I 18 Question(s)

[sbrugby]

I'm trying to figure out the right indices I should be using and I just wanted to double check with ya'll. I plan to be playing 8-deck h17 das. From what I understand the indices of BlackJack Attack 3 on page 213 table 10.1 are best? I'm just a bit confused as to why some of indices differ from table 5.1 on page 62.

For instance:
12 v 5 is -2 on page 62 but its -1 on page 213.
12 vs 2 is +3 on page 62 but its +4 on pg 213

Is this due to testing method or something?

Then there's the risk averse indices on page 375 table 13.11 which I'm not sure if I should incorporate b/c it has some of the same TC index differences with table 10.1

Thanks

[Three]

The tables differ due to different TC calculating procedures. Don's table 10.1 uses flooring in the calculation. Don should chime in to tell you exactly how he did it so you can calculate TC's the same way when you use his indices. Table 5.1 the indices are from Wong's book and he didn't specify how he calculated TC's. It comes down to how you come up with your deck estimates and how you deal with coming up with an integer after doing the math.

Risk averse indices lower risk of ruin which allows you to bet more at the same risk. Ev maximizing indices will have you double your bet to make an extra penny in expectation (EV). This is not a good idea. Doubling should have a higher return before you risk any extra money. If you have a huge BR MAYBE you want to live with the crazier swings involved in EV maximizing indices but I think most in the situation still use RA indices.

You are lucky Don SChlesinger (author BJA3) is a contributor to this site and is happy to help everybody including those with grammar mistakes. LOL

If you include an authors name in the title of a thread they get an Email alert from the site but Don checks regularly anyway. Thread titles can't be edited.

[ZenMaster_Flash]

Don Schlesinger is a true pioneer in Card Counting

His text Blackjack Attack (3rd ed. only) is "must reading"

for the likes of us. It is a truly invaluable resource !

Perhaps his single most important contribution is "SCORE."

[sbrugby]

Thank you very much for the insight.

Table 5.1 the indices are from Wong's book and he didn't specify how he calculated TC's.

So 5.1 tc's are generated in some unknown fashion? Neither the 1994 ProBJ table or the 1981 ProBJ table seem to match these.

BJA3 Chapter 10 specifies half deck estimates w/ flooring, so I'm assuming as long as I'm doing this, these indices apply to me. It says nothing of them being risk averse so I assume they are Max EV. It sounds like without software to generate my own RA indices, I should just stick with these.

You are lucky Don SChlesinger (author BJA3) is a contributor to this site and is happy to help everybody including those with grammar mistakes. LOL
Don Schlesinger is a true pioneer in Card Counting

Yes, I am loving BJA3 and Don seems like an awesome guy.

Perhaps his single most important contribution is "SCORE."

I'm definitely reading this next once I figure this indices stuff out.

[BJGenius007]

I'm trying to figure out the right indices I should be using and I just wanted to double check with ya'll. I plan to be playing 8-deck h17 das. From what I understand the indices of BlackJack Attack 3 on page 213 table 10.1 are best? I'm just a bit confused as to why some of indices differ from table 5.1 on page 62.

For instance:
12 v 5 is -2 on page 62 but its -1 on page 213.
12 vs 2 is +3 on page 62 but its +4 on pg 213

Is this due to testing method or something?

Then there's the risk averse indices on page 375 table 13.11 which I'm not sure if I should incorporate b/c it has some of the same TC index differences with table 10.1

Thanks

For anyone who wants to a second opinion, these indexes are generated by my program. Hi-Lo, S17, 6 decks.
UpCard 2 3 4 5 6
15 -7 -8 -10 -10 -12
14 -4 -5 --8 -8 -8
13 -2 -3 -4 -6 -6
12 3 1 0 -2 -3

Indexes are floored. If TC is greater than or equal to the index, you stand. If TC is less than the index, you hit.

For example, 12 v 6, you stand if TC is -3. You hit if TC is -3.001 since it is floored to -4. Another example, 12 v 2, you stand if TC is 3, you hit if TC is 2.75 since it is floored to 2.

[sbrugby]

I did my best to merge the 8-deck H17 indices from chapter 10 with the catch 22 risk averse indices and came up with this (excluding 10,10 splits). Assuming flooring and half deck TC resolution do these seem right for 8-deck H17 DAS?

