And you have statistical evidence to show how these ASMs are constantly producing bad shoes? Or are these just anecdotal, "Man, I lost a lot of money lately. Must be the ASMs."
This is from #4 post in this thread and I add some comments:Originally Posted by dunknballer21
Group 1: {2,3,4} x 6/13, face cards x 5/13, other cards x 2/13 (face card rich clump)
The goal is to double the intensity of {2,3,4} in this clump so it will increase the chance of getting pairs of 2, 3 or 4 but if you double after split, you like to get a small card. On the other hand, the dealer likes make a hand because of small card rich.
Group 2: {7,8,9} x 6/13, face cards x 3/13, other cards x 4/13 (face card poor clump)
The goal for ASM is to have player three card bust before dealer three card bust unless your three cards is 7/7/7.
Group 3: {A} x 1/3, face cards x 2/13, other cards x 1/2 (face card poor ace rich clump)
The goal for ASM is to reduce player Blackjack possibility. Also after splitting aces, increase the chance to get another ace because of face card poor and ace rich.
Group 4: {5} x 1/3, face cards x 1/3 or 1/6, other cards x 1/3 or 1/2
The goal for ASM is to have players get another five after doubling a pair of five.
Group 5: {6} x 1/3, face cards x 1/2, other cards x 1/6 (face card rich)
The goal for ASM is to have players get two face cards after splitting a pair of six and get two hands of 16.
I should add. The most damage is group 1. By isolating the majority of {2, 3, 4} in one group, it tremendously increases the chance for the Basic Strategy players to bust their hands in playing the other 80% of the shoe (group 2 to group 5) because the lack of small cards.
This is what I look for when I play the first three shoes. Do {2, 3, 4} come out together regardless of TC? If there are more than twelve {2, 3, 4} in a span of twenty six cards (statistically it should be only six cards in a span of twenty six cards) and it happens periodically, ASM is in clumping mode.
Last edited by BJGenius007; 09-18-2014 at 03:23 PM.
Exactly what would you be doubling after you split?
Split 2,2v2-4: What card in this group of 13 next cards gives you a double? One of the 2 mystery cards? You need a 8 or 9.
Split 3,3v2-4: you need one of the two mystery cards to be 7 or 8.
Split 4.4v2-4: Not a split to begin with so you need 1 of the 2 mystery cards to be a 5 or 6 dealer upcard and the other be a 6 or 7 to give you the double.
So much for that theory.
Assume it is 2,2 v 2, the basic strategy says to split. But if the shoe is clumped in zones of five different groups and you are in group 1 zone, even if you get a very scarce 8 or 9 after the split then double, the next card is most likely 2, 3 or 4. Meanwhile the dealer never splits, so in zone of group 1, he will easily make a hand and beat the player who plays correct BS and has 2 hands of 12. Something like 2, 2, 10, 2, 3. Another example is 3, 3, 9, 2, 2. My point is that any card is possible but {2, 3, 4} is the primary cards in group 1 zone. I have seen ploppies never split and hit 12, 13, 14 no matter what the dealer's upcard is. Basically these ploppies play like the dealer. Hit every 12, 13, 14 and the difference is that he or she may stop at 15 and 16. These ploppies turn out to be the winners if ASM is in the clumping mode. Normally I saw BS players run out of their money first and fast, and need to sit and watch their ploppy friends play on these tables. As an AP, if I see two or three rotations of zone 1 early, then TC should be high in the near end of the shoe and I can make up the earlier loss by the spread. But it is not guaranteed. I should add one more thing. It is not a hard evidence. But I have only two blackjack in almost two hours. I side count aces. Ace count is constantly 10 to 12 short in the end of 8d shoe. It happened half of the times that day. For the other shoes, there were 8 to 10 aces in a span of 15 cards. I am saying all shoes I played that day had extreme ace clumping. Statistically it is impossible.
Impossible??? LOL... I've mentioned the following before...
A coin flip... We all know is 50/50... But is it possible that you flip it 10 times and they all land on heads???
The answer is yes... Flip it a million times now... Is it possible that they all land on heads???
Yes again... Not at all likely, "statistically" speaking.
Flipping a coin is a very bad example for my experience. It is more like drawing 400 balls from a box and put the balls into a matrix of 20 by 20. In detail, draw the ball one by one and fill the matrix from the upper left corner to the lower right right corner in row by row fashion. There are only 32 white balls of all 400 balls. What is the odds that at least 27 white balls appears on only 4 rows in every matrix after you repeat this 12 times? (In this example, there are at least 11 rows without any white ball. About 5 rows, each has only one white ball and 19 balls of other colors. Four rows has a lot of white balls.) Unless the person drawing the ball can see the colors of the balls, so after fetching a white ball, he tries to fetch the white ones for the next 19 balls, too. And it is not just the 12 matrices for the day. For the next 50 days you come, you see the similar matrices. Then you go to another casino, and you see white balls are pretty uniformly distributed in 20 rows in each matrix. Like there are always 3 to 6 rows without a white ball and the majority of the rows have very few white balls each. Sometimes you may see a row with more than 10 white balls! But that is like once in a day, not all the times in the first casino!
