Page 1 of 4 123 ... LastLast
Results 1 to 13 of 40

Thread: neptune: trust the math....

  1. #1
    neptune
    Guest

    neptune: trust the math....

    i've greatly enjoyed the discussion engendered by pat saying he has had great success with a 7k bank playing single deck spreading 4:1 and wanting to "go pro".

    most, with the best of intentions, replied that was way too short a bank and any success was pure luck. that was my origional thought, too, til i decided to "run the numbers" and trust the math.

    i simmed the best possible conditions i could imagine: single deck, 1:4 spread, no heat problems, da2, das , nrsa, ns, 60% pen. i also had him wong out @ rc <-6.

    as to risk, in my mind's eye, i thought 2 standard deviations would give one a 97.7% chance of being under or above the curve and a ror of only 2.3%. pretty reasonable, i thought.

    then i spread pat thusly : $100 @ tc +4 or greater, $75 @ tc = +3 , $50 @ tc = +2, $35 @ tc = +1 and $25 @ tc less than +1 .

    i did the sim using (appoligies to norm :-) :-) :-) "only" 40,000,000 hands . the results as follows : wr/100 rounds = $45.41 (round to $45) and sd/100 rounds = $499.89 (round to $500).

    then 2sd's would = $1,000. assuming 100 rounds per hour , and calculating sd for a set number of hours as the sd for one hour times the square root of the number of hours played we get the results for pat;

    after 100 hrs +$14,500 to - $5,500
    after 200 hrs +$23,140 to - $5,140
    after 400 hrs +$38,000 to - $2,000
    after 10,000 hrs +$450,000 to +$350,000 !!!

    all with never slipping below the 7k bank with a 2.3% risk of ruin.

    i was stunned.

    assuming my math assumptions are correct and finding this fine game is possible, pat would have to be "unlucky" to bust....

    thoughts....

  2. #2
    Norm Wattenberger
    Guest

    Norm Wattenberger: A few quick comments


    First, 40,000,000 rounds has no meaning. No meaning. I'm not sure how else to put it. Every number in your post after the number 40,000,000 has no meaning. I'm not trying to be insulting. I'm just saying that I don't bother reading numbers that have no meaning.

    Secondly, you are talking about a game with a SCORE of over 100! I don't believe I have seen such a game outside of BJ pays 2:1. Even in the '70s. Not without getting quickly barred. And where do you go then? (A few bonus exceptions.)

    Thirdly, I'm not sure how you go about regularly wonging a single deck game. It's not like there is another SD game available nearby.

    Fourth, I took the eight minutes needed to run a billion rounds and got a RoR of 7.3%. And that is assuming wonging, single deck with no cover playing green with a spread of 4:1, single-deck with DAS and DA2. Find this game for me and I'll sell my house and put my children into slavery to play

    norm



  3. #3
    Parker
    Guest

    Parker: one very quick comment

    > assuming my math assumptions are correct and
    > finding this fine game is possible,

    It isn't.

  4. #4
    Brick
    Guest

    Brick: trust the fantasy....

    It is amazing how sweet things may appear on paper under the most beautiful conditions imagined by the human mind as we feed the computer information of exactly how we want our picture perfect world to be and let this machine do all the sensational flawless work for us,ah yes,a masterpiece...then we venture out in the real world and soon find out it was nothing but a fantasy. Trust a human brain in action before you trust the math.

    Brick

    i've greatly enjoyed the discussion
    > engendered by pat saying he has had great
    > success with a 7k bank playing single deck
    > spreading 4:1 and wanting to "go
    > pro".

    > most, with the best of intentions, replied
    > that was way too short a bank and any
    > success was pure luck. that was my origional
    > thought, too, til i decided to "run the
    > numbers" and trust the math.

    > i simmed the best possible conditions i
    > could imagine: single deck, 1:4 spread, no
    > heat problems, da2, das , nrsa, ns, 60% pen.
    > i also had him wong out @ rc as to risk,
    > in my mind's eye, i thought 2 standard
    > deviations would give one a 97.7% chance of
    > being under or above the curve and a ror of
    > only 2.3%. pretty reasonable, i thought.

    > then i spread pat thusly : $100 @ tc +4 or
    > greater, $75 @ tc = +3 , $50 @ tc = +2, $35
    > @ tc = +1 and $25 @ tc less than +1 .

