Strategy advantage
  • How do you calculate “Advantage”?
    If the average BS player has a -.5% (minus.005) house advantage does that mean that they will lose $5 for every $1,000 bet? Thus betting flat $10 per hand for 100 hands with a -.5% advantage will result in a loss of $5. (Lets not consider variance) Can I consider that at -.5% advantage?
    What advantage % is the typical card counter going to get? In Fred Renzey book he indicates that a ¾ % advantage (plus.0075) or a swing of 1 ¼ % requires great card counting skills.
  • sage01 said:Thus betting flat $10 per hand for 100 hands with a -.5% advantage will result in a loss of $5.

    Right.

    sage01 said:What advantage % is the typical card counter going to get?

    Generally a card counter can get a 1-2% advantage, although it will depend on his style of play. Most card counters will use other techniques as well in order to get a bigger advantage.

    -Sonny-
  • If I play 100 hands and flat bet $10, total bets $1,000 and win $10 can I say I had a 1% advantage in that game?
    Can I then say (to an expert like yourself) that my system (or skill) gave me a 1% advantage?
  • You can say you ended that session AT a 1% advantage.But thats a far cry from playing
    WITH a 1% advantage.
    You are looking at X amount of hands and dividing your winnings by that amount. That is very different than knowing you have a 1% advantage before every hand is played.
  • I agree, 100 hands do not prove anything.
    How about 1,000 hands, or 5,000 or 10,000.
    If you play, say 3 hours a week, maybe 40 weeks a year, that's 120 hours a year, using 80 hands per hour that's about 10,000 hands a year.
    If you won 2% of the amount bet during that year, can you say you play at a 2% advantage? (Not sure what % of your bankroll gain that represents, probably much more than 2%)
    What is the "long range" that gives some comfort the numbers are correct?
    When can you say you play at an X% advantage
  • When the math tells you so,in advance.
  • NYB said:
    When the math tells you so,in advance.


    That is true, but only to a certain extent. The math will tell you the hypothetical advantage you have under a set of conditions such as rules, spread, etc. played exactly under that set of conditions......BUT, no one can play to that "optimal".

    I get such a kick out of our resident "jammie boy" who likes to think he is an advantage player even though he is still a newbie and only plays 20-30 hours a year, but he posts about scouting the casino, then going to his room and checking CVBX (or whatever it is) for spread, risk or ruin, etc., etc......Totally ludicrous. As if anyone could play to those specific parameters......That guy probably plays "even" with the house, at best. Certainly not to any advantage.......And he knows the "math".

    See what I mean? The math will tell you one thing, but only a rare few will ever even come close to it..........And there are dozens of reasons that are so obvious I won't even state them.......But I can state this: I played for over 40 years and counted for over 30 and I figure I played to only about 1% edge over that period .......Sure, I had the "knowledge/math" to play higher than that, but conditions didn't allow it.

    Regards.....Grifter
  • ...and there you have it.
  • Grifter said:
    That is true, but only to a certain extent. The math will tell you the hypothetical advantage you have under a set of conditions such as rules, spread, etc. played exactly under that set of conditions......BUT, no one can play to that "optimal".

    I get such a kick out of our resident "jammie boy" who likes to think he is an advantage player even though he is still a newbie and only plays 20-30 hours a year, but he posts about scouting the casino, then going to his room and checking CVBX (or whatever it is) for spread, risk or ruin, etc., etc......Totally ludicrous. As if anyone could play to those specific parameters......That guy probably plays "even" with the house, at best. Certainly not to any advantage.......And he knows the "math".

    See what I mean? The math will tell you one thing, but only a rare few will ever even come close to it..........And there are dozens of reasons that are so obvious I won't even state them.......But I can state this: I played for over 40 years and counted for over 30 and I figure I played to only about 1% edge over that period .......Sure, I had the "knowledge/math" to play higher than that, but conditions didn't allow it.

    Regards.....Grifter


    Grifter, you might be an old pro, but you have a _lot_ to learn. The program is CVCX. And I absolutely _can_ play to those rules, and I absolutely can sit at the table and not make a mistake for a couple of hours. I would tell you how I know that, but then that just gives you more slack to make such silly comments.

    If you can't use a computer to help your advantage, that is simply something _you_ are missing out on.

    This is probably more about your not carefully reading what I write, as opposed to trying to be a wiseacre I suppose. I've never said I go back to my room for _every_ table I play at. I have said that if I decide to play at a 6D shoe, that is H17 as opposed to S17, or one with LS vs one without LS, then absolutely I go check out my bet ramp before playing. Only makes good sense to do so. I don't go adjust my ramp if the pen changes by a half-deck or something. But when given the choice of playing at a SD game, A DD game, or a shoe game, all of which have significant rule variations across the casinos I visit, I try to maximize my advantage by taking advantage of easy-to-use tools that are available.

    I've told you previously how much I play, without exaggeration. I have probably had a 20-30 hour year or two. More commonly I play double that or a bit more, only being 3-4 hours from good games in Mississippi, not to mention our annual week+ trip to Vegas.

    If that means I don't know what I am talking about, so be it. But it certainly looks bad for the home team when the home team keeps poo-pooing the very idea of using a computer to play better. This is the computer age. If you choose to not join it, fine. But those that do will be better off, play with a bigger advantage, and play much more accurately with good drill software to help them out. If you choose to ignore technology, more power to you. But don't poo-poo those that choose to use whatever help they can get. And I can tell you I would rather have CVBJ watching my play for a couple of hours as opposed to a human any day of the week. There is no comparison in the accuracy of the two "observers".

    As to the "played for 40 years" nonsense, that is hardly convincing of anything. My dad built houses for 40 years. The process today is _far_ superior, thanks to technology (electric saws, electric/pneumatic nail drivers, laser levels and measurment tools, etc.) Being stuck 40 years ago is not necessarily a good thing...

    think about it...

    I learned to count the old-fashioned way. I now use the modern technology way. Modern is _definitely_ better. But you have to try it to see why, rather than just spouting nonsense... disdaining it without ever having tried it is not the mark of a wise person...

    Just my $.02...

