Question for Norm Wattenberger re CV Index Generator

M

mikeyd

Active Member
#1
#1
Norm,
This question relates to the HiLo indices generated by Southpaw using the CV Index Generator and posted on this message board on December 7.
(6D, .75 Pen, SP to 4, No RSPA, 1 card to split aces, LS, DOA2, DAS, S17).
These indices have been extremely helpful.
While the vast majority of the indices make sense to me, there are a few cases where the recommended action appears to differ from the indices.
For example, in the Surrender Table, 8,8 vs X, the recommended play is Surrender >=1. However, the indices seem to indicate that it is better to surrender even at a TC of 0 (Surrender -0.50, Play -0.536691).
Other examples are Double 9 vs 3 and Double 8 vs 6.
Can you explain how the recommended play was determined based on the indices generated?
mikeyd
 
Southpaw

Southpaw

Well-Known Member
#2
#2
mikeyd said:
Norm,
This question relates to the HiLo indices generated by Southpaw using the CV Index Generator and posted on this message board on December 7.
(6D, .75 Pen, SP to 4, No RSPA, 1 card to split aces, LS, DOA2, DAS, S17).
These indices have been extremely helpful.
While the vast majority of the indices make sense to me, there are a few cases where the recommended action appears to differ from the indices.
For example, in the Surrender Table, 8,8 vs X, the recommended play is Surrender >=1. However, the indices seem to indicate that it is better to surrender even at a TC of 0 (Surrender -0.50, Play -0.536691).
Other examples are Double 9 vs 3 and Double 8 vs 6.
Can you explain how the recommended play was determined based on the indices generated?
mikeyd
Click to expand...
This is the thread that our friend is referencing. The eighth post contains a link to the report generated by CVIndex.

http://www.blackjackinfo.com/bb/showthread.php?t=20292

SP
 
Last edited:
M

mikeyd

Active Member
#3
#3
CV Index Generator

For anyone interested, I raised this same question on another message board, bj21.com, if you want to follow the discussion.

To Southpaw,
In response to my post on that message board, Don Schlesinger raised the following question:
"Are you sure you asked for an e.v.-maximizing index and not a risk-averse one?"
Can you help respond to his question?
I don't know if you are a member there, but feel free to join the discussion & respond directly, or if you let me know the answer, I can respond.
Thanks buddy.
mikeyd
 
Southpaw

Southpaw

Well-Known Member
#4
#4
mikeyd said:
For anyone interested, I raised this same question on another message board, bj21.com, if you want to follow the discussion.

To Southpaw,
In response to my post on that message board, Don Schlesinger raised the following question:
"Are you sure you asked for an e.v.-maximizing index and not a risk-averse one?"
Can you help respond to his question?
I don't know if you are a member there, but feel free to join the discussion & respond directly, or if you let me know the answer, I can respond.
Thanks buddy.
mikeyd
Click to expand...
They are NOT RA indices. I personally use RA indices, but these are not. If they were RA the 8,8 vs. X would have been negative. Instead, it is +1.
 
M

mikeyd

Active Member
#5
#5
Thanks Southpaw.
Would it be possible to generate the HiLo RA Indices for the same game?
(6D, .75 Pen, SP to 4, No RSPA, 1 card to split aces, LS, DOA2, DAS, S17).
mikeyd
 
Southpaw

Southpaw

Well-Known Member
#6
#6
mikeyd said:
Thanks Southpaw.
Would it be possible to generate the HiLo RA Indices for the same game?
(6D, .75 Pen, SP to 4, No RSPA, 1 card to split aces, LS, DOA2, DAS, S17).
mikeyd
Click to expand...
At what TC do you put out your max-bet? This is something that is needed to generate RA indices.

SP
 
QFIT

QFIT

Well-Known Member
#7
#7
Ahh, two things:

On extremely close calls, like 9v3, CVData will select the less risky decision, even if the Risk Averse indexes option is turned off. This is not the same as making RA calculations -- it's just a common sense decision.

On the 88 Split, this is a bug. The index is correct, but the numbers are displayed incorrectly. The porblem only affected 88 surrender, and oddly is part of the hole-carding code. I put up a fix.
 