    > i did the sim using (appoligies to norm :-)
    > :-) :-) "only" 40,000,000 hands .
    > the results as follows : wr/100 rounds =
    > $45.41 (round to $45) and sd/100 rounds =
    > $499.89 (round to $500).

    > then 2sd's would = $1,000. assuming 100
    > rounds per hour , and calculating sd for a
    > set number of hours as the sd for one hour
    > times the square root of the number of hours
    > played we get the results for pat;

    > after 100 hrs +$14,500 to - $5,500
    > after 200 hrs +$23,140 to - $5,140
    > after 400 hrs +$38,000 to - $2,000
    > after 10,000 hrs +$450,000 to +$350,000 !!!

    > all with never slipping below the 7k bank
    > with a 2.3% risk of ruin.

    > i was stunned.

    > assuming my math assumptions are correct and
    > finding this fine game is possible, pat
    > would have to be "unlucky" to
    > bust....

    > thoughts....

  5. #5
    BJ in the Box
    Guest

    BJ in the Box: Long vs Short terms: Challenge to Norm

    > First, 40,000,000 rounds has no meaning. No
    > meaning. I'm not sure how else to put it.

    There seems to be some pervasive misconception regarding long term vs short term on this board.

    Let's suppose Blackjack play can be mathematically modeled as a "biased random walk" (as is typically done in most theoretical studies):
    For every one unit of bet, the average return will be d and the variance will be V. (d will be typically around 0.005 (I suppose it can be up to 0.01 for some "experts" under very favorably conditions), whereas V is much larger number, of order 1 (I presume it'll probably be slightly bigger than 1.0).)

    Mathematically, the "cross-over" between the long and short terms occurs when the "bias/drift" effect is comparable to that of "fluctuation":
    N_c ~ V / d^2
    where N_c is defined to be a critical number of hands under which the fluctuation will most likely "overshadow" the average edge the player (is supposed to) have.

    If we plug in the numbers cited earlier, N_c is around 10,000 to 100,000. If we suppose the advantage player can play 100 hands per hour on average, this number translates to around 100 to 1000 hours, which are the numbers quoted by Neptune (even though he might have been using a very large value for d). (Note, however, that N_c is relatively insenstive to larger values of d which invalidates some of the objections raised against Neptune's numbers. On the other hand, N_c diverges when d approaches zero.)

    Now, Norm, I challenge you: What is the rationale behind your "no-meaning-argument" for 40 million(!) hands? I'm really curious.

    ~Black Jack in the Box.

  6. #6
    Norm Wattenberger
    Guest

    Norm Wattenberger: Re: trust the fantasy....

    > Trust a
    > human brain in action before you trust the
    > math.

    Trust the math - but get the variables right.

  7. #7
    Norm Wattenberger
    Guest

    Norm Wattenberger: Re: Long vs Short terms: Challenge to Norm


    In this case, neptune was taking about a d of .017. An absurdly high number. The standard deviation per 100 hands was a very large 100.7.

    In single-deck card counting we are dealing with a very small edge and a large variance. My complaint is the use of mathematics with inaccuare numbers. We have the ability to generate accurate numbers in a couple of minutes. Why even look at inaccurate numbers?




  8. #8
    Brick
    Guest

    Brick: you missed the point

    > Trust the math - but get the variables
    > right.

    My point was even with a perfect flawless ideal sim of dreamland conditions that a computer has performed for us does not mean we ean achieve these results in the real world. My human brain already knows this through my experiences. Trust me.

    Brick

  9. #9
    Brick
    Guest

    Brick: Re: Long vs Short terms: Challenge to Norm

    Norm, I also dont agree with your statement of meaningless. When we are trying to find an exact index number or variable to the nearest.0000001% then 40 million may not be enough to get accurate results. However if we are trying to get an ev approximation to the nearest .001% then 40 million hands have a very significant value. The reason neptune was way off on his sims was not because he ran 40 million hands. It was due to the variables.

    Brick

    > In this case, neptune was taking about a d
    > of .017. An absurdly high number. The
    > standard deviation per 100 hands was a very
    > large 100.7.

    > In single-deck card counting we are dealing
    > with a very small edge and a large variance.
    > My complaint is the use of mathematics with
    > inaccuare numbers. We have the ability to
    > generate accurate numbers in a couple of
    > minutes. Why even look at inaccurate
    > numbers?

