Quote from: sqzbox on December 30, 2012, 01:29:46 AM
I write my own programs to test theories and concepts before playing.  Guess what!  I still don't play!

Fascinating! What concepts should we avoid in your opinion to save ourself some time?

welcome to the forum! 

Golly gosh - that's an interesting question.  And, surprisingly perhaps, one that has never been asked of me before.  And you know what? as I sit here and think about it I find it rather difficult to answer. Why on earth would that be I wonder, since I have been researching and studying and testing for more than 10 years! 


Computer testing is limited.  Only rule-based or, if you prefer, objective strategies can be tested.  Any strategy that has a subjective component is unable to be fully modelled and so can't be reliably computer-tested.  Also, such strategies as dealer signature, ballistics, etc. can't be tested because they rely on the physics of the system being used - that is, the physics of the specific wheel/ball/conditions under scrutiny and, for some reason, casinos don't like their players to bring computers to the table.  I acknowledge (and believe) that some highly skilled players can win consistently using these techniques but unfortunately I am not one of them. That kind of approach requires a very specific skill set and personality/character that is just not me - I guess I just don't have the patience to develop the skills necessary.


But what I can do is run Monte Carlo simulations of principles that might be able to be capitalised on for real play.  I am trying to find such a principle and while I have lots of ideas I haven't yet proven anything with a large enough edge to defeat the inherent disadvantage of the odds.  I know that there are bucket-loads of mathematicians out there who state categorically that mathematically-based systems cannot defeat the wheel - but, again, I am not one of them.  Maybe my view is based on the romantic notion that "there is more under the sky that has not been discovered ... yada yada etc." but I am convinced that it can be done.  Further, I believe that it HAS been done, but who in their right mind would publish this?  That's another topic of course so I won't discuss that any further here, but one thing I will say - NEVER close your mind to anything.  All through history people have always limited themselves by their closed-mindedness and negativity and time after time have been proven wrong.  What was it that Kennedy said about going to the moon?  Something like "we don't go because it is easy, rather we go because it is hard!" - or some such.  And Einstein's quip? Well, it was exactly that - a quip! Just because Einstein said it doesn't make it so - you have to consider the context.  And I find myself raving again!  Well, again, back to your question.


I have tested all sorts of theories.  All have come back to the probabilistic expectation in terms of results in the end.  So therefore we have to be more clever.  This is not going to be easy.  And furthermore, I doubt that anybody who has solved the problem is likely to publish the complete solution here.  Sorry, but that is a reality.  But we can support each other in terms of generating ideas and testing principles.  Actually turning that into a successful play strategy will be over to the individual of course.  Anyway, what I can say is this - patterns?  forget it.  Anti-patterns? same thing.  Insurance betting? not successful.  Anything commercially available?  nah. Follow the last, opposite of the last, etc. - complete fails. (On this latter style you can find an article I once wrote here - http://xerxx.se/oops/main/reading/rr000020.html ).


What about money management? such as progression, clever in and out, hit and run, etc. Again - insufficient by themselves.  It doesn't matter how you manipulate the chip placement, if the selection strategy doesn't have a flat stake edge greater than the odds disadvantage then they will not be successful long term.  The reason for this might not be that which you expect.  Who are the professionals in this game?  The casinos of course.  Not the dealer, or even the pit boss, but the people who construct and manage the business model.  They know that in reality gambling is all about the bank - basically, who has the biggest.  Anybody with a bank bigger than the casino's, and with unlimited betting, could take them down. Hence the reason for table limits.  If your bet has a limited cap then you cannot go high enough to win a coup.  Think about it - even a simple martingale on even chances will EVENTUALLY get a win - provided the sky is the limit.  But a cap, ANY cap, stops that strategy dead in its tracks.  Again, this is a reality.


So, Bally, on looking back over what I have written here I do not really see a good answer for you.  If you could be more specific in terms of your questions perhaps I could provide better answers - there is just so much that can be studied since the field is so massive.  Also, I am not sure that this particular thread is the right place to go into specifics and maybe start a long discussion.  As I am new to the forum I am not totally familiar with all the options in terms of the boards offered - perhaps Vic could suggest something.  I'm happy to answer specifics as far as I can, and you won't find me a negative soul - but I will douse a fire if I think something is just plain wrong.


regards
Bryan
(and yes, I am a sqzbox player!)

