Index Generation, 16 v 9.

1357111317

Well-Known Member
#1
I have been trying to determine the early surrender (or late surrender, doesn't matter for non tens) zen indice for 16 v 8 and 9. CVdata gives me the index of +1 for 16 v 9which makes no sense. Basic stratagy for this play is surrender so obviously the index has to be negative. Also I have got values anywhere from 15 to 9 for 16 v 8. Any comments on this?
 

1357111317

Well-Known Member
#2
Oh and I guess i should say this before someone points this out, I found the indices on Snyders website but some I can't trust them. Why does the early surrender 16 v 8 indice differ from the late surrender 16 v 8 indice? For 16 v 8 there is no difference between early and late surrender.
 

QFIT

Well-Known Member
#3
1357111317 said:
I have been trying to determine the early surrender (or late surrender, doesn't matter for non tens) zen indice for 16 v 8 and 9. CVdata gives me the index of +1 for 16 v 9which makes no sense. Basic stratagy for this play is surrender so obviously the index has to be negative. Also I have got values anywhere from 15 to 9 for 16 v 8. Any comments on this?
1. What are you using for TC divisor? full or quarter decks?
2. Is this sex decks?
3. RA or EV-Max indexes?
4. Floor, truncate or round?
5. You must also select 8,8, 8 and 9. If you don't, then 8,8 will be taken into account for 16. If you do select them, 8,8 will be omitted. It sounds like you had 88v9 selected once and not another time, which is why you would have received answers of 9 and 15.
6. The index is not necessarily negative just because BS is surrender.
 
Last edited:

1357111317

Well-Known Member
#4
QFIT said:
1. What are you using for TC divisor? full or quarter decks?
2. Is this sex decks?
3. RA or EV-Max indexes?
4. Floor, truncate or round?
5. You must also select 8,8, 8 and 9. If you don't, then 8,8 will be taken into account for 16. If you do select them, 8,8 will be omitted. It sounds like you had 88v9 selected once and not another time, which is why you would have received answers of 9 and 15.
6. The index is not necessarily negative just because BS is surrender.
1. Full
2. 6
3 EV
4. Floor
6. Why is that? Logically that makes no sense to me.
 

QFIT

Well-Known Member
#5
It's not really predictable since the the cards in the hand are counted. This is why the 16vt hit/stand index confuses people. In any case, I get -1 for 16v9 if you also select the 8,8v9 index. The index is right on the border between -1 and zero and could change conceivably depending on penetration and deck estimation resolution.
 

bjcount

Well-Known Member
#6
1357111317 said:
I have been trying to determine the early surrender (or late surrender, doesn't matter for non tens) zen indice for 16 v 8 and 9. CVdata gives me the index of +1 for 16 v 9which makes no sense. Basic stratagy for this play is surrender so obviously the index has to be negative. Also I have got values anywhere from 15 to 9 for 16 v 8. Any comments on this?
Primeman
I just went thru a discussion with Don S regarding indices generated using cvdata for 16 vs 8. After running many sims using different methods the one simple method that worked is this:

Run 16 vs 8,9, 10 & A in all combinations using the Comp-Dependant index generator. What happens in a normal index gen is it also uses 8,8 vs 8,9,10 & A to compute the 16 vs surrender indices.

But we all know that 8,8vs 8 or 9 is an always split but the program doesn't compensate for it by removing it from the equation.

The end result is that the indices are being skewed by the hands that should be split, not surrendered.

Give it a try and let me know what you get.

BJC
 

QFIT

Well-Known Member
#7
bjcount said:
Primeman
I just went thru a discussion with Don S regarding indices generated using cvdata for 16 vs 8. After running many sims using different methods the one simple method that worked is this:

Run 16 vs 8,9, 10 & A in all combinations using the Comp-Dependant index generator. What happens in a normal index gen is it also uses 8,8 vs 8,9,10 & A to compute the 16 vs surrender indices.

But we all know that 8,8vs 8 or 9 is an always split but the program doesn't compensate for it by removing it from the equation.

The end result is that the indices are being skewed by the hands that should be split, not surrendered.

Give it a try and let me know what you get.

BJC
CVData does or does not compensate according to your desires. If you ALSO select the 88 index, it excludes 88 from the 16 index. If you do not, it does not. That gives you the choice.
 

bjcount

Well-Known Member
#8
QFIT said:
CVData does or does not compensate according to your desires. If you ALSO select the 88 index, it excludes 88 from the 16 index. If you do not, it does not. That gives you the choice.
Sorry Norm but I don't understand. I went through 10 differnet setups, from running a complete sim for every hand type to just running a single index gen beat to death for just one hand type and it never produced the result I was looking for.

When I did it in Comp-D index gen it worked perfectly on the first attempt once I set up all the combinations of the hand.

For example, 16 vs 8. Using RPC LS should be +5. It didn't matter how you set up or how many hand types you wanted to generate the index for using the standard cv index generator, the result was +7. So apparently the sim is including the 8,8 vs 8 as a combination of 16 vs 8 which we all know is an always split hand, which is why the result may be skewed. By running a CD index gen, and running every combination of 16 vs 8 (10-6,9-7,A-5) the result was now +5 which was the number I was looking for. And yes, A-5 was not a surrender hand.

BJC
 
Last edited:
Top