Disagree with certain indices?

pooptarts92

Well-Known Member
#1
Ok well I'm not gonna say I "disagree" with this indice, but I use Wong Halves and in Professional Blackjack it says 77 vs 8 is a split at true 2 and above. Now I Can agree with it to a point, since doubling is allowed after splitting, and maybe I can catch some aces on my 7's. But at the same time I feel like a donkey doing this split, whenever it happens (only did it once, didn't work out). Does anyone care to elaborate on this one? Also, share some of your indices that you find "weird" or "gut wrenching". I'm not saying this is a bad split or anything, I just don't see the big picture behind it is all. Perhaps it's similar to splitting 88 vs a 9 at high TC's (less than 7 of course), just a bit more favorable than hitting/surrendering.
 

UK-21

Well-Known Member
#2
You should be able to get a handle on the question by working out the probability of drawing advantageous cards and disadvantageous cards on both hitting 7,7 and drawing two cards on your split 7s. It will of course alter slightly depending on the TC and number of decks left in the shoe.

To answer your subsidiary question, I have recently moved my stand on 12v2 from TC+3 to TC+4 - I use hi-lo and at TC+3 if all cards are 10s it is correct to stand, if one of the three additional high cards is an ace it's an even call, and if two of the three additional high cards are aces, the maths suggest to hit as there's a lower chance of busting out with a 10.

The answer to this conundrum is, of course, to keep a side count of aces but on a six deck game the instances where this will offer information to alter your play are pretty limited. It's been discussed at some length, and on shoe games the consensus is that the advantage of keeping an ace side count isn't worth the extra effort. The incidence of 12v2 is just 750 times per 100,000 hands on average, and at a TC+3 (8% of plays) this works out to 60/100,000 or six opportunities in every 125-ish hours of play? Not worth getting too excited about.
 

k_c

Well-Known Member
#3
pooptarts92 said:
Ok well I'm not gonna say I "disagree" with this indice, but I use Wong Halves and in Professional Blackjack it says 77 vs 8 is a split at true 2 and above. Now I Can agree with it to a point, since doubling is allowed after splitting, and maybe I can catch some aces on my 7's. But at the same time I feel like a donkey doing this split, whenever it happens (only did it once, didn't work out). Does anyone care to elaborate on this one? Also, share some of your indices that you find "weird" or "gut wrenching". I'm not saying this is a bad split or anything, I just don't see the big picture behind it is all. Perhaps it's similar to splitting 88 vs a 9 at high TC's (less than 7 of course), just a bit more favorable than hitting/surrendering.
Without looking into the details (relative to Wong Halves) I would say there is probably a window of TC = +2 to TC = +X where splitting 7-7 v 8 DAS is the best play. When TC is greater than whatever value X turns out to be then splitting would no longer be best. X is probably a value high enough that realistically it would virtually never be encountered.
 

HockeXpert

Well-Known Member
#4
From my recollection, Bryce Carlson in BJFB says to split 77v8 is basic strategy and the index to hit is -1 (the equivalent to -.5 or 0 for hi-lo) for AOII when playing single or double deck. The reasoning is two-fold imho.

1. the outcome of the 77 hand is somewhat dependent on the number of sevens left in the deck

2. you are praying for a 3 or 4 as your first card on each seven

It must be a very close play (without checking BJA) and I do recall reading a thread somewhere about this a long time ago. Anyway, I would guess that the ra index for this might be quite different.

I would be curious what some of the experts say about this hand because I have wondered the same thing.

HockeXpert
 

QFIT

Well-Known Member
#7
21forme said:
Norm - A suggestion... As H17 games are becoming more ubiquitous, could you add that as an option to your charts?
I've moved on to the next thing. But, CVData can create these charts.
 

pooptarts92

Well-Known Member
#8
UK-21 said:
You should be able to get a handle on the question by working out the probability of drawing advantageous cards and disadvantageous cards on both hitting 7,7 and drawing two cards on your split 7s. It will of course alter slightly depending on the TC and number of decks left in the shoe.

To answer your subsidiary question, I have recently moved my stand on 12v2 from TC+3 to TC+4 - I use hi-lo and at TC+3 if all cards are 10s it is correct to stand, if one of the three additional high cards is an ace it's an even call, and if two of the three additional high cards are aces, the maths suggest to hit as there's a lower chance of busting out with a 10.

The answer to this conundrum is, of course, to keep a side count of aces but on a six deck game the instances where this will offer information to alter your play are pretty limited. It's been discussed at some length, and on shoe games the consensus is that the advantage of keeping an ace side count isn't worth the extra effort. The incidence of 12v2 is just 750 times per 100,000 hands on average, and at a TC+3 (8% of plays) this works out to 60/100,000 or six opportunities in every 125-ish hours of play? Not worth getting too excited about.
Yeah I just checked the surrender section, it includes 2-card combos and the best TC surrender for them. For 77 vs 8 it's true 9. Thanks for bringing that to my attention, it would rarely happen but when it does I'd save a nice chunk of quarters lol.
 