Thank you Bryan for a very well thought out reply. You raised loads of interesting points that I hope will be discussed by us all at some point in the future.

regards

bally

Bryan, anything I can help you with, just shout.

As for the functionality just say what you want to achieve and we'll respond 

Bryan

Hi Bryan,

Thank you for your post,it was very insightful.

Sqzbox -" After all, it SEEMS sensible to "follow the trend" which is really just the last decision, or, if you want to be a bit more sophisticated, follow the decision before last which captures a proportion of two types of trend (all the same, or alternating)."

I read your posting at the link referenced. That and what you have written here are refreshing reading to say the least. Have you considered a form of detachment from bet selection, nothing more than a guess, and the effectiveness track? Some people refer to these conditions as win streaks and losing streaks. What about using the bet selection process as a tool to produce favorable periods for the purpose of exploiting them.

I'm also curious what it was about patterns and anti-patterns that gave you the impression that they are worthless or useless? Have you ever exploited a perfect occurring pattern?

Bryan, can I just say what a breath of fresh air your writing is and how right you are..!

A very warm welcome and I for one look forward to your next clear-sighted and insightful installment..! 

Thanks so much for the warm welcome - I look forward to many thought-provoking discussions. Perhaps a brief word or two on patterns might be appropriate since Gizmotron asked my thoughts regarding them.


I think it has already been mentioned in other posts that the human brain is a pattern-matching machine. This is vital to our success in the game of survival. Without this ability every animal would fail and die out. We do it without thinking and it is extremely powerful. We see patterns in everything, which themselves are parts of larger patterns, and so on ad infinitum - there is no end to what we can spot if we actually apply a little mindpower to the exercise.  As a tool in the game of roulette perhaps it is useful but can it be exploited for consistent success?


A friend of mine recently referred me to an excellent book that I would recommend highly to you also.  Fooled by Randomness by Nassim Nicholas Taleb. I am in the process or re-reading it and this time making the effort to jot down in summary form the gems contained therein. 


Patterns are just structural overlays on what has already occurred.  There is no evidence to support the contention that once a pattern has been found it will continue - or even if a pattern appears to be emerging then outcomes will go on to form the pattern predicted.  Roulette outcomes are random - period.  Consider the coin tossing exercise (because it is easier to visualise a binary when performing a thought experiment rather than the possibility of one in 37). If we have, after a few tosses, a pattern of any kind that we spot, what is the likelihood of the next toss being the continuation of that pattern? 0.5 of course - no matter what pattern we have spotted.  So there is no advantage.  Anti-patterns are just the logical inverse and again is 0.5 so again no advantage.


Having said all that, I do acknowledge that when playing a game of chance some controls are necessary (you have to have a starter and a stopper at the very least right?) and patterns may be a useful structural overlay that can provide them for you - but make no mistake (in my view at least) these control overlays are not predictive or in any way provide an edge.  They are simply mechanisms to provide structure to your play.  Consider a calendar for example (they seem to have been in the news a lot lately).  The universe doesn't know that today is Monday, or that it is the 31st Dec, or that it is new year's eve.  This is a man-made structural overlay on the passage of the days that helps us coordinate our activities as a single human in the activities of the race as a whole - nothing more.  It doesn't help us determine whether or not tomorrow is going to be wet or dry (neither do the forecasters actually and they have a lot of science at their fingertips). 

Have I ever exploited a perfect occurring pattern?  I have to confess I have not.  The reason being, it is my belief that a pattern may continue or it may not and the likelihood of such is exactly equal to the probability of the game - therefore no advantage exists.  The research I did on this is in that article you read and, although it was about baccarat, I think the principle applies to all games.


Regarding your comment on detachment and streaks - there was a British guy whose name escapes me who wrote a nice little strategy called Gambler's Luck I think it was. He basically stated exactly that - sit around with a relatively harmless little strategy until your luck comes in and then leverage that as best you can.  This is not a bad strategy at all really, but for me and my rather scientific mind, that is too subjective and doesn't fit into my frame of reference of studying the game using logic and computer simulation. 


regards
Bryan

sqzbox, thank you for answering all my questions. My goal is to convince you that there is more to detachment than you have explained. You see, the smart player must accept the truth that attempting to follow trends offers no advantage, as you have clearly stated. Hence the need for total detachment from any bet selection process. I'm not talking about rule based or progression but any randomness characteristic. There is no advantage and no prediction offered.