Insurance	Count >= +3; get insurance
16 vs 9	Stand if count >= +5; else hit
16 vs 10	Stand if count >= 0; else hit
15 vs 10	Stand if count >= +4; else hit
13 vs 2	Stand if count >= -1; else hit
13 vs 3	Stand if count >= -2; else hit
12 vs 4	Stand if count >= 0; else hit
12 vs 5	Stand if count >= -1; else hit
12 vs 6	Stand if count >= -3; else hit
11 vs Ace	Double Down if count >= -1; else hit
10 vs 10	Double Down if count >= +7; else hit
10 vs Ace	Double Down if count >= +4; else hit
9 vs 2	Double Down if count >= +1; else hit
9 vs 7	Double Down if count >= +4; else hit
12 vs 2	Stand if count >= +4; else hit
12 vs 3	Stand if count >= +2; else hit
8 vs 5	Double Down if count >= +5; else hit
8 vs 6	Double Down if count >= +3; else hit
A8 vs 5	Double Down if count >= +1; else stand
A8 vs 6	Double Down if count >= 0; else stand

[Fireman]

I just started to scan sbrugby's chart from bottom to top and on the first two items I found a discrepancy. You should stand on A8 vs 6 and A8 vs 5, not hit. I did not look at anything on his chart after that.

[apkevy]

I did my best to merge the 8-deck H17 indices from chapter 10 with the catch 22 risk averse indices and came up with this (excluding 10,10 splits). Assuming flooring and half deck TC resolution do these seem right for 8-deck H17 DAS?

Insurance	Count >= +3; get insurance
16 vs 9	Stand if count >= +5; else hit
16 vs 10	Stand if count >= 0; else hit
15 vs 10	Stand if count >= +4; else hit
13 vs 2	Stand if count >= -1; else hit
13 vs 3	Stand if count >= -2; else hit
12 vs 4	Stand if count >= 0; else hit
12 vs 5	Stand if count >= -1; else hit
12 vs 6	Stand if count >= -3; else hit
11 vs Ace	Double Down if count >= -1; else hit
10 vs 10	Double Down if count >= +7; else hit
10 vs Ace	Double Down if count >= +4; else hit
9 vs 2	Double Down if count >= +1; else hit
9 vs 7	Double Down if count >= +4; else hit
12 vs 2	Stand if count >= +4; else hit
12 vs 3	Stand if count >= +2; else hit
8 vs 5	Double Down if count >= +5; else hit
8 vs 6	Double Down if count >= +3; else hit
A8 vs 5	Double Down if count >= +1; else hit
A8 vs 6	Double Down if count >= +1; else hit

Where did you find those indices ? I find " Doubling down on A8 vs 6 if tc > +1 else HIT " very confusing ????

[sbrugby]

Where did you find those indices ? I find " Doubling down on A8 vs 6 if tc > +1 else HIT " very confusing ????

I just started to scan sbrugby's chart from bottom to top and on the first two items I found a discrepancy. You should stand on A8 vs 6 and A8 vs 5, not hit. I did not look at anything on his chart after that.

Oops haha, I got a little trigger happy with copy and paste. Ty for pointing that out.

I updated that in the post and changed A8 vs 6 index to zero from 1 b/c I realized it would be lower since my table would be for H17.

[Bodarc]

Hi sbrugby

A8 v 5 and A8 v 6 are not included in the Ill 18. What you are missing is 10 10 V 5, Split = 5 else stand, and 10 10 V 6, Split = 4 else stand. Some people do not use 10,10 vs 5 and 6 because of heat problems.

I believe A8 V 5 should be 2 instead of 1.
10 V 10 should be +4 instead of +7
9 V 7, I use 3 instead of 4
12 V 2 is 3

You will find if you run sims that you won't always get the exact same indices if you run them more than once. It is my understanding that being off a little on an index won't make any difference over the long term. The indices I gave you are the ones I use.

I hope this helps.

[DSchles]

Sorry to be late to the party; I just got back from a short vacation. Will try to summarize:

1. The page 213 indices in BJA3 should be as accurate as you can get. As previously stated, they are floored, using half-deck resolution for the multi-deck games. There are separate values for H17, when they differ from S17, and they are e.v.-maximizing.

2. Wong's indices from the original article are different because, a) they were derived by Wong, and b) he used a different methodology for rounding (see p. 254 of the later editions of Pro BJ for his explanation). In any event, my p. 62 footnote 7 reference to Wong, p. 173, is to the "Bible" edition of Pro BJ, not the later editions. Don't forget, the original article was written over 30 years ago.

3. Some risk-averse indices are the same as their I18 counterparts; others aren't. So, the Chapter 13 section on risk-aversion lists those indices along with the Catch-22 additions to the I18, namely doubling 8 vs. 5 and 6 and doubling A,8 vs. 5 and 6.

4. Finally, you didn't include the Fab 4 surrenders, which are nonetheless part of the indices I furnished. So, in all, there are 22 + 4 = 26 e.v.-maximizing indices and seven r-a indices (shaded in gray) that are different from the I18 ones. In all, when people refer to BJA3 furnishing only 18 indices for, say, multi-deck, there are, in fact, more than double that amount, as there are the 33 (26 + 7) referenced above, plus the four more for H17. So, 37 in all.

Hope this helps. Write back with further questions.

Don