Question for Adde

[Syntax]

What is the rationale behind the changes you suggest to the basic strategy? I don't doubt them, I've been doing well with them, just curious of how you came up with it. Thanks

[CCass]

I will give it a shot.

66 vs 7 - Dealer most likely has 17, splitting 6's will most likely give you 2 16's, the worst hand possible in BJ. Why take 1 marginally bad hand (12) and turn it into 2 very bad hands.

77 vs 8 - Again dealer probably has 18. You split and hit two face cards, and you have 2 17's against an 18 and lose two hands. Better to play the 14 and only lose once.

A7 vs 4/5/6 - You are doubling your bet against the 3 weakest cards the dealer can have up, and will still have 18 after your double card a lot of the time.

Hopefully Adde will agree with my answers.

[MKR]

I'm not Adde, but the changes he suggests get autoplay to use the best strategy for the game as given by the wizard of odds at Microgaming BJ Strategy (http://wizardofodds.com/software/reviewmicrogaming.html). The wizard says the strategy gives the house an .3606% advantage.

MKR

[Syntax]

Thanks guys.

[Adde]

It's nice to have stand-ins. /images/graemlins/smile.gif

Not making these adjustments are not THAT awful EV-wise. I'd say you lose some 0.1% tops maybe (in higher house edge), but it sure feels better to have it all correct when you hand the action over to the autoplayer.

Adde

[Syntax]

I wonder why its not +EV to double on an 8 vs a dealers 6. The rules allow double on any two cards.

[Lazymeatball]

I'm pretty sure most of the basic startegy has been developed or at least perfected by doing very large sample computer simulations sometime in the late 60's or 70's. It goes a liitle bit beyond just assuming that if the dealer is showing a 7 then he most likely has a 10 down. The idea is to calculate the EV of every possible situation and determining which decision has either the greatest positive EV or the smallest negative EV.

For example (I believe) splitting 8's isn't done because starting with an 8 is a good hand, but because a 16 is just so awful, so in this case splitting the 8's shows a smaller loss than playing the 16.

[Homer]

What is the rationale behind the changes you suggest to the basic strategy? I don't doubt them, I've been doing well with them, just curious of how you came up with it. Thanks

You aren't going to like this answer, but the reason is that it has a higher EV.

Some smart dude sat down at a computer years ago and determined the best play mathematically for each possible situation. Other smart dudes verified this, and now we trust them and take it for granted.

If you really wanted to, you could create a program and come up with basic strategy on your own, for peace of mind.

-- Homer

[M.B.E.]

[ QUOTE ]
I wonder why its not +EV to double on an 8 vs a dealers 6. The rules allow double on any two cards.

[/ QUOTE ]
It is +EV to double on an 8 against dealer's 6. However, just hitting has even higher EV. How come? Because, as Homer said, someone worked it out by computer and it just turns out that way.

But if that answer doesn't satisfy you, consider this: about 15% of the time you draw one card, you will catch a deuce or trey. When you are allowed to take a further card after that, you will frequently hit 19, 20, or 21; hence 15% of the time you'd do quite a bit better by just hitting rather than doubling. Also taking one card the best you can hope for is 18 or 19; even with a 6 showing the dealer will sometimes beat that.