4,4 vs 5 or 6

[Jabberwocky]

Just remember this simple rule. Splitting always takes precedence over doubling.

[Norm]

Just remember this simple rule. Splitting always takes precedence over doubling.

yeahthat.gif

[ZenKinG]

There is a separate index for a 4,4 v5 or 6. And you would split rather than double in a high true count typically, with of course the exception of 5,5 v 2-9, you obviously do not want to split that, but instead double it because it's a 10 and splitting it would result in two stiffs.

For a 4,4 v 4, 5 or 6 though, splitting gets more money on the table than a double down in the long run so therefore splitting will be the option in a high true count. Look for the number in your charts, there will be a separate index for splitting pairs and a separate number for a total of 8.

[Thirdbaseman]

[QUOTE=Tthree;149303]Each count differs in how fast gain is accumulated for each index play after the index is exceeded due to each counts correlation to the play. Norm has a great tool for HILO users to see the gain for certain hands as the index is exceeded:

http://www.card-counting.com/cvcxonlineviewer3.htm
Look at 44v5:
The split becomes profitable at TC -1 and reaches maximum gain by TC +3 at less than 0.1% gain over hitting.
The 8v5 double becomes profitable at TC +4 and surpasses the gain for splitting at TC +8 and adds to the profitability a lot with each TC increment increase.

Tthree,
I don't understand your comment. It looks to me like splitting 4,4 vs 5 is always better than doubling 8 vs 5. At +8, splitting has an advantage of .4, while doubling has an advantage of .3.
Maybe it's because I don't understand why the no double and no split advantages aren't the same.
In any event, shouldn't you always split 4s vs 5 rather than double?

[Dieter]

In any event, shouldn't you always split 4s vs 5 rather than double?

Splitting 4's doesn't make sense if you can't double after the split.

At least, that's what I gleaned from the books.

[Three]

Tthree,
I don't understand your comment. It looks to me like splitting 4,4 vs 5 is always better than doubling 8 vs 5. At +8, splitting has an advantage of .4, while doubling has an advantage of .3.
Maybe it's because I don't understand why the no double and no split advantages aren't the same.
In any event, shouldn't you always split 4s vs 5 rather than double?

Thanks for the catch. I assumed that the green lines were the same which obviously they aren't. Most likely the difference is from 4,4 being different from most other hard 8 totals. I am not sure if Norm even included the 4,4 in the hard 8 possibilities but if he did it would represent less than 20% of the total data. Using a CDA there are situations were doubling is far better but a level 1 count has no hope of distinguishing them. Really a side count of 7's or 6,7 block is the only way. I use a level 2 count and the correlation is much different. But for HILO the SBA RA index play for 4,4v6 is double at TC +7 or more. It is not the EV maximizing index but as a part of RA indices allows you to make more money by betting more at the same risk increasing overall profits by up to 3% (when using a family of RA indices including this play) while not increasing RoR. At this point lower risk is worth more than the gain in EV from splitting. RoR is very important when it comes to ebbs and flow of your BR. And if you are still resizing your bets with BR growth you will hit resizing goals faster and with more certainty which should set up an exponential increase in the value of RA indices over time. I think most here are not like KJ, who really can't increase his bets by using RA indices as he is betting as much as he feels the casinos will tolerate. For players like that EV maximizing indices are worth more and use of RA indices with Aversion Ratio greater than 1.