  10. #10
    neptune
    Guest

    neptune: Re: Long vs Short terms: Challenge to Norm

    > In this case, neptune was taking about a d
    > of .017. An absurdly high number. The
    > standard deviation per 100 hands was a very
    > large 100.7.

    > In single-deck card counting we are dealing
    > with a very small edge and a large variance.
    > My complaint is the use of mathematics with
    > inaccuare numbers. We have the ability to
    > generate accurate numbers in a couple of
    > minutes. Why even look at inaccurate
    > numbers?

    >>>>LD

    hi norm,

    first of all, thanks for your insightful/inciting remarks :-) :-) :-)

    i'm not a math genius, but although i cannot spell well.... i do read ok.

    i am looking at wong's bj analyzer right now as i re ran the sim that i spoke of yesterday and guess what ... in a stunning turn of events the numbers @108 million rounds agree with the numbers @ 4o million rounds with < 1 % :

    wr/100 rounds = $45.69 ; se = $0.48
    sd/100 rounds = $499.93

    this was again : betting $100 @ tc +4 or better, $75 @ tc +3, $50@ tc+2 , $35 @ tc +1 and $25 the rest of the time down in count.

    when you ran your 500 million rounds, were your results different ???

    by the way, when i get back from racing tonight , i'll post what wong's bja calls @ 500 million rounds and we'll see if its close to the 40,000,000 figure. if it isn't, i'll happily munch on (lance) humble pie :-) :-) :-)

    anyone want any action on this :-) :-) :-)

    but, i know, if the figures for 500 million rounds coincides with the "meaningless" 40 million rounds, we'll just be "lucky" again.

    i guess i'm just one of those infinite number of monkey's sitting behind one of those infinite number of typewritters and one of them randomly comes up with "hamlet". :-) :-) :-)

  11. #11
    neptune
    Guest

    neptune: Re: A few quick comments

    > First, 40,000,000 rounds has no meaning. No
    > meaning. I'm not sure how else to put it.
    > Every number in your post after the number
    > 40,000,000 has no meaning. I'm not trying to
    > be insulting. I'm just saying that I don't
    > bother reading numbers that have no meaning.

    > norm

    >>>>>neptune

    hello norm,

    i realize you are not trying to be insulting, just giving your opinion.

    i do have a story for you, though, about a fellow named mr van winkle.

    mr. van winkle loves blackjack. he loves it so much in june of 1958 he sits down at a 6 deck table at the flamingo (did they exist in 1958 ???)

    much like his brother rip who could sleep for years, this van winkle was a real "headstone" ; he didn't get up to eat, didn't get up to go to the bathroom... he just kept playing 100 rounds per hour every hour without fail.

    eisenhauer, nixon, ford, carter, reagan, bush , clinton and bush again came and went, but mr van winkle kept playing.

    yesterday, he got up and stepped away from the table .

    the first thing he did was call norm, to evaluate his play. he told norm he played such and such a way and had norm sim it to 500 million rounds.

    the question was , did norm tell him is lifetime of continueous play (40,000,000) rounds was "meaningless" as compared to the results of his 500,000,000 sim and had no statistical corrolation ???

    if that's the case norm, then you fellas are selling snake oil. if 40,000,000 rounds of play are meaningless, what's the point of simulating at all if a lifetime of continueous play (40,000,000 rounds)doesn't reasonably approximate the "true" outcome (divined by 500,000,000 rounds) :-) :-) :-)

  12. #12
    DD'
    Guest

    DD': n0 is the term normally used in Bj circles (nt) *NM*


  13. #13
    Norm Wattenberger
    Guest

    Norm Wattenberger: Yes and

    We must be careful not to confuse two numbers. n0 and the number of rounds of simulation required to accurately determine the values used in calculating n0. A far higher number.

Page 1 of 4 123 ... LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

About Blackjack: The Forum

BJTF is an advantage player site based on the principles of comity. That is, civil and considerate behavior for the mutual benefit of all involved. The goal of advantage play is the legal extraction of funds from gaming establishments by gaining a mathematic advantage and developing the skills required to use that advantage. To maximize our success, it is important to understand that we are all on the same side. Personal conflicts simply get in the way of our goals.