    BTW 1% is a good number, although stating that by itself is a bit misleading. The rules can change that. Poor penetration can change that. If you are not capable of watching the payouts carefully, dealer error can change that. Alcohol can change that. Etc. If you can find a game where you can track the shuffle, the number goes up. If you find a dealer that reveals his hole card frequently, the number goes up. If you find the right promotion going on (matchplay, snappers pay 3-1 or whatever) the number goes up. But most don't appreciate what that 1% means. Long-term it does mean you win $1 for every $100 bet. But realistically, it means nothing for a weekend, because you can lose $100 for every $100 bet short-term, or you can win that amount. The variance is much more of an issue than the advantage to most casual players, because it is the variance that convinces them that counting doesn't work, or that basic strategy or a progression is enough to win...
  • stainless steel rat said:
    The program is CVCX. And I absolutely _can_ play to those rules, and I absolutely can sit at the table and not make a mistake for a couple of hours..............


    SSR - Exacly when was it you fell off the turnip truck?

    So you play absolutely perfect and follow the exact betting ramp that CVCX told you to, raising and lowering your bets in exact lockstep with the count to get the "optimal"?.....Sure you do, and the Pope is a Baptist.

    Or if you do actually that, then you are incredibly naive or incredibly misinformed about advantage play blackjack.

    Regards.....Grifter
  • uh o.. battle of the titans lmao
  • Grifter said:
    SSR - Exacly when was it you fell off the turnip truck?

    So you play absolutely perfect and follow the exact betting ramp that CVCX told you to, raising and lowering your bets in exact lockstep with the count to get the "optimal"?.....Sure you do, and the Pope is a Baptist.

    Or if you do actually that, then you are incredibly naive or incredibly misinformed about advantage play blackjack.

    Regards.....Grifter


    Then I guess the Pope's religion is a bit different than what everyone else believes...

    You do know that CVCX can be told to give "rational bet ramps" right? So that you don't go from $3.51 to $8.44 if the count goes from +1 to +2???

    As I said, technology works, if you use it...
  • Casino Royale said:
    uh o.. battle of the titans lmao


    not really. Just that some would prefer to live in a glass house and throw stones, which is foolish. For those that don't use CVCX/CVData and CVBJ, you should give 'em a try to see what you are missing. It's really nice to either ask "what is my advantage and the optimal betting ramp required to reach that?" or "what is my advantage and hourly win rate if I use _this_ betting ramp instead of the optimal ramp?" (for example, in a DD game, you might want a simple $5 at <= 0, $10 at 1, $20 at +2, and $40 at +3. It isn't optimal, but it is easy to do. how much does it hurt your bottom line? wouldn't that be nice to know rather than just guessing?<br />
    It is pretty interesting to see how much a subtle change to your play can erase that +1% advantage in a heartbeat... Eventually you will remember the results and not need to look 'em up unless you have a new game in town. But for the first couple of years of play, the questions cvcx can answer can save you a ton of money and improve your play as a result...

    This is for those interested in moving into the technology-arena. The rest can continue to use handsaws and claw hammers and work much harder for the same return, and continue to guess what their bottom line will be...
  • Not taking sides here, but:

    Automation for the "SAKE OF AUTOMATION" is seldom a good idea and generally is an overkill when it comes to an environment with ever changing conditions (too many variables). Thus, the casino and the need to apply common sense to your prototype/model/pilot/strategy. I think this is Grifter's point and it applies to a real world approach to the many differences from one casino to another or even in the same store. Game selection that best fits your overall abilities is a very big consideration.

    However, automation used for comparative analysis, learning and/or as a teaching tool is a very good thing because it does not involve immediate risk. It should be used in a controlled environment to help to arrive at a strategy. I think this should be SSR'S point, but with qualifications as to where and how to use it. In a casino environment, many adjustments are necessary because of rule differences, games available, number of players, etc,etc. Casino management is after your money in many different ways.

    Theory alone will not suffice without the art.
  • i agree with SSR, i use CVBJ, its an amazing tool
  • Somehow this thread's focus has shifted from optimal play to the viability of computers. I'm going to address the computers with this post, then get the thread back on track with the second post.

    Casino Royale.....Yes, CVBJ is a great tool. I was one of the first to buy it and use it back in the mid-90's.

    Everyone..... Yes, computers are fantastic tools for today's generation to learn how play blackjack, and if you are just starting out by all means purchase CVBJ and CVCX. Just guessing, I would imagine you can learn as much in a year as it took my generation five/six years to learn.

    SSR.... Your inexperience is showing again, so here is your factoid for the day. I agree that CVCX is probably a great learning tool for you......but don't think you have discovered Wonder Bread with it, and don't think that just because my generation doesn't use it as much that we are working with handsaws and claw hammers. CVCX didn't invent the wheel, SSR, they just improved on it. Way back in the mid-80's Snyder and RGE published books that did for my generation essentially what CVCX is doing for you.......Those books gave 350 different combinations of ramps with various deck penetrations, and expected WR's, standard deviations, etc.........we memorized the ones we used regularly and were comfortable with, and kept the rest in Day-Timers to refer to if needed.

    End of Computer Discussion In This Thread......If anyone wants to respond, do so in a new thread...........My next post (later this morning) will get this thread back on track regarding "Optimal Play".
  • To Get This Thread Back On Track........The original question was about playing to a certain "edge" (EV, Expected Value).

    The EV published for any method is the "optimal" for that method at a given ramp, spread, rules, indices, etc., etc. To obtain that optimal one would have to play absolutely perfect....Never making a counting mistake, never making a playing mistake including indices, and ramping the bets in exact lockstep with the count.

    My original statement, and I say it again..... Nobody can play a method to the "optimal" (in real casino play).

    Ray, you are exactly right. I am trying to describe the "real world" of playing blackjack. Sure, you could physically play to the optimal, but how long would you last before you were 86'd?.....So physically possible, yes; but "real world" possible, definitely no.

    Regards.....Grifter
  • Grifter said:
    To Get This Thread Back On Track........The original question was about playing to a certain "edge" (EV, Expected Value).