All you have of use is the current nature and characteristic of the effectiveness. Every loss is a signal that you are not in a perfect win scenario. I believe it's useful to attack all winning steaks and to back down from chaos and losing streaks. Have you done much work on this? 

You clearly have no use for trends. This is exactly what you have decided to do without. Perfect patterns and trends occur. While you successfully bet on them the casino is like your own personal ATM machine. That is a fact. I know of no other opportunity gambling that is better than that. All it takes is having a strategy that pays for when it does not work.

Download Ebook: Fooled by Randomness by Nassim Nicholas Taleb in private area for friends:

http://betselection.cc/private-general-board/download-ebook-fooled-by-randomness-by-nassim-nicholas-taleb/

Another one in private area for forum colleagues.

Download to the book The Black Swan: The Impact of the Highly Improbable:

http://betselection.cc/private-general-board/download-ebook-the-black-swan-by-nassim-nicholas-taleb/

Thanks Wannawin - I hope people take advantage of this and give them both a jolly good read.  The text is nicely written and reads like a novel but don't be fooled - there are a lot of vital concepts in there even though the vehicle he references are the financial markets - the same principles apply to our subject matter here.  In particular - take notice of the references to "expectation" and what that really is.  Also the "Black Swan" - this is the statistical anomaly, or rare event that nobody expects to fall victim to but always do.  The Black Swan is why progressions fail if they are being used to attempt to overcome an odds disadvantage.  And if you want to tackle something really heavy going, try "Treatise on Probability" by Keynes (1920). Now, I don't pretend to completely understand this brilliant work - it is way above my head. But reading it can still give you some insights into how really clever people tackle the problem, and perhaps indicate that a s..tload of work has already been done and is available to us if we wish to research it.


Gizmotron - I hear you.  I think I understand what you are saying - that tracking or measuring "effectiveness" and adjusting, tuning, etc. until you are in-phase with what is transpiring can lead to a significant and real advantage.  I believe you.  I haven't been successful personally with that approach - perhaps I lack the necessary intuition.  Maybe my focus on the scientific approach has shut that down in me, or maybe it was never there, or perhaps it just withered and died due to lack of use ("use it or lose it" as they say).  This approach is, I would suggest, an entirely subjective one and therefore one that I couldn't put in a virtual lab and test.  For that reason (and only that reason) I haven't pursued that approach. 


regards
Bryan



           

Wow, what an interesting post. My gut reaction is to tell you that two things made me seek my success by working very hard at this. Simplification & patience is the secret. It is a simple thing to look at my chart and to see the best occurring trends. It is a simple thing to probe the gambling session to see if a string of wins is forming. If so then the trends are holding up. If those conditions continue then I attack and keep attacking. If they fail then I back off. To do this I must avoid digging deep holes. I back off to a single minimum side bet when any change occurs that changes the win streak. Without going into it in fine detail, that is all there is to it. It's also as far as I've taken it.

I've given away my software charting method and my list of randomness characteristics. They are posted here on this forum. I've clearly stated that following complex characteristics of trends and patterns is only a device to deliberately manipulate the effectiveness track.

In my opinion patience is almost impossible to use absent playing experience. That includes trying to guess when to attack the effectiveness streaks.


Sqzbox -" Gizmotron - I hear you. I think I understand what you are saying - that tracking or measuring "effectiveness" and adjusting, tuning, etc. until you are in-phase with what is transpiring can lead to a significant and real advantage. I believe you. I haven't been successful personally with that approach - perhaps I lack the necessary intuition. Maybe my focus on the scientific approach has shut that down in me, or maybe it was never there, or perhaps it just withered and died due to lack of use ("use it or lose it" as they say). This approach is, I would suggest, an entirely subjective one and therefore one that I couldn't put in a virtual lab and test. For that reason (and only that reason) I haven't pursued that approach. "

Sqzbox -" This approach is, I would suggest, an entirely subjective one and therefore one that I couldn't put in a virtual lab and test. For that reason (and only that reason) I haven't pursued that approach."

I'm working on it as a computer simulation & as a validation of proof for peer review. The algorithms source code will be the subject of review. My language of choice is LiveCode. It's one of the best xTalk languages. The code is almost the same as reading commented code. So the common structure, custom functions and common methods will easily be identifiable. If successful I'm sure there will be people that reconstruct it in other languages. Attempting to write human intuition and selection is no easy task.