Dan Provonost of DeepnetTechnology (www.deepnettech.com)reported using his BJ Audit software an advantage of 0.41% at the true count of 1 for the game of 8 decks, S17, DAS, 75% pen (no LS?), full Hi-Lo indices spreading 1-10. The advantage number seemed over rated to me, I always thought that we play a break even game at TC of 1. Any truth about this? (you can read the full report at http://www.bjinsider.com/newsletter_45_bet.shtml)
CVCX gets .32%. The exact value depends on the exact settings.
An "exact" TC of 1 is virtually a break even game. Many sims and software is being set up to show the AVERAGE ev from a TC of "1 to 1.99". As a result you will often see sims over estimating the "true" ev of a specific TC. I was never very fond of this and tend to refrain from raising bets untill the TC is well over the exact tc of 1.00.
By doing this I have decreased variance and increased EV.... my results seem to be doing much better.
Brick
> Dan Provonost of DeepnetTechnology
> (www.deepnettech.com)reported using his BJ
> Audit software an advantage of 0.41% at the
> true count of 1 for the game of 8 decks,
> S17, DAS, 75% pen (no LS?), full Hi-Lo
> indices spreading 1-10. The advantage number
> seemed over rated to me, I always thought
> that we play a break even game at TC of 1.
> Any truth about this? (you can read the full
> report at
> http://www.bjinsider.com/newsletter_45_bet.shtml)
First, sims that have been correctly run do not 'over-estimate' anything. To ignore the math and add in your feelings will reduce your efficiency.
Second, if you are making decisions based on ?exact? TCs instead of TCs that include a range, you are not playing in the manner in which strategies are developed.
Third, CVData allows great variability in exactly how TC is calculated. Not only can you set the resolution to full, quarter, half or exact deck, you can set it differently for different sections of the shoe. You can set flooring, rounding or truncation to customize the calculation that you are discussing. You can set the divisor to cards seen, cards in tray or cards dealt. You can set the method of deck estimation. You can set the TC calculation times to before bet, before first decision, before insurance, or before every close decision.
I?m sorry, but I?m a little tired of posts claiming that sims don?t accurately portray human play. If your sim doesn?t, get another sim.
Now that I have that off my chest, let me get to your assertion. You state An "exact" TC of 1 is virtually a break even game. Well, an ?exact? TC of 1 virtually doesn?t exist(Think about it.) If you are speaking theoretically, I suppose you could plot the value of all exact TCs and try to determine an exact crossover. But, I would bet that the plot won?t be linear. And, the point of your post seems to be that we should play realistically as opposed to theoretically. What matters realistically is the EV at the counts as a human counts them for a particular strategy. For Hi-Lo as published, this would be by truncation and TC>=1 and <2 would count as +1. This isn?t a short-cut taken by sims, it is a sim accurately portraying human play.
norm
Of course I trust the math,otherwise I would have gave up blackjack 10 years ago. There is nothing(I'm aware of) in my post that implies I dont trust the mathematics and bet on gut feelings or hunches. What is more important than trusting the math is to trust and believe in yourself. Without this faith, you have nothing.
There is no knocking of sims and software on my part,I just thought it would be nice to bring it the attention of those who care, that many sims express TC's as (1.00> to <2.00 = TC of 1). It's just something to think about,that's all.
If someone has an RC of 7 with 7 decks remaining which is a TC of 1 not 1.90 and think they almost have a .5% advantage while choosing to bet that % of kelly,then that's their choice,but not mine. My strategy seems to be a bit more conservative in nature. I prefer to wait for a 1.5 TC range before getting hasty.
> First, sims that have been correctly run do
> not 'over-estimate' anything. To ignore the
> math and add in your feelings will reduce
> your efficiency.
> Second, if you are making decisions based on
> ?exact? TCs instead of TCs that include a
> range, you are not playing in the manner in
> which strategies are developed.
> Third, CVData allows great variability in
> exactly how TC is calculated. Not only can
> you set the resolution to full, quarter,
> half or exact deck, you can set it
> differently for different sections of the
> shoe. You can set flooring, rounding or
> truncation to customize the calculation that
> you are discussing. You can set the divisor
> to cards seen, cards in tray or cards dealt.
> You can set the method of deck estimation.
> You can set the TC calculation times to
> before bet, before first decision, before
> insurance, or before every close decision.
> I?m sorry, but I?m a little tired of posts
> claiming that sims don?t accurately portray
> human play. If your sim doesn?t, get another
> sim.
> Now that I have that off my chest, let me
> get to your assertion. You state An
> "exact" TC of 1 is virtually a
> break even game. Well, an ?exact? TC of 1
> virtually doesn?t exist
> If you are speaking theoretically, I suppose
> you could plot the value of all exact TCs
> and try to determine an exact crossover.
> But, I would bet that the plot won?t be
> linear. And, the point of your post seems to
> be that we should play realistically as
> opposed to theoretically. What matters
> realistically is the EV at the counts as a
> human counts them for a particular strategy.
> For Hi-Lo as published, this would be by
> truncation and TC>=1 and norm
Posting Permissions
-*-
Bookmarks