    The EV published for any method is the "optimal" for that method at a given ramp, spread, rules, indices, etc., etc. To obtain that optimal one would have to play absolutely perfect....Never making a counting mistake, never making a playing mistake including indices, and ramping the bets in exact lockstep with the count.

    My original statement, and I say it again..... Nobody can play a method to the "optimal" (in real casino play).

    Ray, you are exactly right. I am trying to describe the "real world" of playing blackjack. Sure, you could physically play to the optimal, but how long would you last before you were 86'd?.....So physically possible, yes; but "real world" possible, definitely no.

    Regards.....Grifter


    Although I agree that optimal play as described by computer sims is not quite the same as optimal casino play, I do think with team play you can be very close to what is computer optimal. Such as with spotters that flat bet with no need for bet sizing, leaving it just a simple counting game for them where mistakes are much less likely to occur. Then calling in a BP that only bets big with a small spread due to the fact that they do not play the disadvantaged hands. Its also easy for the BP's to play the indices due to the fact they only play the positive ones. As far as the negative indices for the spotters, they could play them, but I would rather they walk away from tables that are too negative anyway. As long as its played well this method is still a very effective way to play. Its got built in cover and shared duties so no one player is overloaded with too much brain strain. Nothing is perfect, but I like this as a way to get close.
  • Grifter said:
    To Get This Thread Back On Track........The original question was about playing to a certain "edge" (EV, Expected Value).

    The EV published for any method is the "optimal" for that method at a given ramp, spread, rules, indices, etc., etc. To obtain that optimal one would have to play absolutely perfect....Never making a counting mistake, never making a playing mistake including indices, and ramping the bets in exact lockstep with the count.

    My original statement, and I say it again..... Nobody can play a method to the "optimal" (in real casino play).

    Ray, you are exactly right. I am trying to describe the "real world" of playing blackjack. Sure, you could physically play to the optimal, but how long would you last before you were 86'd?.....So physically possible, yes; but "real world" possible, definitely no.

    Regards.....Grifter



    Sorry, but I have sat in the Beau, the Copa, the Plaza (and others on Fremont) and spread $5-$40 in a DD game, and played _exactly_ by the count and optimal (rational) betting ramp. Granted if I were spreading $100-$800, that would not work very long. But at $5-$40, or at times $25-$200, I rarely get noticed in places where people walk up and bet 10 benjis on one hand.

    But, and here's the _BIG_ but... if you don't play optimally, do you know what it costs you? I can answer that question to the penny using CVCX/CVData... I don't have to guess. And I can determine if the necessary cover turns a fair game into a poor or losing game without having to learn "the hard way". It is amazing how such small cover plays as "don't reduce on a win and don't increase on a loss" hurts your bottom line. And it is nice to know how bad the effect is before employing it in a real game. And the answer is different as the rules change (LS/NoLS, etc)...

    So yes, I run the sims, I look at the results, I play around with playing indices (16 vs 10 which is often a give-away play for counters) and so forth to see what it will do to me without having to find out the hard way.

    I have seen a couple of friends make ugly mistakes by employing cover plays and winning more than usual and concluding "this works well.." when, in fact, they just had a bit of positive variance and the cover plays they were using had drastically hurt their advantage.

    My point is, it is nice to know, rather than to rely on guesswork, which can lead to problems. Perhaps _you_ don't have to guess. I've counted long enough I don't do a true-count conversion on Hi-Lo. I somehow "know" the proper bet and index based on the running count, from many many hours of practice and play. But most don't practice as much as I do, and many play just a couple of times a year. And they are not going to "know" by experience, so software is by far the best solution to improve their winning chances.

    Negative comments about simming games and rules and such are just nonsense... Negative comments about using CVBJ for practice is more of the same... Both are highly useful for 99% of the people reading here. Your comments more commonly apply to the 1% that already know whatever it is you are talking about.

    I'm an educator, where the "why" is _just_ as important as the "what or how"... The old "give a man a fish or teach a man to fish" comes to mind here. I'd rather teach...
  • Bojack1 said:
    Although I agree that optimal play as described by computer sims is not quite the same as optimal casino play, I do think with team play you can be very close to what is computer optimal. Such as with .......... Nothing is perfect, but I like this as a way to get close.


    Bojack - Good point......and I agree that with team play you could get very, very close to optimal....Grifter
  • Grifter said:
    Bojack - Good point......and I agree that with team play you could get very, very close to optimal....Grifter


    Yes, or even backcounting for the individual players who, for whatever reason, can't team play. Don't play negative counts, big bets and small spreads. The major difference is opportunity for quicker long term results and lower ROR for team. A lone wolf goes hungry at times (few or no hands per hour is common).
  • It has been interesting reading the various computer (and non computer) comments.
    What was driving my question was:
    I have a copy of Blackjack Tracker by Bob Hubby. It is out of print, having been written some years ago. Bob was a Casino blackjack dealer. He and some friends played about 55,000 hands of BJ and recorded the results. The book shows the Wins, Loss, DD, BJ, -DD, P, etc. It does not show you the cards, just the results of the hands. Thus you have columns of data with the results of playing each hand. He also included the Knock-Out count at each hand.
    I have written a program in Microsoft “Excel” to use the data. I enter the hand results and the computer program varies the bet (per my selected instructions) and tracks the $$ results. What I was interested in was a goal for the various “playing techniques” I was trying. (I have entered about 10,000 hands and bet about $130,000, at a $10 table, with an average bet of $15)
    When do I gain confidence, at what %% advantage, that I have the correct “techniques”?
  • sage01 - Wow, Blackjack Tracker is a real flashback for me. Bob put in a ton of work putting that book together and publishing it......Unfortunately for him, Casino Verite was just getting popular about the time it came out; and most players started using CV for their testing.

    Sage, I hate to tell you this, but Blackjack Tracker is not going to tell you when you have the right "techniques", or tell you when your "technique" is playing to an "advantage". Even if you use the K-O counts shown, you won't learn how to improve your "technique".

    There is only one way to play blackjack to an "advantage" at this stage of your blackjack education.......You must learn how to count cards, and after you learn the basics, you must learn to be very good at it under casino conditions. It doesn't matter which counting method you select, so take your choice of any "Level 1" method, and learn everything about it.......and practice, practice, practice.

