Index generation from BJA #3

[lij45o6]

Using D10 on page 515 of BJA, I am attempting to compute specific TD index using Snyder's Algebraic Approximation Method: http://www.blackjackforumonline.com/...algapprox.html to compute Hi-Opt II indicies for both index departures and MP generation.

So far, my adventure has been fruitless.

Using the data on D10 and computing for 16 vs 10 S17 1D, the values I compute for the AAM are:

M_1 = -0.4459
P_HOII = 112
i_1 = -70.696
T = -2

Index = (M_n * P_c) / i_n + T

I compute the index 16 vs T:
H, S: -1.29

Which is both incorrect and absurd.

Somewhere, I am computing the wrong variable but I don't know where.

[Three]

Sometimes approximations aren't very accurate.

[lij45o6]

Sometimes approximations aren't very accurate.

What would be the "correct" way of generating indices using BJA numbers?

[Three]

I just know I have used the best approximation method to try and figure things out before, only to find out they weren't accurate enough to be of any use for what I was trying to do. The latest incident, Don warned me that the best way to do what I wanted to do was through simulation (That is also true for what you are doing. Sims replaced the mathematical approximation for a reason). Don helped with the best approximation since the series of sims for my research project would take so long. As usual he was right and the approximation was too approximate to be of any use.

[lij45o6]

I just know I have used the best approximation method to try and figure things out before, only to find out they weren't accurate enough to be of any use for what I was trying to do. The latest incident, Don warned me that the best way to do what I wanted to do was through simulation (That is also true for what you are doing. Sims replaced the mathematical approximation for a reason). Don helped with the best approximation since the series of sims for my research project would take so long. As usual he was right and the approximation was too approximate to be of any use.

Okay so, I have to build a simulator to generate the correct indices? What about multi-parameter deviations as well?

[Stealth]

Sometimes approximations aren't very accurate.

Did you get that from Yogi Berra?

or

Is that the same as "we have no undetected security breaches?"

[lij45o6]

Did you get that from Yogi Berra?

or

Is that the same as "we have no undetected security breaches?"

?????

[Three]

Okay so, I have to build a simulator to generate the correct indices? What about multi-parameter deviations as well?

If you don't want to use the ones that have already been written for that purpose.

[Three]

So far, my adventure has been fruitless.

Using the data on D10 and computing for 16 vs 10 S17 1D, the values I compute for the AAM are:

M_1 = -0.4459
P_HOII = 112
i_1 = -70.696
T = -2

Index = (M_n * P_c) / i_n + T

I compute the index 16 vs T:
H, S: -1.29

Did you do the inner product for the EoRs for the play or for betting?

[SteinMeister]

What would be the "correct" way of generating indices using BJA numbers?

I use Norm's CVData simulation software to generate my indices.

[Three]

You also used m1 value for the threshold to reach for the play being considered. You do realize that is the m-value for single deck. Double deck is m2. Six deck is m6 and so on.

[Jack Jackson]

Okay so, I have to build a simulator to generate the correct indices? What about multi-parameter deviations as well?

Theres a online program you can use to figure out the MP indexes if your willing to put in the work and know what your doing..In case your interested heres the link


http://www.bjstrat.net/cgi-bin/cdca.cgi

[lij45o6]

Did you do the inner product for the EoRs for the play or for betting?

I used the EOR's for playing the specific hand match-up data found in BJA #3, pg 515.