What about this guy?
- Thread starter Warlord
- Start date
So this guy is saying that he has a super edge just because he decides bet size and modifies basic strategy according to the exact composition of the remaining cards.
http://www.bewersdorff-online.de/black-jack/
This calculator does what this guy claims he can do. Is this edge that much more super than normal card counting with simple basic strategy modifiations (according to the true count)? Is it worth all this mental harshness to attain this extra edge? Can this guy calculate the exact composition and also have in mind the best decision that each composition suggests? I think it is more probable that he is simply lying.
Beware of the above calculator. Someone told me that it calculates wrongly the edge of the player. Can someone confirm whether it gives the correct decision tables though?
http://www.bewersdorff-online.de/black-jack/
This calculator does what this guy claims he can do. Is this edge that much more super than normal card counting with simple basic strategy modifiations (according to the true count)? Is it worth all this mental harshness to attain this extra edge? Can this guy calculate the exact composition and also have in mind the best decision that each composition suggests? I think it is more probable that he is simply lying.
Beware of the above calculator. Someone told me that it calculates wrongly the edge of the player. Can someone confirm whether it gives the correct decision tables though?
Last edited:
OK I see what the calculation is doing. It is based on a fixed probability of drawing a card based upon initial shoe composition. This is a quote from the site in the link -
"The computation is an approximation based on the assumpation that the actual probability distribution of remaining cards is fixed during the following game. All results are shown depending on the dealer's up card and in total."
In other words effects of removal are not included in the computation. For example the same calculation is done for 1 full deck, 2 full decks, 3 full decks, ......, any number of full decks. Likewise the same calculation would be done for 1 five removed from a single deck and 2 fives removed from a double deck. This is basically an inifinite shoe computation. I think the overall EV for S17, NDAS, no resplit infinite full shoe is about -.69% as compared to the web program's -.77%, so it's close but still not right.
If you want something that uses all of the information to get exact values, check out my programs. The only values that aren't necessarily optimal are split values but they are still exact calculations and accurate approximations of optimal. In order to compute splits in a reasonable amount of time I use a fixed strategy algorithm. In my optimal comp dependent program for split EVs I have 2 options: 1) use optimal strategy of non-split hands and apply it post-split and 2) use optimal strategy on the first hand of a split and apply it to all subsequent split hands. In computing splits in my total dependent program, the hard and soft standing numbers are fixed. For example when 8-8 v 10 is split the hard standing number = 17 and soft standing number is 19. This is basic strategy and the present version uses basic strategy for overall computations. You can manually change the hard/soft standing strategies for individual hands, though, and you will find that you are slightly better off standing on hard 16 after splitting 8-8 v 10 for any number of decks up to program's limit of 8. The comp dependent program's deck range is 1 to 41,297,762 so a virtual infinite deck could be created by inputting a very large number of decks. For a very large number of decks, the removal of a relatively small number of cards has virtually no effect on calculated values and optimal comp dependent strategy becomes the same as total dependent strategy. Infinite deck strategy for 8-8 v 10 is to hit to hard 17/soft 19 after splitting regardless of whether the computation is optimal or total dependent.
As for your question as to whether exact composition could realistically be used at the table, "No way Josê.
"
"The computation is an approximation based on the assumpation that the actual probability distribution of remaining cards is fixed during the following game. All results are shown depending on the dealer's up card and in total."
In other words effects of removal are not included in the computation. For example the same calculation is done for 1 full deck, 2 full decks, 3 full decks, ......, any number of full decks. Likewise the same calculation would be done for 1 five removed from a single deck and 2 fives removed from a double deck. This is basically an inifinite shoe computation. I think the overall EV for S17, NDAS, no resplit infinite full shoe is about -.69% as compared to the web program's -.77%, so it's close but still not right.
If you want something that uses all of the information to get exact values, check out my programs. The only values that aren't necessarily optimal are split values but they are still exact calculations and accurate approximations of optimal. In order to compute splits in a reasonable amount of time I use a fixed strategy algorithm. In my optimal comp dependent program for split EVs I have 2 options: 1) use optimal strategy of non-split hands and apply it post-split and 2) use optimal strategy on the first hand of a split and apply it to all subsequent split hands. In computing splits in my total dependent program, the hard and soft standing numbers are fixed. For example when 8-8 v 10 is split the hard standing number = 17 and soft standing number is 19. This is basic strategy and the present version uses basic strategy for overall computations. You can manually change the hard/soft standing strategies for individual hands, though, and you will find that you are slightly better off standing on hard 16 after splitting 8-8 v 10 for any number of decks up to program's limit of 8. The comp dependent program's deck range is 1 to 41,297,762 so a virtual infinite deck could be created by inputting a very large number of decks. For a very large number of decks, the removal of a relatively small number of cards has virtually no effect on calculated values and optimal comp dependent strategy becomes the same as total dependent strategy. Infinite deck strategy for 8-8 v 10 is to hit to hard 17/soft 19 after splitting regardless of whether the computation is optimal or total dependent.
As for your question as to whether exact composition could realistically be used at the table, "No way Josê.