    I would suggest you purchase Blackjack Bluebook II by Fred Renzey, and then read it, read it again, and then read it again.........It is an excellent book for beginners to experts, and gives you all the fundamentals plus at least four counting methods.

    Hope this helps.......Grifter
  • sage01 said:
    It has been interesting reading the various computer (and non computer) comments.
    What was driving my question was:
    I have a copy of Blackjack Tracker by Bob Hubby. It is out of print, having been written some years ago. Bob was a Casino blackjack dealer. He and some friends played about 55,000 hands of BJ and recorded the results. The book shows the Wins, Loss, DD, BJ, -DD, P, etc. It does not show you the cards, just the results of the hands. Thus you have columns of data with the results of playing each hand. He also included the Knock-Out count at each hand.
    I have written a program in Microsoft “Excel” to use the data. I enter the hand results and the computer program varies the bet (per my selected instructions) and tracks the $$ results. What I was interested in was a goal for the various “playing techniques” I was trying. (I have entered about 10,000 hands and bet about $130,000, at a $10 table, with an average bet of $15)
    When do I gain confidence, at what %% advantage, that I have the correct “techniques”?


    Let me add one or two comments to follow behind Grifter's advice.

    1. You _MUST_ get used to the idea of the variance in this game. The advantage is pretty thin if all you do is count cards. You can certainly win, but the variance is difficult to grasp until you spend some time studying results and using a simulator. You can play 10,000 hands and be way up or way down, and still be within "expectation" because of the wild variance we see in this game. (more in a minute).

    2. CVCX/CVData is the way to get your questions answered. 50K hands is nothing in the world of blackjack when you want to know what the actual expectation is for a game with the variance we experience. TO understand the variance, you should use CVCX and sim some small number of rounds (I am not sure how low you can go, and I'd have to reboot my linux laptop to bring up windoze to run CVCX and see) like 5,000. Do this several times and notice the final result for each run. That's ugly.

    With CVCX/CVData, you can run _billions_ of rounds of 21, rather than thousands or tens of thousands, to get a better view of what to expect. Just remember that whatever the result says, you have to take it with a large grain of variance salt, because you can't actually play enough rounds to get to that stable result. Getting to 55,000 rounds is a _lot_ of play for one person, unless you are talking about years of play. Assuming about 1,000 hours, you'd have to play at least 3 hours a day for the entire year. Possible if you live in a town that has casinos, unlikely if you have to travel significant distances to get there and you are not a "pro" that plays for a living.

    There's no better learning tool than CVCX and CVBJ. One gives you tremendous insight into how the rules, BS indices, counting system, playing accuracy, betting strategy, cover plays (if any) will affect your ability to win any money. The other will coach you to the point that errors are almost non-existant and make counting second-nature. When you can play other games and find yourself counting as though it were 21, you have reached that stage.

    It is really hard to explain or understand how a small change to your playing strategy (which includes betting, cover, indices, etc) affects your bottom line. It might seem that "this is a fairly rare occurrence, it can't hurt..." but when you test that hypothesis with CVCX you go "holy sh**, batman, that hurts..." That's the value of the software... And the nice thing is, learning how bad a particular idea is doesn't cost you a penny beyond the nominal cost of the software. Takes almost no time to run a decent sim (and many useful sims are already provided so that the computation time is zero for those) and since you don't risk any money, it is hard to get any better than that...
  • Thanks for the input! Do you question the results of Blackjack Tracker??
    This past summer I was nowhere near a Casino. However, I wanted to sharpen my game for the winter season. I have Casino Verite, BJ 4.2.3 and played the game most nights from June to October. The goal was to sharpen my BS skills. It worked; I can play BS on a hand without consciously thinking about it.
    However, I noticed that I would get long runs of losses. Maybe 7 or 8 or more hands without a win. When I got home this fall I reviewed BJ Tracker & did not see either the very long runs of losses or the frequency of the losses. Got to thinking, maybe the random number generator had a hitch in it, or the order of the cards in the 6-deck Casino game was not the same as a computer random number generator.
    I can count a little bit using the KISS II system. However, I do not have the bankroll or the fortitude to vary the bet from 1 to 10 units in my local 6-deck game. I am trying to vary the play of the hands using the count. But want a simpler betting system(less $$ swing). Thus the work on Blackjack Tracker. And I’m making progress!
  • Stainless Steel Rat,
    I don't have CVCX/CVData, sounds like I should look into it!
    Thanks,
  • sage01 said:
    Thanks for the input! Do you question the results of Blackjack Tracker??
    This past summer I was nowhere near a Casino. However, I wanted to sharpen my game for the winter season. I have Casino Verite, BJ 4.2.3 and played the game most nights from June to October. The goal was to sharpen my BS skills. It worked; I can play BS on a hand without consciously thinking about it.
    However, I noticed that I would get long runs of losses. Maybe 7 or 8 or more hands without a win. When I got home this fall I reviewed BJ Tracker & did not see either the very long runs of losses or the frequency of the losses. Got to thinking, maybe the random number generator had a hitch in it, or the order of the cards in the 6-deck Casino game was not the same as a computer random number generator.
    I can count a little bit using the KISS II system. However, I do not have the bankroll or the fortitude to vary the bet from 1 to 10 units in my local 6-deck game. I am trying to vary the play of the hands using the count. But want a simpler betting system(less $$ swing). Thus the work on Blackjack Tracker. And I’m making progress!


    I don't know enough about it to question the results. My point is that 55,000 rounds seems like a not, but the variance is large enough that one could end up ahead or behind. And the real problem with using this approach is that the 55K hand outcomes are fixed. It would be quite easy to define a progression to beat that string of results, even though it would not work in real games... 55,000 is a miniscule sample-size when there are litterally millions of trillions of possible game outcomes over that many rounds.

    The problem with smaller bet spreads is a much lower win rate. In blackjack, win rate and bankroll variance go hand-in-hand, and that is what makes the ride so wild. you bet big when you have the edge, but when you lose it drains your bankroll like mad.

