Perfect index for insurance

Hi there,

What is the perfect insurance hi-lo strategy number for 6D game? 217/72 (~3.01389)? The number 3.01389 means that for any 52 cards of the shoe, on average, high cards will be more then low cards by 3.01389. Middle cards are not tracked by hi-lo system. So, on average, we could suspect that there are 12 middle cards in every 52 cards. Thus, in a pack of 52 cards, high cards will be 21.506945 ((52-12)/2+3.01389/2). Four of five high cards will be ten valued cards. Thus, the number of tens will be 17.205556 (21.506945*4/5). But 17.205556/52 is less then 1/3. So, it looks like that the index 3.33(3) is the precise number.

It will be very kind if somebody throw some light on my ignorance.

Blackjack Rookie

Blackjack Rookie,

A perfect insurance index would tell you when the ratio of non-X's to X's in the remaining pack is exactly 2:1. Thus, to have a perfect insurance index, a count system has to count all X's in one group, and all non-X's in the other group.

One such unbalanced, running count, perfect insurance count begins with an initial running count (IRC) of -4*(number of decks) and counts all X's as -2 and all non-X's as +1. When the PIC is zero, insurance is an even bet; when the PIC is positive, insurance is +EV; and when the PIC is negative, insurance is -EV.

For example, say you're playing a heads-up SD game. On the first round after the shuffle, you're dealt 5-6 vs. the dealer's A. Now, the PIC is -4+3 = -1, so DON'T take insurance. However, if you happened to peek the bottom card of the deck when the dealer offered you the cut, and you saw that is was a 7, NOW the PIC is exactly zero, so insurance is a coin flip. Furthermore, if you ALSO managed to see that the burn card was a 2, NOW the PIC is +1, and you SHOULD take insurance.

Now HiLo does not have a perfect insurance index because of two shortcomings: first, it lumps A's in with the X's; and second, it doesn't count the 7's, 8's, and 9's. Furthermore, the calculation you outlined in your post is flawed, because you assumed that the neutral cards ALWAYS represent (12/52)nds of the undealt pack, but this is not the case.

Hope this helps!

Dog Hand

If you count all NON-TENS as +4 and all TENS as −9's you will have a perfect insurance calculation.

If the simple running count equals +16 the Insurance bet is a toss-up, but good for variance-reduction and general risk-aversion.

Any Running Count exceeding +16 is an automatic rush to take insurance.

Understand?

This simple Ten Point Count is also pretty good for bet-sizing, but is pretty worthless for playing decisions.

It is best used by a partner playing at the same table where s/he is in charge of bet-sizing and you are responsible for playing decisions with a strong count e.g. Hi-Opt II, A.O. II, Zen, etc.

The two of you covertly signal info back and forth, (subtly and non-verbally); being sure to alternate who, at the table, will be the "big player" and frequently switching the role of "Big Player" to keep the pit confused.

Yikes! W.T.F. I must be intoxicated. I am revealing Pro secrets on a Free Forum.

A perfect side count for Insurance purposes would count Tens as -9 and all other cards as +4. But, I know of no one that uses this

An unbalanced count found in the 1981 version of Wong's Professional Blackjack. Here, Tens are counted as -2 and all other cards as +1. Insure if the count is greater than four times the number of decks. The Ten count can also be used for the Bust Out bet if you can find that rule anywhere.

This info courtesy of the nonpareil Norm Wattenberger.

Dog Hand,

Thanks for your expert response. But what I meant was the TC at which taking insurance is a coin flipping in a long run. I think I have not phrased my question correctly. Maybe it is because English is not my native language. I should have used the word �theoretical� instead of �perfect�. So, what I am interested to know is if the theoretical index for insurance in 6D game is 217/72 or 10/3.

Blackjack Rookie

FLASH,

Thanks for your readiness to help. I know that any count that compares tens to non tens will do perfect. Such are the counts you describe � (4/-9) and (1/-2). But I�ll have to revise your sentence �This simple Ten Point Count is also pretty good for bet-sizing, but is pretty worthless for playing decisions�. I think you are wrong. The BC of these systems is only 0.7226. And taking in account that main counter�s gain comes from betting, it is not for nothing that Wong says in his PB that 10-count strategies are obsolete. As to the playing strategy, these counts have PE of 0.6117. It�s not bad.

Blackjack Rookie

It's the former if you don't keep a side count of aces and adjust. It's the latter if you do.

Don