Last edited by BJGenius007; 09-18-2014 at 08:40 PM.
Once someone says something is happening you look for it and even though the frequency of occurrence hasn't changed you notice it more than you used to. You believe it is occurring more. At a craps table this is apparent. 11 are said to come in clumps. They come in clumps just as often as any other number but one an 11 follows an 11 it is pointed out reinforces the ploppy superstition. 7 out is said to happen within 4 rolls of the dice going off the table. What are the odds of 7 out in 4 rolls?
1/6 + 5/6*1/6 + (5*5)(6*6*6) + (5*5*5)/(6*6*6*6) = (216 + 180 + 150 +75)/1296 = 47.9%.
Almost half the time. That is more than enough to reinforce the myth especially since the times a 7 doesn't show aren't noted like when the myth has outcome reinforce myth.
60 hours is a small sampling so even if you did all the extra to eliminate confirmational bias, what is the expected amount of these clumps and what is the SD for this event over that time frame. I am sure it is well within the statistical expectation and unremarkable.
Welcome to the crazy swings inherent to blackjack. It sounds like you were long overdue to experience this. No matter how strong the counting system you use you will hit periodic losing periods. A literal pro like KJ has had some ridiculously long losing periods. Most of us look at how we can improve our decisions and game selection to minimize this. It may not be why you experienced tough results but it will help future performance but the ploppy minded try to figure out how the casino is cheating them when they experience normal downswings that are inevitable if you play long enough. Then the worst thing can happen, you start figuring ways to change optimal play to deal with the imaginary threat causing the adjustment to lower your EV rather than having your study on finding your own weaknesses improve your EV.
Every possible deck composition is an equal part of what is random. No deck composition exists that is not an equal part of the whole of what is random than any other composition. Translated the odds of the cards coming out sorted by rank and suit are the same as the odds of any other distribution that is defined by rank and suit. You just see it as an odd distribution. This illustrates why once you look for a distribution its occurrence is given more significance.
I make a lot of money off the cards being clumped. If they weren't counting wouldn't work. If the casino wanted to cheat a counter they would make sure the RC never got very positive. That would make a counter be almost a flat better and kill all the advantage from betting correlation. In shoe games that is were most of your advantage comes from.
Look at the Genius's clumps and what you would need to know to determine if the clumping was more than random predicts:
What are the odds that you would see this 13 card clump. What is the standard deviation for the odds?
What are the odds of the clump lasting more than 1 cycle and the standard deviation.?
How often do you see this clump ?
Another 13 card clump with the same questions.
The least common denominator for this clump is 78 cards so to keep the ratios pure you need a 78 card clump. Of course he wasn't talking about that.
The same types of questions need to be asked to show significance.
This clump has two different clumps as one clump. The least common denominator is 6 cards for both clumps. One clump has 2 T's and 2 other cards in the 6 cards while the other has 1 T and 3 other cards. What are the odds that any of these while be the next 6 cards or a multiple of 6 cards? What is the standard deviation for these frequency of occurrences?
Another 6 card clump.
The same applies.
Then with that all said how different fro the defined clumps can the sequence of cards be to be perceived as the said clump when it is actually not the exact specified clump? What is the standard deviation of this undefined clump?
When you get to this final statement you need to see how many standard deviations off expectation your results are to seeing from the expected distribution. Remember standard deviation is relative to sample size and not a constant. It is an exponential function proportional to the square root of the sample size; eg:
In a sim at 100 hands HILO has an EV of 50.32 and a SD of 39.248 but the 1 hand SD is 3.9248. It is not the same SD or the same SD/hand. It is the SD of 1 hand times the square root of the number of hands in the sample size. 100 times the sample size has 10 times the standard deviation. 10 is the square root of 100.
How does your observation of clumping stack up to the expectation and standard deviation for your sample size? You are expected to perceive a higher frequency once the event changes in your mind from part of the norm to something odd to watch for and take note of. That is confirmation bias and almost everyone would be guilty of the same.
Yes my sample size is small. roughly 60 hrs for the past two months. But in these 60 hrs, I see the cards pattern different from when I played before my long vacation. As an AP I have been through bad swings, I mean really bad swings. But the way I have been losing recently is very different from my previous experience which is several years. I mean for the past two months I literally spent time studying the cards in each shoe and they were all like that. Because of the pattern/grouping, I lost much much more quickly even when the count was pretty much neutral, like slowly eating away the money. Like I mentioned, I also asked other basic strategy players and also the dealers and they noticed the difference too. My original goal for posting this thread was seeking advice for how I change my play when cards are clumped. I am not accusing all ASMs are rigged. As I have been playing 8 decks many years counting down 300 something cards for each shoe, I can distinguish if the cards are randomly distributed or they are grouped in similar values. And lastly, I have lost way more than this time in the past and I didn't post on this forum or blamed ASMs. Again, everyone can have their own experience and opinion. Thanks.
Posting Permissions
-*-
Reply With Quote
Bookmarks