    Once again, CVCX can come to your rescue by telling you exactly what a smaller spread will do against a particular game. For a 6D shoe, anything less than 1-12 will be _very_ marginal. If you want to give me some game rules and a counting system, I can give you the CVCX results for spreads from 1-2 to 1-20...
  • Hi SSR, Thanks for the input
    I was surprised by your comment that “It would be quite easy to define a progression to beat that (55K hands) string of results, even though it would not work in real games”
    If you have the time, I would like to take you up on the offer to run some sims with CVCX. Let’s try about 10 + runs of anywhere from 10K to 55K hands. I assume they all run as one long session. Also does it plays the hands & not just assumes a % gain? What is the logic of the software? How does it go about the problem? I would be interested in the variance of the final results. Maybe we could treat the results like a standard deviation.
    I have entered and run about 10K hands of the 55K listed in the book. What I’m finding is the results (even in this small number) seem to be converging.
    Let’s try, 6 decks, I play $10 and max bet is $50 (1 to 5 spread) if that is too small I could change tables and move up to a 1 to 6 spread. It is a good game, about 85% pen, (less than 1 deck behind the cut), double any two, split to 4 including aces, double after split and late surrender. Stand 17, BJ pays 3 to 2. Bjmath.com puts it at EV= -.402225%
    Have a great Holiday, looking forward to the results
  • sage01 said:
    Hi SSR, Thanks for the input......I was surprised by your comment that “It would be quite easy to define a progression to beat that (55K hands) string of results, even though it would not work in real games” .......


    Sage01.....Ignore that statement, he can NOT do it......He is just blowin' smoke again.

    SSR....... Last warning! This forum is sick and tired of you pretending to be a blackjack expert. If you do not know what you are talking about, the don't post about it. Until two days ago, you had never heard of Bob Hubby and Blackjack Tracker, and now you are can define a progression to beat those 55K hands because it would be "quite easy".......Well, put your money where your mouth is, because I KNOW two of the players cannot be beat even with a count............

    Here's the deal:
    - I will bet you $1,000.00 you cannot define a progression that can beat those 55K hands......You have five days to do it (just send your progression to me and I will run it....I have all 55K hands from BJ Tracker on Excel).

    - If you cannot back up your statement, then post an apology to Sage01!!

    Pretty simple isn't it........So as Nike says, "Just Do It". (one or the other).

    Regards......Grifter
  • Grifter said:
    Sage01.....Ignore that statement, he can NOT do it......He is just blowin' smoke again.

    SSR....... Last warning! This forum is sick and tired of you pretending to be a blackjack expert. If you do not know what you are talking about, the don't post about it. Until two days ago, you had never heard of Bob Hubby and Blackjack Tracker, and now you are can define a progression to beat those 55K hands because it would be "quite easy".......Well, put your money where your mouth is, because I KNOW two of the players cannot be beat even with a count............

    Here's the deal:
    - I will bet you $1,000.00 you cannot define a progression that can beat those 55K hands......You have five days to do it (just send your progression to me and I will run it....I have all 55K hands from BJ Tracker on Excel).

    - If you cannot back up your statement, then post an apology to Sage01!!

    Pretty simple isn't it........So as Nike says, "Just Do It". (one or the other).

    Regards......Grifter


    OK, since that's your preferred tone of conversation, are you dumb as a rock? You don't think that once I know the outcome of each and every hand, in the order they will be played, I can't produce a betting progression that will beat that? I don't even need a progression. And yes, it is possible that for some sequences, any simple progression might fail. But I imposed no constraints on whether it would be simple, or a progression that is 55,000+ bets long.

    You apparently can't see the forest for all those trees around you, and you try to hone in on a minute detail that not even an important aspect of the discussion... which was about the quality of result when basing it on 55,000 rounds versus billions of rounds with CVCX. There is no comparison between the two results. CVCX wins hands down. That was _the_ point we were discussing before you decided to jump in with pointless comments... As far as "two of the players can't be beat even with a count..." Do you have _any_ idea what that proves? That the 55,000 games are atypical. And using atypical hands to draw any conclusions is not exactly "good science". Or do you believe card counting doesn't work? That actually makes my point even stronger.

    I'm not blowing any sort of smoke at all. You, on the other hand, are apparently inhaling some that is pretty potent... The "I'll bet you xxx" ranks right up there with the dreaded triple-dog-dare-you argument... And it is just as effective...

    I do believe I _clearly_ said that I was not familiar with his work. I was drecting my comment to a fixed series of results. I'd suggest you either take something, or stop taking something, whichever will help your attitude and ability to follow a discussion...
  • SSR- The only way to beat those 55K hands with a progression is to have it change in relation to a predetermined sequence of cards. In that case, we are talking about many progressions rather than a single defined sequence and that amounts to many strategies rather than one.

    Suggest you give up the denial and admit you mis-spoke just a tad................
  • Ray said:
    SSR- The only way to beat those 55K hands with a progression is to have it change in relation to a predetermined sequence of cards. In that case, we are talking about many progressions rather than a single defined sequence and that amounts to many strategies rather than one.

    Suggest you give up the denial and admit you mis-spoke just a tad................




    There _is_ a pre-determined sequence of cards already. Did you overlook/miss that _key_ point here???

    That was the complete point I was making. Using that as a basis for analyzing betting or counting systems is poor science at best... it is badly flawed. And if it is true that 2 of the players in the records can't win even with counting, what exactly does that tell you about those 55,000 hands? Tells me they are no good for studying anything, else I would conclude card counting doesn't work and move on.

    In a worst case, the progression required to beat that set of hands is a sequence of 55,000 unique bets. In a best case scenario, a much shorter "pattern" will do the job.

    This is not a difficult job for a computer to figure out. My chess program regularly searches billions of moves in a couple of minutes. That's more than enough to exhaustively try different betting approaches to find the shortest progression that will beat 55,000 unique hands...

    I don't see why this is so hard to understand, at least for a computer scientist familiar with searching enormous tree search spaces, it doesn't represent much of a computational challenge...

    And again, that was not the main point being made, which seems to be getting overlooked at _every_ response...

