
BJFan: SCORE and backcounting
How do I calculate SCORE for a backcounting game?
For a play all game I use the formula SCORE= ExpVal^2/Var*1000000 but in the case of a backcounting game, as I get the ExpVal and Var values for the actually played hands, the formula leads me to very high values, so I guess that there must be a way to adjust it to get reasonable values, perhaps using the values of TotalHands and PlayedHands?
Thanks,
BJFan

Don Schlesinger: Re: SCORE and backcounting
> How do I calculate SCORE for a backcounting game?
> For a play all game I use the formula SCORE=
> ExpVal^2/Var*1000000
I'm sure you realize it, but, as you've written it above, the one million term is in the denominator, and you mean it to be in the numerator.
> but in the case of a backcounting
> game, as I get the ExpVal and Var values for the
> actually played hands, the formula leads me to very
> high values, so I guess that there must be a way to
> adjust it to get reasonable values, perhaps using the
> values of TotalHands and PlayedHands?
Yes, you need to adjust both e.v. and variance for the hands actually played per hour, and not the hands actually seen per hour. You have no e.v. or variance on hands that you watch! :)
Take a look at the chapter 10 charts and look at the numbers. The relationships should become clear to you.
Don

BJFan: Re: SCORE and backcounting
> I'm sure you realize it, but, as you've written it
> above, the one million term is in the denominator, and
> you mean it to be in the numerator.
Yes, Don, it's in the numerator.
> Yes, you need to adjust both e.v. and variance for the
> hands actually played per hour, and not the hands
> actually seen per hour. You have no e.v. or variance
> on hands that you watch! :)
My values of ExpValue and Variance ARE for actually played hands, not for the hands actually observed and if I insert them in the SCORE formula I get very high values for SCORE (i.e. 190 for HiLo 4.5/6 18 wong in +1), so my question is how do I tweak this high value to take into account that I only play a fraction of the hands. I tried multiplying ths SCORE by HandsPlayed / HandsObserved but I'm not happy with this (BJA says the SCORE should be 40).
> Take a look at the chapter 10 charts and look at the
> numbers. The relationships should become clear to you.
That's what I did and I get the right values for play all but for backcounting the SCORE seeked should be about 40 and I get 190 as I told you.
Thanks,
BJFan
> Don

Don Schlesinger: Re: SCORE and backcounting
> Yes, Don, it's in the numerator.
> My values of ExpValue and Variance ARE for actually
> played hands, not for the hands actually observed and
> if I insert them in the SCORE formula I get very high
> values for SCORE (i.e. 190 for HiLo 4.5/6 18 wong in
> +1), so my question is how do I tweak this high value
> to take into account that I only play a fraction of
> the hands.
You just said above that the values are for hands actually played. So, you're observing 100 hands per hour, playing, say, 27 or so, and you have a certain dollar win and dollar standard deviation per 100 hands observed.
> I tried multiplying ths SCORE by
> HandsPlayed / HandsObserved but I'm not happy with
> this (BJA says the SCORE should be 40).
> That's what I did and I get the right values for play
> all but for backcounting the SCORE seeked should be
> about 40 and I get 190 as I told you.
Please supply the hourly win and standard deviation, and the number of hands you're playing per hour.
Don

Don Schlesinger: Re: SCORE and backcounting
What software are you using to generate your win rate and s.d? Shouldn't that same software be producing the SCORE for you?
I checked both with CVCX and BJRM 2002, and both gave the same 40+ SCORE result.
Don

BJFan: Re: SCORE and backcounting
> What software are you using to generate your win rate
> and s.d? Shouldn't that same software be producing the
> SCORE for you?
The software is my own homemade simulator and I'm in the stage of bsckcounting and multiple hands. I'll do some extra tests and if needed I'll ask you again, thanks for your interest.
> I checked both with CVCX and BJRM 2002, and both gave
> the same 40+ SCORE result.
> Don
Posting Permissions
 You may not post new threads
 You may not post replies
 You may not post attachments
 You may not edit your posts

Forum Rules
Bookmarks