Need experts to confirm

I posted some comments in two topics at advantageplayer.com some time ago, which I think are different from the other opinions. Since there were no response, I would like to try here.

Topic #1, was about insurance bet
http://www.advantageplayer.com/blackjack/forums/bj-main/webbbs.cgi?read=17762

No, not at all. What I mean is that, for each one TC increase in Hi-Lo, the increase in the value of the insurance wager goes up by virtually the same amount. Just look at Norm's green, er, um, LINE! :-)�KDon

see the section in Theory Of Blackjack chapter 3 called "Insurance is 'Linerar'."�KSonny

Insurance is a linear function. The penetration doesn't matter�K.Don

But I think ��penetration is a factor of Insurance calculation��

http://www.advantageplayer.com/blackjack/forums/bj-main/webbbs.cgi?read=17799
Ratio of neutral cards
Discussed with Richard Reid at BJMath long time ago, I seem to remember that...,
In the first half shoe, the higher the TC(+ or -, away from 0), the higher the ratio of neutral cards
In the next half shoe, the higher the TC, the lower the ratio of neutral cards

http://www.advantageplayer.com/blackjack/forums/bj-main/webbbs.cgi?read=17807
For a 6 decks game, using hi-lo,
the ratio of neutral cards on TC+10 at 208 cards left,
is different from the ratio of neutral cards on TC+10 at 52 cards left.
So penetration is a factor of Insurance calculation.

Hope someone can confirm if I am correct or not.
I will post about the other topic later on.

Thanks

I wrote a program that computes the shoe composition based upon the HiLo running count and penetration and accounting for whatever removals the user inputs. It is based on a weighted average of all possible subsets at a given penetration. EV for the insurance bet is positive when the percentage of tens > 1/3. Insurance data for 1, 2, 4, 6, and 8 decks is posted on my simple site www.bjstrat.net for all possible 2 card HiLo hands versus dealer up card of ace.

kc

KC, it seems that you are the only one who cares about the penetration in the sim program outcomes.
Although there is only a very small change in the ratio of neutral cards in different penetrations, I predict some of the strategy indexes have to be adjusted.

Enumerating insurance data for a given running count/pen/removal combo requires only the computation of the percentage of tens present for that particular case. Perhaps there is a statistically "best" overall index. I guess that's what is output by a sim.

kc