Perfect insurance count

Kasi

Well-Known Member
#1
Can any of you smart guys out there design a perfect count system for a Spanish 21 8 deck game that pays 5-2 on insurance?

Something comparable to all non-10's as +1 and all 10's as -2 for a regular BJ game.

Just seems like it should be possible?
 

jack.jackson

Well-Known Member
#2
Kasi said:
Can any of you smart guys out there design a perfect count system for a Spanish 21 8 deck game that pays 5-2 on insurance?

Something comparable to all non-10's as +1 and all 10's as -2 for a regular BJ game.

Just seems like it should be possible?
Wait a couple days and I'll go over to my secret underground labaratory, mix up some special chemicals I've been working on, and Ill let you know what I come up with.
 

Kasi

Well-Known Member
#4
jack said:
Wait a couple days and I'll go over to my secret underground labaratory, mix up some special chemicals I've been working on, and Ill let you know what I come up with.
:grin: I think you're off the hook now lol.

But maybe you could figure how often quipper's system above would actually get to a >=+96 RC and/or how how much reduction in house edge could be expected.

But that would take some really serious chemicals :)
 

Kasi

Well-Known Member
#5
quipper said:
non ten = +2
tens = -5
starting at an RC of zero, take insurance when RC is greater than 96!

Thanks quipper - we must be on the same page because I came up with non-ten=+1 and 10's = -2.5 and take insurance when RC > +48!

Think I like yours better - who likes fractions anyway lol.

I'll probably be saying to myself "dam - was the RC 65 or 75 last round?" lol.

Anyway it's not like I'll have much else to think about while playing so at least it'll give me something to do lol.

Thanks again for helping a stranger.
 

k_c

Well-Known Member
#6
Reverse tags

To me the initial running count is the key. If insurance paid 2.5 to 1 = 5 to 2, then when the ratio of tens to non-tens = 2/5 the insurance bet is even. In a single spanish 21 deck there ar 36 non-tens and 12 tens. Tagging the non-tens=-2 and tens=5 gives an initial running count = 36*(-2)+12*5=-72+60=-12. Initial running count (IRC)=numDecks*(-12). Whenever the current running count=0, the insurance bet EV=0 at 2.5 to 1 odds.

This can be done for any odds. If insurance paid 3 to 1, tag non-tens=-1 and tens=3. For normal deck and 2 to 1 insurance odds, tag non-tens=-1 and tens=2. In each case when running count > 0, insurance is +EV.

I like to tag negative cards with a negative sign and positive cards with a positive sign instead of the other way around. If done that way, current running count can always be figured by looking at the present shoe comp. Once initial running count is figured, current running count goes up by (neg tag) when a negative card is dealt and down by (pos tag) when a positive card is dealt. If done this way, any counting system can be true counted.

k_c
 

Kasi

Well-Known Member
#7
k_c said:
If done this way, any counting system can be true counted.k_c
Thanks k_c that's an interesting appoach.

Also I liked your overall approach making it clear you probably arrived at the same count I did in 15 seconds instead of like an hour of fussing around like I did lol.

Must be nice actually knowing what you're doing :)

Was it you or Josh that did all that composition stuff of what cards would make up a count? Anyway, if you have a way of determining how often such a count may actually happen in say a 6/8 shoe let me know if you could.

No big deal - I'll probably find out after I don't get it in the first hour or 2 lol.

And, if I do, I bet I lose the bet anyway lol.

I just thought I might give Double Attack Blackjack a try in AC in a few weeks - I like to try new games.

So, unless, you can come up with a some regular counting system for the game, (seems like it might be at least a little easier to do that than reg SP21 since there's none of that 678 payoffs etc), I'll probably just start at RC=0 and go up lol.
 
#8
k_c said:
To me the initial running count is the key. If insurance paid 2.5 to 1 = 5 to 2, then when the ratio of tens to non-tens = 2/5 the insurance bet is even.



WHEN the the ratio of tens to non-tens = 2/7 the insurance bet is even.
 

Canceler

Well-Known Member
#12
Oh-oh! I think I'll leave it.

davidpom said:
My advice (take it or leave it) - avoid the insurance bet. It's a bad bet, period, counted or not.
Considering that insurance is the most valuable index play there is, why would you not make use of it?
 

QFIT

Well-Known Member
#13
davidpom said:
My advice (take it or leave it) - avoid the insurance bet. It's a bad bet, period, counted or not.
I'll leave it. See http://www.blackjackincolor.com/cardcountingcover6.htm. Correct Insurance is top of the pink area. No Insurance is top of the green area. With six decks and typical penetration, no Insurance reduced SCORE by 12.3%. Perfect Insurance added another 11.5%. Single deck, no Insurance reduced SCORE by 29.2% and perfect added another 10.5%.
 
#14
QFIT said:
I'll leave it. See http://www.blackjackincolor.com/cardcountingcover6.htm. Correct Insurance is top of the pink area. No Insurance is top of the green area. With six decks and typical penetration, no Insurance reduced SCORE by 12.3%. Perfect Insurance added another 11.5%. Single deck, no Insurance reduced SCORE by 29.2% and perfect added another 10.5%.
Ok, I've no issues with this for a skilled advantage player. As long as they have sufficient money management / bankroll to ride out the standard deviation "storms". Sorry, I should have mentioned that caveat in my earlier post. The issue with insurance, and doubling / splitting for that matter, is that it requires further player bankroll - and doesn't always work out. Thus whilst you might improve your long term game by taking COUNTED insurance, you could affect your SHORT TERM bankroll negatively by putting more money on the table (especially if you're overbetting).

For the AVERAGE player (i.e. a typical non-counter giving the house a 2% theo) then I stand by my original quote: Insurance is a bad bet.
 

QFIT

Well-Known Member
#15
SCORE takes into account standard deviation. Insurance is a critical rule. Which is why the UK games are barely playable without very large spreads.
 

mjbballar23

Well-Known Member
#17
davidpom said:
Just as well I can no longer play the UK games then. Barring does have its advantages after all. :)
David, just for future reference everybody on this forum knows you have been barred from all UK casinos so you dont have to repeat it in every post:p
 
Top