Sharky

Well-Known Member
#9
BS for DD says split 6's vs. 7 and 7's vs. 8...constantly see this one misplayed

Edit: for DAS only
 
Last edited:
#10
+3.5

UK-21 said:
You should be able to get a handle on the question by working out the probability of drawing advantageous cards and disadvantageous cards on both hitting 7,7 and drawing two cards on your split 7s. It will of course alter slightly depending on the TC and number of decks left in the shoe.

To answer your subsidiary question, I have recently moved my stand on 12v2 from TC+3 to TC+4 - I use hi-lo and at TC+3 if all cards are 10s it is correct to stand, if one of the three additional high cards is an ace it's an even call, and if two of the three additional high cards are aces, the maths suggest to hit as there's a lower chance of busting out with a 10.

The answer to this conundrum is, of course, to keep a side count of aces but on a six deck game the instances where this will offer information to alter your play are pretty limited. It's been discussed at some length, and on shoe games the consensus is that the advantage of keeping an ace side count isn't worth the extra effort. The incidence of 12v2 is just 750 times per 100,000 hands on average, and at a TC+3 (8% of plays) this works out to 60/100,000 or six opportunities in every 125-ish hours of play? Not worth getting too excited about.
Just use +3.5 for 12v2.
 

UK-21

Well-Known Member
#11
Certainly could take an average of the three situations. All borderline stuff really - one of a number of situations where there's little in it, and it's just not worth getting excited about, especially if like me you're a casual player who isn't looking to make any serious money from counting cards.
 

FLASH1296

Well-Known Member
#12
QFIT,

I do not fathom the following, that appears on the referenced page you cited:

"Note: Some indexes depend on other indexes. That is, an index can change depending on whether or not you use other indexes."

Could you elaborate and provide examples ?
 

assume_R

Well-Known Member
#13
FLASH1296 said:
QFIT,

I do not fathom the following, that appears on the referenced page you cited:

"Note: Some indexes depend on other indexes. That is, an index can change depending on whether or not you use other indexes."

Could you elaborate and provide examples ?
I'm not sure if this is what Norm was referring to, but consider this:

You have 12v5 and the TC is extraordinarily low, say -10. The correct move is to hit that 12v5. What if you get an A? Then you have 13v5. If you were using your 13v5 index, and hit, you could potentially get a winning hand of 19 or 20. If you weren't using your 13v5 index, you're standing on that 13. Thus your 12v5 index could possibly be dependent on if you are going to use a 13v5 or 14v5 index (since the probability of winning a 13v5 or 14v5 is dependent on if you're using those indices or not)

Just a simple example off the top of my head, but I think it makes sense...seems like it wouldn't make too much of a difference though.
 

FLASH1296

Well-Known Member
#14
I do not think that that was what QFIT was indicating in his note.
However, respecting assume_R's keen numeracy prompts me to post:

In the example given, hitting 12 vs. (any) 'stiff' at negative T.C.'s is a simple 'no-brainer'.
The play is made if I am an Ill18 player OR if I am an obsessive with all indices memorized.

I fail to see where a basic strategy departure can be dependent upon other than the current index via True Count.
 
Last edited:

QFIT

Well-Known Member
#15
FLASH1296 said:
QFIT,

I do not fathom the following, that appears on the referenced page you cited:

"Note: Some indexes depend on other indexes. That is, an index can change depending on whether or not you use other indexes."

Could you elaborate and provide examples ?
Very common. In creating a DD index, you compare doubling with hitting. But, depending on the hit/stand indexes, you may or may not rehit. You must create the Hit/stand indexes before you create the DD indexes. For that matter,you need to create the hit/stand indexes for a hand of 16 before you create the index for 15. Even more common is the effect on splits. You need to know the hit/stand and DD indexes first.
 

assume_R

Well-Known Member
#16
Yeah, I think I chose a poor example to use in my example, but what I was trying to say is that for whatever index you are currently generating, you need to know the win rates for all the "outcomes" of that index. QFIT correctly pointed out that it matters mostly for double downs and splits. I should have used a "Split" example in my post :(.

Better example (but same basic concept) based on qfit's response: so if you are wondering whether to split your 3's vs 2, you have to take into account the win rate if you happen to get a 5 on those 3's, 6 on those 3's, a 7 on those 3's, and so on. So therefore you'd need to first generate your 9vs2 index first to know the win rate of your 3,6 v 9.

Another example: should you double your 10v10? Well, if you happen get a 5, would you hit that 15v10? You need to take into account what would happen if you happen to get a total of 12, 13, 14, ... after you double, and see your win rates for each one to know if it's "worth" it to double vs. to just hit.
 
Top