    The question was about the viability of basing conclusions on 55,000 hands vs billions. The difference is staggering...
  • Grifter said:
    just send your progression to me and I will run it....I have all 55K hands from BJ Tracker on Excel.


    Can you make the spreadsheet available online? I wouldn't mind taking a crack at it. :)

    As SSR said, it would seem like a simple job to design a winning progression system when the outcome of the hands is already known.

    -Sonny-
  • ...he writes, running for cover as he hits "submit reply"...

    dangerous territory here when someone is overlooking the basic key point that the outcome of each hand (win/lose/push) is known in advance.

    But it is nice to know that some "get it". This certainly is not a heavy computational load for today's processor speeds...
  • Grifter:
    Which of the two hands of the four can't be beat with card counting?
  • SSR:
    If you have the time, I would like to take you up on the offer to run some sims with CVCX. Let’s try about 10 + runs of anywhere from 10K to 55K hands. I assume they all run as one long session. Also does it plays the hands & not just assumes a % gain? What is the logic of the software? How does it go about the problem? I would be interested in the variance of the final results
    Let’s try, 6 decks, I play $10 and max bet is $50 (1 to 5 spread) if that is too small I could change tables and move up to a 1 to 6 spread. It is a good game, about 85% pen, (less than 1 deck behind the cut), double any two, split to 4 including aces, double after split and late surrender. Stand 17, BJ pays 3 to 2. Bjmath.com puts it at EV= -.402225%
    Looking forward to the results
  • Sonny said:
    Can you make the spreadsheet available online? I wouldn't mind taking a crack at it. :).....-Sonny-


    Sonny - If I had done the work, sure I would send it to you, but I didn't. A player friend of mine spent a ton of time creating it and gave it to me "for my use". Sure there is nothing secret about it, and anybody could do it if they wanted to take the time, but posting somebody else's work just wouldn't be "right". I get the feeling you have been around casino's and gaming enough to understand, call it "honor among thieves" or whatever.....Capiese?

    Regards.....Grifter
  • Grifter said:I get the feeling you have been around casino's and gaming enough to understand, call it "honor among thieves" or whatever.....Capiese?

    I understand. Is there anything special about the BJ Tracker hands or would any string of 55k hands suffice? I wrote a BJ simulator that can give the results of every hand, so a list of 55k hands (or any number for that matter) would be very simple to compile as a text file. Would any random* string of 55k hands be acceptable for your challenge?

    -Sonny-

    * - Technically the hands wouldn't be “random” since they would be dealt from a shoe with a predetermined shuffle point. However, that would make the results of the hands more realistic.
  • sage01 said:
    Grifter: Which of the two hands of the four can't be beat with card counting?


    Sage01.....If my memory serves me right it is Players 2 and 3, definitely Player 2.....You don't lose big, but you do lose. BUT read this next part carefully because it is about "real" players vs jammie boys like SSA, who stated those hands were atypical (or whatever he said).

    When you divide all those hands in BJ Tracker by the four players, that loss by Player 2 would only equate to about three months of "real" play by a "real" player.......AND there isn't one real player on this board who hasn't gone through a losing stretch of that long or longer.....Hell, for example Ray went through a losing stretch of almost a whole year last year.

    Regards.....Grifter
  • Grifter said:…it is about "real" players vs jammie boys like SSA, who stated those hands were atypical (or whatever he said).

    I believe he said that the results of those 55k hands were atypical of the long-term expectations, which makes them useless for analyzing any betting system. The fact that it is just a small subset of hands that is not representative of real world conditions is his main point. The fact that those 55k hands can be manipulated to show that a winning system is an overall loser while a losing system is profitable makes it a complete waste of time. Your explanation of how extreme the variance can be is exactly right and shows why BJ Tracker is not a worthwhile tool.

    Perhaps your challenge can shed some more light on this topic...

    -Sonny-

    P.S. - Did anyone notice how I was completely uninterested in this thread until money was involved? :)
  • stainless steel rat said:
    There _is_ a pre-determined sequence of cards already. Did you overlook/miss that _key_ point here???.........In a worst case, the progression required to beat that set of hands is a sequence of 55,000 unique bets.......I don't see why this is so hard to understand, at least for a computer scientist familiar with searching enormous tree search spaces,.......


    Ray.....Jeez, are you a dense Kentucky hillbilly or something? Don't you know that 55,000 hand progressions are used everyday by real players in real casinos with real chips at a real table?.....Hell, I thought you were a lot smarter than that.

    But I sure bet you realize it now, because a "computer scientist familiar with searching enormous tree search spaces" just enlightened you.

    ROFLMAO.......Grifter
  • Grifter said:Don't you know that 55,000 hand progressions are used everyday by real players in real casinos with real chips at a real table?

    SSR never said that a 55,000 hand progression was realistic, only that is was a “worst case” scenario for beating your challenge. Technically he beat your challenge mathematically. Does he need to write down an arbitrary 55k hand progression before he gets paid or does he win by theory alone? And does that mean I can’t get paid if I win the challenge too? :)

    Also, keep in mind that a 55,000 hand progression doesn’t necessarily need 55,000 unique “moves” to be a complete strategy. For example, a Martingale player who parlays through 10 losses might make it through 1,000 hands without ever reaching his max bet. That simple 10-step strategy lasted for 1,000 hands without being completed. A more conservative progression system could easily last much longer with only a few “moves” involved. That is the nature of repeated patterns. Sometimes only a few instructions can cover a large amount of situations. Even most of basic strategy can be summed up in a few simple sentinces even though it covers hundreds of unique hands.

    -Sonny-
  • From my point of view I don't believe that the approach being proposed here has any intelligence whatsoever. A damn fool can develop a progression to beat a billion hands if he knows the outcome of each and every hand. That is not real world nor does it require an analytical program to arrive at the solution. Why do the experiment if you already know the solution?


    For example:

    Here is a very simple pregression that will do the trick and I only need to know the outcome of the first hand. Lets say it is a win.....three step progression with the other 54,999 hands in any sequence. Be my guest and shuffle after each hand if you like.


    $55,ooo-$1.00-$1.00....... (Win 55K first step and go to step two, step two if lose stay at step two, if win go to step three and win or lose go back to step two and repeat.) I will not double or split. If I lose 54,999 hands I will still win one dollar.

