a 10 in your hand, it's obvious that the odds of the dealer who is showing an Ace actually having a ten-valued card is 15 of 52 or 28.84%. The casino's edge on this bet comes about from the fact that they'll be paying you only 2 to 1 for what is basically a 34 (non-10s remaining) to 15 (10s remaining) proposition. To avoid over-complicating this calculation, just let it suffice to say that the casino's edge on the insurance bet is 8%. Consequently, if you take even-money, you're giving up 8% of what your insurance bet would be and, presuming you'd bet half of your original bet (you may, at most casinos, insure for less), the net cost is 4% of your original bet.

In spite of that very big house edge, I still recommend that non-counters who find they have a bet that is some multiple of their minimum bet out there when this happens (say a $20 bet at a $5 table) to go ahead and "lock up" a guaranteed profit. My primary reason for this is, if you're not counting, the casino has a long-term advantage over you, regardless of how you play and/or bet, but this win may give you some additional ammunition, if only for a little while.

However, if you are counting, most "experts" will tell you to never take even-money unless the count warrants it. Are they correct? Let's find out.

Insurance and the Counter

The card counter has a distinct advantage at insurance bets because s/he is, in essence, tracking the 10s in a deck, albeit in a less-than-ideal manner. If one is using the Hi/Lo count that we teach here, Aces get lumped in with the 10-valued cards, so we aren't measuring the ratio of 10s to non-10s, which is needed to make "perfect" insurance decisions. But, we can get pretty close by using the True Count (TC) to tell us if making the insurance bet is wise. For a single-deck game, the TC should be 1.4 or higher and for all multi-deck games, I usually recommended taking insurance at a TC of 3 or more, just to keep things simple. Unfortunately, you won't win every insurance bet that you make, because the Hi/Lo count has an "insurance efficiency" of only 80%. What this basically means is that you'll be correct only 4 times out of 5 when you make an insurance bet at the stated True Count or higher.

But being right or wrong is just a part of the insurance dilemma. The other part is that you'll often be making the insurance bet when you have a lousy hand, like 14 against a 10 and that's not something most gamblers do. Sure, many will insure a 20; at least that makes sense to most people, but a 14? Seeing a player that insures any hand with a larger-than-normal bet on it starts sending signals to the casino personnel and that is not a good thing, because the signal says "counter". And yet, it's very important to make the insurance bet because it is, as I mentioned earlier, the single most important Basic Strategy variation, even exceeding the famous hit/stand with 16 versus a 10. So, can we camouflage our use of this bet and still get the majority of the return it offers? Perhaps.

First of all, we can always accept even-money when the True Count says to insure and we, ourselves, hold a natural. That move will certainly make us look like the average gambler, but it's actually a fairly rare event. On average, you'll be dealt a natural about once every 21 hands in a multi-deck game and the dealer will have an Ace up 1/13th of the time (actually a little less than that because you already have an Ace in your hand). Multiply the two together and the probability of getting a natural versus a dealer's Ace is about once every 273 hands. If you're playing at tables with less than three other players plus yourself and you're getting a rate of 125 hands per hour, this is going to happen only about once every 2-3 hours.

Far more often, you'll have a hand of 12 to 16 versus an Ace; about 1 hand in 44, so that will happen three or four times an hour. Now understand that the count won't be high enough to warrant making an insurance bet on all of those hands, but it'll happen often enough and it just looks so weird that we need to come up with some sort of strategy to handle the situation. It's also safe to assume that you'll have one of your bigger bets out there when this happens (because the count will likely be in plus territory), so not taking insurance on the hand can cost you some serious $$$. What's a counter to do?

First of all, do not try to fool the dealer with some sort of lame excuse ("Gee, I better insure because you've been hot tonight.") because it's not the dealer who's going to be your problem. And, if you have less than $50 bet, you're not going to attract a lot of attention anyway, unless you're playing in some little hole-in-the-wall casino in rural Nevada. The real problem will be either one of the pit critters or the eye-in-the-sky, so verbal camouflage is of little value.

I think it's much more important to establish yourself as a player that insures a lot of hands for whatever reason; probably because you're simply a "wild-ass" gambler. This can be done very cheaply at the minimum bet level, especially if the casino where you play uses chips in the denomination of $2.50. Instead of tipping the dealer with those chips, use them to make some "dumb" insurance bets when you have a minimum bet placed, especially if you can do it on the first hand that's dealt after the shuffle. Even non-counting pit critters know the count in a multi-deck game can't usually get high enough on the first round to warrant the insurance bet, so it'll add to your image as a "gambler." Hey, it's all just luck anyway, isn't it? (Just kidding there, folks. We know darn well it isn't all just luck!) Anyway, insure those 12s and 15s! The worse the hand, the better. You will win some of those, so it's not exactly like throwing the chips down the drain. Remember the 8% cost I talked about in the beginning? Well, 8% of $2.50 is 20 cents, so your actual cost isn't exactly huge. Generally, you'll be safe when you insure the hands of 9 to11 and 17 to 20, because a lot of gamblers do that. Just don't get carried away and start doing it on a count that doesn't make the bet profitable. Keep the camouflage bets for when you've got the minimum out there. One more hint: A lot of dealers try to "help" the table by quickly offering insurance and then peeking at their hole card. If you are making a lot of weird insurance bets, they won't pass you by and that's important for those times when it's worthwhile to make a "real" insurance bet.

I guess the best way to sum this up is to try and make your insurance bets appear to be random hunches on your part. By making insurance bets a (somewhat) regular part of your game, you'll be much more inclined to make them when they really count. It's just too important to give up.

I'll see you here next time.

  

RETURN TO THE CURRENT BLACKJACK ARTICLES

2007 Articles

2006 Articles

2005 Articles

2004 Articles

2003 Articles

2002 Articles

2001 Articles

2000 Articles

1999 Articles

1998 Articles

1997 Articles