Maybe someone can explain this one to me. . .

[SPX]

It seems apparent to me that we don't like it when the dealer hits Soft 17 because he/she stands a greater chance of improving his hand that he has of busting.

So why, when a dealer hits Soft 17, do we double on 11?

I'm sure there's some reasoning behind it but at the moment it seems to defy logic.

[KenSmith]

First, I'll clarify your question... You're talking about when you have 11 and the dealer has an Ace up. Basic strategy is to just hit if the dealer stands on soft 17, but to double if the dealer hits soft 17.

It does seem backwards, since the dealer makes stronger hands with the H17 rule. However, he does bust more often. (In S17, an Ace busts 16.70% of the time. In H17, an Ace busts 20.13%.)

For whatever reason, the increased bust percentage is apparently more important than the strength of the average made dealer hand, at least for this particular decision.

Just for reference, here are the complete outcome results for an Ace up in a 6-deck game: (after peeking, and no blackjack)

S17: Bust 16.70%, 17: 18.81%, 18: 18.92%, 19: 18.89%, 20: 18.94%, 21: 7.74%

H17: Bust 20.13%, 17: 8.28%, 18: 20.66%, 19: 20.68%, 20: 20.73%, 21: 9.53%

[SPX]

Originally Posted by KenSmith

First, I'll clarify your question... You're talking about when you have 11 and the dealer has an Ace up. Basic strategy is to just hit if the dealer stands on soft 17, but to double if the dealer hits soft 17.

It does seem backwards, since the dealer makes stronger hands with the H17 rule. However, he does bust more often. (In S17, an Ace busts 16.70% of the time. In H17, an Ace busts 20.13%.)

For whatever reason, the increased bust percentage is apparently more important than the strength of the average made dealer hand, at least for this particular decision.

Just for reference, here are the complete outcome results for an Ace up in a 6-deck game: (after peeking, and no blackjack)

S17: Bust 16.70%, 17: 18.81%, 18: 18.92%, 19: 18.89%, 20: 18.94%, 21: 7.74%

H17: Bust 20.13%, 17: 8.28%, 18: 20.66%, 19: 20.68%, 20: 20.73%, 21: 9.53%

Yeah, after looking back over it I guess I didn't state my question too well . . . but you got it for me.

So it seems to me that this must be a very marginal play. I used to mostly play in MS where it was 6D S17, but now play in Wendover where it's single deck H17 and generally find myself weary of making plays that deviate from the BS that I have been playing for so long.

In fact, I'll admit that this is one play that I've just ignored. I got an 11, the dealer shows an Ace? Hit it. Maybe I'll start doubling but it will take some time to get over the psychological element of doing so.

[EasyRhino]

According to the Wizard:

http://wizardofodds.com/blackjack/ap...9-1dh17r4.html

If you get, say, 6,5 vs A, and don't double, it costs you about 9% (14% v 23% advantage). That's a pretty big chunk (For comparison, 12 v 2 is around 0-6%, 16v10 is even closer).

It's worth noting that the extra return reflects the results of doubling your bet. However, it also means that doubling only slightly reduces your odds of winning that hand... while doing so with twice as much money.