    Ridiculous, like in laughable, yes it is , but this discussion is totally childish to begin with.
  • Ray said:From my point of view I don't believe that the approach being proposed here has any intelligence whatsoever.

    I agree completely, but it’s worth $1,000 to anyone who agrees with you. That alone makes this thread very valuable! *lol*

    Ray said:A damn fool can develop a progression to beat a billion hands if he knows the outcome of each and every hand.

    Exactly. It would be much more exciting if there was a limit on the number of steps in the progression system. For example, if the progression system was limited to 10 steps (or less) before repeating itself it would be pretty darn tricky to make it last for very long. Unfortunately, that would still be a complete waste of time…unless ZG wants to go “double or nothing.”

    -Sonny-
  • Sonny said:


    I agree completely, but it’s worth $1,000 to anyone who agrees with you. That alone makes this thread very valuable! *lol*



    Exactly. It would be much more exciting if there was a limit on the number of steps in the progression system. For example, if the progression system was limited to 10 steps (or less) before repeating itself it would be pretty darn tricky to make it last for very long. Unfortunately, that would still be a complete waste of time…unless ZG wants to go “double or nothing.”

    -Sonny-


    Sonny- Don't you agree that we have better things to talk about? What I'm trying to say here is that sometimes the end does not justify the means. Like who gives a big rats "A".
  • During one cold winter week Grifter was kind enough to let me try my hand at beating a limited number of those 55K hands with a progression. I used a modified verision of a D'....(can't remember the guys name, French maybe). I was able to beat hundreds of hands without knowing the hand sequence and on several different trials. My progression had some logic that tried to capture the likely sequence and resonable consistency (due to having so many cards) of a 6D game. Well I got close, but close don't count. In an actual casino I was able to win a little and then lose more than I had previously won (low stakes). I was able to conclude that this progression like all others must in some way or another rob Peter to pay Paul. Further, the closer I got to winning 48.5% of the hands, the more likely I was to win.

    The fact that a progression player will lose at the same rate as a flat bettor tells you all you ever need to know about progressions. Namely, that regardless of the structure each an every out-come has an equal possibility to occur and over time that is exactly what happens. But the house has that tiny edge. Thus, the net of all progessions is a big fat ZERO. They are all the same, but you do lose more money because of the higher average bet.
  • You’re explanation of progression systems is perfect. There is no system that will show any benefit in the long run, and often they show a bigger loss than flat betting. However, in the short run there are progression systems that can beat certain subsets of cards. We’ve all heard people say “I’ve been using this system all year and I’ve made tons of money. It definitely works!” It may have worked for the hands that person played, but against other subsets (most subsets in fact) the same system will fail.

    If the goal is to design a progression system with a long term positive expectation then there is no solution. However, Grifter’s challenge was to create a progression system that beats a specific series of hands. As long as the number of hands is relatively small there will be systems that can be developed to show a profit. Only when the number of hands is large enough to approach a normal distribution will the systems begin to fail. When the number of hands is small and the outcomes are known a priori, there are probably several progression systems that will beat the game. That was the $1,000 mistake.

    -Sonny-
  • Sonny said:


    I understand. Is there anything special about the BJ Tracker hands or would any string of 55k hands suffice? I wrote a BJ simulator that can give the results of every hand, so a list of 55k hands (or any number for that matter) would be very simple to compile as a text file. Would any random* string of 55k hands be acceptable for your challenge?

    -Sonny-

    * - Technically the hands wouldn't be “random” since they would be dealt from a shoe with a predetermined shuffle point. However, that would make the results of the hands more realistic.


    there appears to be one thing that is unusual, in that over 55K hands, two of the four players can't win at those positions even when using counting. This suggests that either the games played were very poor making for a really large N0, or else the 55,000 hands are atypical.

    For this particular argument, I don't see why any 55,000 hands would not suffice, but I doubt it will "make the point" since it wouldn't exactly the same 55,000 hands mentioned...
  • Grifter said:
    Ray.....Jeez, are you a dense Kentucky hillbilly or something? Don't you know that 55,000 hand progressions are used everyday by real players in real casinos with real chips at a real table?.....Hell, I thought you were a lot smarter than that.

    But I sure bet you realize it now, because a "computer scientist familiar with searching enormous tree search spaces" just enlightened you.

    ROFLMAO.......Grifter


    I am probably laughing harder than you, because you have some sort of reading issue that lets you overlook key words here and there. Please show me _anywhere_ where I wrote that a real player in a real casino
    would use a 55,000 point progression? Or did you miss the point where I _clearly_ said that once the hand outcomes are known, it is easy to produce any sort of random system that will win? And are you incapable of understanding that in a real casino, I won't know the outcome of the hands in advance?

    I don't think you are that dense. I think you just like to argue for no real reason, and try to use semantical misdirection to keep the argument alive. Again, for clarity, my position has _always_ been:

    (1) no progression will work. Oscar's Grind that you seem to like included.

    (2) once hand outcomes are known, it is not hard to create a special counting system or special betting system that will beat _that_ series of hands. Even though the system might well be a "losing system" when applied to a real game.

    (3) using 55,000 hands to study betting or playing systems is simply poor science. Period. Might have been interesting in its day, but it is not interesting in the day of CVCX/CVData. Period.

    I've never said more than the above, regardless of your lame attempts to imply that I did. Other things you have misstated:

    (1) you can't play perfectly by the count in a real casino. Strange because I have done it _many_ times. And I clearly gave my betting ramp to supply the stakes I played at. Playing black, no you can't spread 1-8 in a DD game without cover. Playing $5-$40 or even $25-$200 you can, at the right locations. I can point you to a well-known poster on other BJ sites (not here for obvious reasons) that has played at one of the same casinos I played at (before it was blown across US 90 by Katrina) and he spread $5 to $40 and even $50 in a single-deck game, without getting the boot. Just because you can't do it doesn't mean everybody can't. I played $5 to $40 (or a bit higher) all over Nevada (Including laughlin at your suggestion) without any notice, using my one-hour or less session strategy.

