Norm,
I can't understand why the Custom Bets give a higher SCORE than the Optimal Bets in this example.
Thanks in advance.
BJFan
No idea. Something to do with the combination of transfering sim info from CVData to CVCX, moving from one to two hands, forcing a min bet, and the very high RoR. I'll have to play around with it.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
The lower bet with a minimal advantage lowers RoR and N0. SCORE is a function of N0. 1,000,000/SCORE = N0. I have to make sure use my terms right. SD is lowered because of both the frequency of that count and the sketchy edge. Using a linear count many of these bets are made when you do not have an advantage. Some are made with a large advantage they average to the win/loss%. Typically the average about 1/3rd in from the bottom of the range with a large cluster of events just below the mean and more scattered results that range much farther from the mean above the average. The net result averages to the mean but most of the time you are betting with a slightly smaller advantage. While far fewer times you are betting with a much higher advantage. The result is you are better off being more conservative at this level to maximize SCORE and minimize N0 and RoR but to maximize EV you want the optimal betting ramp. Personally I would use the custom with that high a RoR. Actually I wouldn't bother playing that game. It sucks.
None of this makes any sense. First, it says that the spread is 1-16, but it isn't: it's 1-32. Next, it says 100% backcounting, yet you are playing negative counts. Then, then are two SCOREs given for the same game! For any given game, there can be only one SCORE. If you change the bets, you change the hourly win rate; you do NOT change the SCORE! If these are C-SCOREs, then say so; if they aren't, then there can't be two SCOREs for the SAME game!
Finally, if you bet optimally, you can't produce a higher SCORE by betting less than optimally. That defies the very definition of the term and makes zero sense. There are several things that are screwed up here, and none of the attempts at an explanation are satisfactory.
Don
Don, the screen was taken from CVData as it appeared, if it says Backcounting 100% then I don't know why, there's no backcounting at all, all rounds are being played. And I agree with you, the spread should be 1-32 but the screen shows 1-16.
Without taking into account the goodness of the game, I guess that there's someting wrong with the algorithm that determines the optimal betting, and that's what I was asking Norm.
BJFan
If you use options to limit the ability of the software to give the optimal spread, it will do what it can, but cannot possibly calculate the SCORE as defined.
Backounting 100% means no backcounting as 100% of hands are played as per the documentation.
There have been many arguments about what spread means when changing number of hands. I have no position on this. CVCX uses per hand instead of total bet to avoid the concept of fractional spreads.
If you send me an export, I can look at it.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
Posting Permissions
Reply With Quote
Bookmarks