"
Last edited:
Hitting 16 vs 6 is not so strange. You must hit at less than -13 TC. And if you know the exact composition, you should hit in cases when the TC is e.g. -12. But ... this should happen at a low money bet of his, and not a high money bet. How much had he wagered at this bet? Can someone see?
Lol, I just saw it, it was a 300 pounds bet, thus this should not happen. 300 pounds means that the TC was positive, and not negative. Perhaps he is implying shuffle tracking as he said "I had a good sense of the deck by now". But hey, 16vs 6 with a positive count, the dealer shuffles the cards you know!
Lol, I just saw it, it was a 300 pounds bet, thus this should not happen. 300 pounds means that the TC was positive, and not negative. Perhaps he is implying shuffle tracking as he said "I had a good sense of the deck by now". But hey, 16vs 6 with a positive count, the dealer shuffles the cards you know!
Last edited:
So, kc, it is better to use that calculator than counting with the halves system and use the basic strategy modifications according to the true count, right? (however these modifications are based on the hi-low system and not the halves, lol).
Any idea how much extra edge I gain by using this calculator instead of counting the way I do, regarding 8 decks, 50% penetration? (lol, yes, I am counting at live dealer online casinos).
Any idea how much extra edge I gain by using this calculator instead of counting the way I do, regarding 8 decks, 50% penetration? (lol, yes, I am counting at live dealer online casinos).
I guess you're looking to use a computer to maximize EV while you play. In a brick and mortar casino this is illegal. I'm not sure about online but I don't see how they would know if you were using a background program. 8 decks, 50% pen probably has pretty limited opportunities, though. The best that can be done is to know optimal play EV before each and every round and to make the optimal play each and every time. A program that can do this is better than any counting system. I don't know how well the calculator in your link would accomplish this. I wouldn't know how well a perfect program would expect to do without doing a sim and by the same token I think you'd need to sim how you use your calculator to get a picture of what to expect.
I do not understand why they are all standing? do they think they are on tv and are at the world series poker and all of are all in? another thing how can u memorize for six decks which is impossible. That is what the majority of casinos in this country play. I am not sitting down at a bj table while everyone else is standing. :laugh:
In a shoe game, 80% of the profit comes from knowing when to bet big or small. HiLo has a betting correlation of over 98%, meaning a simple 1 level count is pretty damm close to perfect for betting.
The I18 captures about 80% of the profit possible from deviating from basic strat, even using an inefficent Playing count like HiLo. Additional deviations earn you less and less rapidly.
Knowing the exact composition and making some wild play and gaining an extra .000001% ? How much is that really worth? How many hours would you have to play to justify the cost of the computer, let alone the hours of practice to do it mentally?
Computer perfect play vs a good counter -- 2 cents an hour, knock yourself out.
The I18 captures about 80% of the profit possible from deviating from basic strat, even using an inefficent Playing count like HiLo. Additional deviations earn you less and less rapidly.
Knowing the exact composition and making some wild play and gaining an extra .000001% ? How much is that really worth? How many hours would you have to play to justify the cost of the computer, let alone the hours of practice to do it mentally?
Computer perfect play vs a good counter -- 2 cents an hour, knock yourself out.
Computer perfect play was the genesis of why blackjack became an advantage opportunity in the first place mainly thanks to Thorp. Now most seem to take this kind of thing for granted and go so far as to intimate that it is foolish to approach the game from this point of view Basic attitude is that virtually everything is known and the only way to answer anything is by brute force sim (no disrespect for sims intended,) but the parameters of what is being simmed generally derive from what a computer calculation has determined to be relevant.
(Dead link: http://www.bjstrat.net/odd_removed.html) _Here's an example of a computation where all of the odd cards in an 8 deck shoe have been removed._ A perfect play player has a 12.67% advantage. I'm sure your response will be that the chances of ever encountering this situation are nil even if you were capable of recognizing it and playing it perfectly so move on, and you will be right. But is there some system that counts odd cards versus even cards that is workable? I don't know the answer to that. The point is that most just want to learn the game and have questions about it and having a tool that can truthfully show any specific situation without delay is a valuable and possibly time saving asset in finding a direction. In conclusion I would like to say .... there is no conclusion. :devil:
Basic attitude is that virtually everything is known and the only way to answer anything is by brute force sim (no disrespect for sims intended,) but the parameters of what is being simmed generally derive from what a computer calculation has determined to be relevant.(Dead link: http://www.bjstrat.net/odd_removed.html) _Here's an example of a computation where all of the odd cards in an 8 deck shoe have been removed._ A perfect play player has a 12.67% advantage. I'm sure your response will be that the chances of ever encountering this situation are nil even if you were capable of recognizing it and playing it perfectly so move on, and you will be right. But is there some system that counts odd cards versus even cards that is workable? I don't know the answer to that. The point is that most just want to learn the game and have questions about it and having a tool that can truthfully show any specific situation without delay is a valuable and possibly time saving asset in finding a direction. In conclusion I would like to say .... there is no conclusion. :devil: