Quote Originally Posted by k_c View Post
I do not think that the late surrender decision for 8-8 vs. T using HiLo is necessarily best determined by true count.
I just cherry picked LS versus T from my data. Full set of data in a text file is here: http://www.bjstrat.net/Downloads/HiLo_88.zip

Code:
Count tags {1,-1,-1,-1,-1,-1,0,0,0,1}
2 decks, s17, NDAS, 8-8 vs. T late surrender

Cards Remaining          RC                TC ref
(before up card)
___________________________________________________________

102-98                   Do not surr       --
98-77                    >= +3             +1.61 to +2.03
76-9                     >= +2             +1.39 to +13.00
<9                       No subsets or not enough cards
k_c
Apparently HiLo is not a good count for this hand, because this hand heavily depends on the density of the two cards 2 and 3, not that much on the density of 10s. Based on your calculation, I can roughly estimate a deviation HiLo TC index for this hand, ~+3.

Also, I like the idea of using RC for this hand. For 2-deck games, I have been using RC for insurance decisions with a lot of success. If the RC is +5, +4, +3, +2, and +1, I buy insurance.

For this hand, if the RC is +5, +4, +3, and +2, I surrender.