Dear members,
I used CVIndex to generate the Late Surrender indicies for 16 vs 8 and 15 vs 8 for the Red-7 strategy for 6-deck and it gave a higher index to surrender at for the 16 vs 8 than for 15 vs 8. I found that strange. Anyone know why? Thanks.
From the configuration, go to Tools-Export, create an export file, and e-mail to [email protected].
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
I am not sure why but observations are:
1) The 6 is a bust card for 16 but counted as a low card. 1.5 out of 6.5 low cards are bust cards for the 16 and 0.5 out of 6.5 low cards are bust cards for the 15. This should work in the other direction than a required explanation would need.
2) There are no 8's in a hand of 16. The 8 is the worst card for the dealer to have in the hole. Sometimes composition dependence can occur. This is more common with multi-card indices which really aren't an issue for regular BJ and never an issue for surrender. The possible hands for 16 are 80% T,6 and 20% 9,7 (8,8 is a pair of eights not a 16). They are all bust cards for your 16. The 9 and T would make the dealer's hand as a hole card as a 17 or 18, the others would give the dealer a 14 or 15 to hit. Being sure 1 of these cards is out of play because it is on the table has a small effect on the possibilities beyond the count. The possible hands for 15 are 66.7% T,5 (1 bust card for you and 1 gives you 20. For the dealer T makes a 18 and 5 has the dealer hitting 13) and 16.7% 9,6 (1 bust card and a card that makes you 21. For the dealer there is one less card that will make a 17 or have him hitting 14) and 16.7% 8,7 (Both cards bust you and have the dealer hitting 15 or 16).
3) If you use integer indices one might be barely under the integer while the other just makes it. i.e. They almost have the same decimal index value but have different integer index values.
4) My bet is some kind of error and your instinct is right.
Yup. As Norm mentioned, 8,8 is, unfortunately, included in such calcs, just as 7,7 is included for surrendering 14, unless you expressly exclude it, which you shouldn't have to do. Since, depending on rules (DAS or not) the surrender number for 8,8 vs. 8 is much, much higher than for the other 16s (+4), including 8,8 skews the overall index to the upside.Originally Posted by Wino
Don
This is in the documentation. CVData is designed for flexibility. You are allowed to simulate the situation in either manner as both situations can occur. Strategy generation should not be considered by newcomers. Most new strategy efforts are designed to deal with the increasing complexity of new games and the increasing difficulty in environment that APs experience.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
Posting Permissions
-*-
Reply With Quote
Bookmarks