Upon double checking, it would appear that what I wrote above is wrong. Although there are more cards that can help your double than not (i.e., you would win your double more than you'd lose it), the benefit from being able to hit more than once outweighs the value of the double. Sorry about that!
Don
Yes, I was also wondering why you should double in this situation and had in mind that the optimal play for 10 and 11 versus 17-20 is to hit. As a general rule, these optimal hole card strategy tables (e.g. from WizardOfOdds) say that doubling is never favorable against a dealer pat hand. The only situation which might be different is when the dealer has soft 17 and the H17 rule applies. I have not seen any HC strategy table addressing the S17 versus H17 issue, so I am not sure if there are any differences. Also, the surrender rule is often not addressed but I read that surrendering stiff hands versus 10,11 and 19,20 is favorable.
I think correct hole card strategy is:
Double hard 10-11 vs. soft 17 if dealer hits soft 17
Hit hard 10-11 vs. soft 17 if dealer stands on soft 17
Surrender hard 15-18 vs. dealer 19
Surrender hard 12-19 vs. dealer 20
Surrender anything except player BJ vs. dealer BJ
k_c
Due to my simulations done with CVData, your suggestions regarding Double/Hit hard 10-11 vs. soft 17 seem correct,
but the difference in EV is very tiny (just about 0.02 percent, for both S17 and H17 cases). I have not taken a look
at the variance but would suggest that hitting is better due to lower variance, even in the H17 case.
Furthermore, the sim runs indicate that versus dealer 19, not only 15 and above should be surrendered,
but also 13 and 14 at least. Even for 12 the difference is again so small that I would do it, because surrender lowers variance.
On the other hand, I assume that surrendering 17 and above, whilst mathematically advantageous, would look quite suspicious.
These are values I get dealt for 14 vs. 19 from top of full single deck:
These are values I get dealt for 15 vs. 19 from top of full single deck:Code:T-4 vs. T-9 hit: -44.78%, surrender: -50.00% 9-5 vs. T-9 hit: -45.73%, surrender: -50.00% 8-6 vs. T-9 hit: -47.44%, surrender: -50.00% 7-7 vs. T-9 hit: -51.25%, surrender: -50.00% T-4 vs. 8-A hit: -46.19%, surrender: -50.00% 9-5 vs. 8-A hit: -47.16%, surrender: -50.00% 8-6 vs. 8-A hit: -48.88%, surrender: -50.00% 7-7 vs. 8-A hit: -52.78%, surrender: -50.00%
Code:T-5 vs. T-9 hit: -51.98%, surrender: -50.00% 9-6 vs. T-9 hit: -51.63%, surrender: -50.00% 8-7 vs. T-9 hit: -47.46%, surrender: -50.00% T-5 vs. 8-A hit: -53.24%, surrender: -50.00% 9-6 vs. 8-A hit: -52.97%, surrender: -50.00% 8-7 vs. 8-A hit: -48.80%, surrender: -50.00%I'm sick of losing to your pat hands?On the other hand, I assume that surrendering 17 and above, whilst mathematically advantageous, would look quite suspicious.
k_c
I think 6 decks is even more definitve. I am using hit strategy where hard total < 19 is hit versus 19.
Code:T-4 vs. T-9 hit: -47.71%, surrender: -50.00% 9-5 vs. T-9 hit: -47.88%, surrender: -50.00% 8-6 vs. T-9 hit: -48.15%, surrender: -50.00% 7-7 vs. T-9 hit: -48.75%, surrender: -50.00% T-4 vs. 8-A hit: -47.93%, surrender: -50.00% 9-5 vs. 8-A hit: -48.10%, surrender: -50.00% 8-6 vs. 8-A hit: -48.37%, surrender: -50.00% 7-7 vs. 8-A hit: -48.97%, surrender: -50.00%k_cCode:T-5 vs. T-9 hit: -51.97%, surrender: -50.00% 9-6 vs. T-9 hit: -51.92%, surrender: -50.00% 8-7 vs. T-9 hit: -51.27%, surrender: -50.00% T-5 vs. 8-A hit: -52.16%, surrender: -50.00% 9-6 vs. 8-A hit: -52.11%, surrender: -50.00% 8-7 vs. 8-A hit: -51.46%, surrender: -50.00%
Yes, but I forgot to mention that my EV percentages are related to the overall EV, not only to the surrender cases. That is, the probability of the occurrence of the situation is taken into account. I just compared the IBA (Initial Bet Advantage) results produced by the CVData sims of the hc strategy with and without surrendering in those cases.
Bookmarks