    So one _can_ play exactly by the count, use the I-18 (and then some) playing indices perfectly, and survive, at the right stakes, with some good time management.

    2. You seem to think technology is over-rated and not necessary. Programs like CVCX/CVData and CVBJ are incredibly useful for beginners or experienced players. Then we don't have to rely on guesswork. Or on "this has worked for 40 years." Flathead motors worked for 50 years as well, but OHV/OHC motors are far better.

    One of my favorite discussions centers around the physics "law of entropy". If you are not getting better, you are getting worse. It really is that simple. I try to use tools to make me better, because sitting around and poo-pooing computer-related discussions means that one is getting worse, since he is not getting better.

    Hopefully the few that read messages here will figure out that the software is really "worth a look". Even though I am sure you consider Norm another "jammie boy computer nerd" also....
  • Ray said:
    From my point of view I don't believe that the approach being proposed here has any intelligence whatsoever. A damn fool can develop a progression to beat a billion hands if he knows the outcome of each and every hand. That is not real world nor does it require an analytical program to arrive at the solution. Why do the experiment if you already know the solution?



    As I have said _many_ times, the forest has gotten lost in the trees. Back to my _original_ point. Using a series of 55,000 hands to study betting systems, playing systems, etc, is simply "bad science" when compared to using billions of rounds of simulations. 55,000 hands can be beaten by a reasonably simple betting system. By reasonably simple, I would suggest something I could carry with me in my pocket as "reasonably simple. At worst case, 55,000 unique bets would do the trick, but more likely there is a much simpler strategy, since we can use a computer to analyze the results. A simple idea. How many times, in that 55,000 hands, do you win 8 in a row and then win the next, vs how many times do you win 8 in a row and lose the next? If you win 8 in a row and then the 9th more often than you lose it, that's all you need. Flat bet until you win 8 in a row then bet the table max. Since you will win more of those than you lose, you come out a big winner.

    So it could be "that easy" since we already know the 55,000 results in question. But that is not the point...

    The point was that 55,000 hands is too little if you are going to try to draw conclusions on playing/betting/counting strategies based on those results. If I run 20 billion rounds thru a sim, I can probably still find a pattern to beat _those_ pre-determined results. But the pattern won't work on the _next_ sim, or the next 20 billion rounds, because they won't be the same hands.

    My original point was that using the results from that book is not just inferior to today's options, the results are pretty meaningless overall because the number of hands is too small.

    I've not yet had a year-long losing streak. I have had one that lasted beyond 3 months, but then I have only been playing 8-10 years now so I am sure the "worst" and "best" days are yet to come my way. But one thing is for sure, I am not going to ever play 20 billion rounds, and by that number of rounds, the standard deviation is extremely small and the results are far more useful/dependable.

    That was my point. CVCX/CVData are _far_ superior to using the data mentioned. And I said _exactly_ that if you look at my original post. Grifter just went ballistic and so far out in left field he left the stadium completely, and zeroed in on the "progression comment" (which is trivially correct as I just explained) rather than on the point I was making. Which leads me to believe he simply wanted confrontation, rather than exchange. He's been doing a lot of that.





    For example:

    Here is a very simple pregression that will do the trick and I only need to know the outcome of the first hand. Lets say it is a win.....three step progression with the other 54,999 hands in any sequence. Be my guest and shuffle after each hand if you like.


    You are now "getting the point". Using the same 55,000 hands over and over, while altering your basic strategy for betting or playing will not work. Because eventually you will stumble onto a system that beats _that_ specific sequence of hands, but loses horribly in the next set of 55,000 results.





    $55,ooo-$1.00-$1.00....... (Win 55K first step and go to step two, step two if lose stay at step two, if win go to step three and win or lose go back to step two and repeat.) I will not double or split. If I lose 54,999 hands I will still win one dollar.

    Ridiculous, like in laughable, yes it is , but this discussion is totally childish to begin with.


    No disagreement there. But there is a basic underlying point, which I clearly tried to make, that seems to get lost in the forest each time Grifter jumps in...

    The worry here is that suppose something simple like Oscar's happens to work on those 55,000 hands? Or suppose there are not enough losses so that the Martingale works? Is the conclusion that "this works" of any use to anyone? Do you think there's a chance in hell the Martingale would work over 20,000,000,000 hands? Of course not.

    But we can't seem to get "there" from "here" without tons of static and nose covering up the primary data signal...
  • Ray said:
    Sonny- Don't you agree that we have better things to talk about? What I'm trying to say here is that sometimes the end does not justify the means. Like who gives a big rats "A".


    I give a big rats "A" if someone is interested in using something that clearly won't work well, much less given the alternatives available today...

    Bottom line was and still is, use real simulation software, not something from the dark ages that can/will give misleading results when used as the original poster intended...

    If that point isn't important, then nothing posted here is...

    The old "to the man who has a hammer, everything looks like a nail" should not apply to everyone. We have screws, hex-bolts, internal/external torx, torx-plus, security-torx, 5-point torx, etc. Newer tools do better than the old hammer in _many_ cases. On a modern car, a hammer is worthless. Results based on 55,000 hands are worthless given the alternatives...

    And one might wonder why there is so little technical discussion posted here after seeing this debacle???

    Fortunately I considered Grifter's $1000 bet to be total nonsense, else I would hassle him for years to come for the payoff... Sonny's idea of a "limit" on the progression would make it more interesting,since these kinds of problems are common in computing (traveling salseman problem, knapsack problem, etc). But the bet is already lost since I can beat those 55,000 pre-determined hands easily enough... even if it takes 55,000 unique bets. When no constraints are placed on a bet, it seems a bit childish to then complain "but who could use a 55,000 term progression in a real casino?" That question itself is pretty stupid, because anyone with an ounce of sense would realize that the casino would not deal you the same 55,000 hands...

    I've tried to answer each and every pointless point he raised. Without resorting to incessant name-calling arguments, which is the mark of someone that has already lost the debate...

    If you, or Grifter, or anyone else wants to use that old stuff to study betting/playing/counting systems, more power to you. I hope most of us use something far more modern, accurate, and usable... I know I do...

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!