Strategy gain taken in account or not in this calculator?

K

k_c

Well-Known Member
ThodorisK said:
http://www.bewersdorff-online.de/black-jack/

Give the calculator any example, and you get at the bottom of the page the value for "Expected gain at the beginning of a game"

Does this value take in account the strategy gain or not?:confused:
Click to expand...
It is incorrect.

Decks=1, no resplit, no double after split, no removals

Results from calculator
EV vs up cards (2 through ace)
8.9% 12% 15.3% 18.8% 22.3% 14.2% 5.7% -4.1% -17.4% -33.8%
Overall: -.773%

(I assume double on any 2 card total, cannot double on any number of cards, 1 card to split aces.)

Correct optimal results
Dealer: S17
EV vs up cards (2 through ace)
9.995% 13.59% 18.19% 23.35% 23.94% 14.53% 5.441% -4.383% -17.08% -36.03%
Overall: +.0248%

Dealer: H17
EV vs up cards (2 through ace)
9.970% 13.62% 18.25% 23.37% 24.22% 14.53% 5.441% -4.383% -17.08% -38.90%
Overall: -.1684%
 
T

ThodorisK

Well-Known Member
Do you find the error of the wrong (according to you) calculator to be a particular stable error? Because if yes, I can still use that wrong calculator.

Besides any calculators, are there any tables on the net that show the player's edge which corresponds to each particular true count, taking in account the strategy gain?
I think I just found such a one, but it is for particular rules. Better than nothing though:http://www.blackjackincolor.com/truecount3.htm
By "indexes" they mean the modifications of the basic stratgy according to the count? Or not?

Hm, I'm not satisfied with this. Any better charts or free calculators?
 
Last edited:
K

k_c

Well-Known Member
ThodorisK said:
Do you find the error of the wrong (according to you) calculator to be a particular stable error? Because if yes, I can still use that wrong calculator.
Click to expand...
I'm too lazy to check to see if there is any consistent error. The calculator purports to be able to get EV for any shoe state (take into consideration cards removed) but how could it include strategy departures without referencing any counting system? My programs also consider cards removed. One of them can compute EV for perfect optimal play but that is not the same thing as using indices for a count. Optimal play is better because the perfect play is always made. The other just computes overall EV using total dependent basic strategy. This gives it the ability to compute overall BS EV for any single shoe composition at a time. Therefore, I decided to sim many of these instances (below) to see what I get. Adding the ability to use indices for a given count could be added to the second program.

ThodorisK said:
Besides any calculators, are there any tables on the net that show the player's edge which corresponds to each particular true count, taking in account the strategy gain?
I think I just found such a one, but it is for particular rules. Better than nothing though:http://www.blackjackincolor.com/truecount3.htm
By "indexes" they mean the modifications of the basic stratgy according to the count? Or not?

Hm, I'm not satisfied with this. Any better charts or free calculators?
Click to expand...
In the Weekend Warriors II thread I used my above second program to sim each round with exact calculation of pre-round overall EV using basic strategy. (Dead link: http://www.bjstrat.net/WW2.html) _ Results here_. I don't think anyone has ever done a sim like this using actual calculated EV. Granted that this applies to the WWII rules and pen and does not include any index variations. If index variation was included only one count at a time could be referenced.

I'm new at doing sims. I used the random number generator in C++. C++ is a respected language and I assume it has a decent RNG but I have seen discussions about them. Anyway I think it gives a decent picture of what to expect in the WWII game albeit using basic strategy. Different decks/rules/pen would vary as would using indices but I don't think using indices would drastically affect results.

I, too, am looking to answer some of the questions you are asking. Hope this helps.
 
You must log in or register to reply here.
Top