M

mikeyd

Active Member
#8
#8
Norm,
Thanks for your responses.
I understand re 8,8 Split, but I am still confused re 9v3 where the recommended DD is >=0, yet the ev at -1 is higher for DD (0.090573)
than for No DD (0.088943). Is CVData saying that it is less risky and thus better to hit at -1 rather than DD even though the ev is lower? Can you explain how this is determined given that it's ev maximizing and not RA indices?

Southpaw,
Max bet at TC of +4 as follows:
0 1 unit
+1 2 units
+2 5 units
+3 8 units
+4 10 units

Thanks.
mikeyd
 
Southpaw

Southpaw

Well-Known Member
#9
#9
mikeyd said:
Norm,
Thanks for your responses.
I understand re 8,8 Split, but I am still confused re 9v3 where the recommended DD is >=0, yet the ev at -1 is higher for DD (0.090573)
than for No DD (0.088943). Is CVData saying that it is less risky and thus better to hit at -1 rather than DD even though the ev is lower? Can you explain how this is determined given that it's ev maximizing and not RA indices?

Southpaw,
Max bet at TC of +4 as follows:
0 1 unit
+1 2 units
+2 5 units
+3 8 units
+4 10 units

Thanks.
mikeyd
Click to expand...
Unfortunately, CVData does not allow you to input a custom spread when generating RA indices; rather, I believe it assumes an optimal spread based upon when you place your max-bet. However, your spread is close enough to optimal for all intents and purposes.

SP
 
Southpaw

Southpaw

Well-Known Member
#11
#11
Mikeyd,

The index generator has been running all night. It is coming near its completion. Should have the results to you by tonight.

Best,

SP
 
Southpaw

Southpaw

Well-Known Member
#13
#13
mikeyd said:
Thanks alot Southpaw.
Really appreciate it.
MikeyD
Click to expand...
Here are your RA indices for Hi-Lo, based on your specified parameters:

https://docs.google.com/document/d/12ZFSDiPXXBBUCAuzrRAu9TbOq5lCfFLmWsjGvQcDL_k/edit?hl=fr#

P.S.-- I did make a few inconsequential assumptions. One being that I assumed you use 1/2 deck resolution. (You still divide by the number of full-decks, although you may be dividing by 2.5, 3.5 and etc.) If this does not truly reflect your playing style, it is inconsequential. Furthermore, I assumed you truncated your TC.

Best,

SP
 
M

mikeyd

Active Member
#14
#14
Thanks Southpaw.
These RA indices are a wealth of information.
I plan to follow your advice and surrender 8,8 vs X at the RA index of -3, rather than the EV index of +1.

One observation is that this simulator output shows many apparently contradictory results, such as DD 10 vs X where the recommended play
is >=7, even though the indices seem to indicate it is better to DD
at TC of +6 (DD 0.069558, no DD 0.050806).
I know Norm said that where the EV difference is insignificant, CVData chooses the less risky decision.
I wonder if this applies for RA indices also, or if the riskiness is already built into the indices calculated.
I also wonder what is the numerical threshold for being considered insignificant.
Norm, can you address this?
Thanks.
MikeyD
 
M

mikeyd

Active Member
#17
#17
Norm,
I am simply trying to follow the flow of information from the alternative EVs at each TC to the recommended action.
However, it appears that there is another step, which is not displayed in the output that I have seen, which determines the decision point.
Can you explain this other step?
For example, for DD 10 vs X, after the program determines the EVs at +6
of .069 for DD & .058 for no DD, and the EVs at +7 of .080 for DD & .055 for no DD, what is the calculation to determine that the precise optimal point to DD is +7? I understand that this additional step relates to risk, but is there a mechanical calculation to arrive at the precise trigger point, such as a certain percentage over the crossover point?
Sorrry for being so granular, but I would just like to better understand your outstanding software.
MikeyD
 
QFIT

QFIT

Well-Known Member
#18
#18
I'm afraid I don't know of any detailed explanation. When I have time, maybe I'll write one up, if no one else finds one or writes one first.
 
You must log in or register to reply here.
Top