hot /cold machines

[cddenver]

Without getting into a discussion of "nothing is purely random" (those rnd generators are, in fact, programmed by humans who are not random),  our job in any probabilistic field is to take advantage of positive scenarios, or "good luck" while minimizing our exposure to negative scenarios or "bad luck."

 
Thanks for posting, and welcome to the forum.
 
No one can argue that RNGs are written by imperfect humans, but the numbers coming out of the RNGs used in casino machines are required by law to meet specific standards of randomness as defined in the regulations.  In a nutshell, the probabilities have to match those one would see if an actual deck of cards was being used for each hand.  In addition, RNGs are required to run continuously, with thousands of numbers generated and discarded every second between plays.

This is possible,  It is even probable to do this.   Now, the next objective is to design for yourself a system of interacting with this endeavor that gives you the greatest probability of winning.   A system designed to maximize your advantage when you have the advantage, and has you betting the minimum at ALL OTHER TIMES.  (the winning bonus of the max bet to hit the max payout royaldoes not compensate for what you lose by betting the max at all negative sceanrio times). 

 
I've modeled several non-optimal strategies that are geared towards increasing the probability of winning in a session.  They do indeed allow a player to win more often, but in the ones I've checked the bottom-line dollars over all play come out worse than with optimal play.  That's because the average dollars won in a session played with a short term strategy are less than the average dollars won when using optimal play, and the average dollars lost in a session played with a short term strategy are larger than the average dollars lost when playing an optimal strategy.
 
If this is something that you've developed on your own I'd be interested in seeing the specifics of your strategy - enough info to set up a spreadsheet of my own.  We've already got quite a few threads on short term strategies.  If you're not sure if it's something that's already been written about I can post some links to threads discussing them and you can take a look.  The reason I mention that is strategy threads can get pretty long, and there isn't any sense in reinventing the wheel.

[optimalf]

John Robison, a nationally recognized expert on the operation of
video poker and slot machines, cautions that "random" should not imply
outcomes that are to a player's personal satisfaction every time:
"There are cycles on machines, but who ever said that cycles
disprove randomness?" Robison has written. "On the contrary, streaks
(called local non-randomness) are a requirement of randomness.
"If an event is random, anything can happen, including long cold
streaks, long hot streaks, short streaks and choppy results."I copied this from an article on video poker.    Just so I am not the only one saying it, but if i was the only one saying it that doesn't mean it could be wrong.The trick is to design a method that signals you to the advent of these streaks, and make a plan to act accordingly (The Plan).   Granted, nothing can predict the future.   Also, gamblers need to be cautioned as stock traders do, that in our lifetime we will never play enough hands, make enough bets, of take enough trades to even come close to approximating the "long term".   Only insurance actuaries get close, because they have large enough sample size, and even they are wrong sometimes in big ways.  Casinos get close because they have the other side of the deal.  But if they were so right on with numbers and prediction, tell me, why would any of them file for bankruptcy?  See, even the law of large numbers has some outlier events that are business killers.   But... As an individual player, you are not in the ballpark of large numbers, never will be, so you cannot adopt a strategy that includes that in it's premise if you want to survive.   This is the primary fault of the AP method in my estimation.   Insert wrath of mainstream here.....The same is true in stock trading, where forever the media, publishing companies, books, have indoctrinated countless generations into the theory that the "buy and hold" method of investing and trading actually works.   The same is true of the real estate associations that convinced people that "buying a home is the best investment you will ever make."Obviously history has shown this is not the case.   I have made my living in stocks using strategies that we created, extrapolated and crafted from different methods: arbitrage, short selling.  These are not things they media wanted people to know about over the years.  Nor were they strategies the average investor could enact with limited capital and typical clearing fees.  Anyway,,,sorry back to the point.   This "plan" would involve nearly any set of arbitrary rules that cause you to reduce or increase your risk accordingly.    As the expert said, just because you are using random numbers to generate card dealing in vp doesn't mean there are not streaks...as i have said as well, there in fact ARE these streaks and they do exist.   I set up my own method for validating this assertion as well.     My method for picking them off is crude and personal to me, and probably wouldn't  work for any of you most likely because it probably wouldn't fit your personality, tolerance for risk (heat), or logical reasoning.   So because of that i won't detail it here.   If you even believe im not wrong, you can find the way.  That's the work you must do, because you need a method that you will follow.   Those tend to be the ones you create and validate yourself. One place to start would be to define very precisely what a streak is...simply..what is a streak?  what does it look like at the beginning? and then define what a streak isn't?  remember you can't predict the future, the only thing you can do is differential those two things and act accordingly.   You will be wrong a lot of times but make it cost a little.   When you are right figure out how to maximize it.   I have done this successfully for a few years, but I also know that tomorrow it could all end...the next few years could be losers, but up until now it works...and so i continue.Thank you. 

[shadowman]

You said one thing that is absolutely true ... "nothing can predict the future". Hence, there is no "plan" that is better than any other.  Also, you are wrong about your view of the "long term". Here's some information that should help you understand:
 
http://members.cox.net/vpfree/Bank_NO.htm
 
As you can see, the "long term", or in this case having results within one standard deviation, is a function of the amount of edge. Clearly, casinos would have no problem reaching this definition of "long term".
 
Casinos file for bankruptcy just like any other business. Their income does not cover their costs. All machines could pay back 10% in a casino but if no one came into the casino it would go bankrupt.
 
You are absolutely right that streaks are part of randomness. I've mentioned that many times. But, as I previously stated, streaks are also random in length and there's no way to predict them. If you have had success which is better than random, then consider yourself lucky, because that's all it is.
 
Finally, I understand RNGs better than most. I have a couple right here on my computer than I use. Since these programs are trying to simulate true randomness they would have streaks within their output. However, these particular streaks would never be evident on a VP machine. Since the RNGs run continuously the streaks any individual sees are based entirely on their own timing. They sample less than 1 RNG generated value out of hundreds of thousands. It is truly a random sampling of a random set of numbers. Hence, it is not the machine that determines any persons results, it is their own timing.

[shadowman]

Let me take the notion of streaks one step further. As stated previously all RNGs have streaks. Since the RNG in a VP machine is programmed it will have streaks as well. However, no player will EVER see one of these programmed streaks. A bad streak of say 50,000 hands would whiz by a player while they took a sip of water. A really, really bad streak of say 200K hands would require two sips. You never see them.
 
All streaks a player does see is the result of their own timing. And, since they are randomly sampling a random set of numbers, they will also have streaks. However, there is no way these streaks are in any way tied to the RNG of the VP machine. Change a person's timing slightly and they will have different results. All it takes is sneeze, an interruption of any kind, and the rest of the results for an entire day will change. What does that do to any "plan"?
 
BTW, those who write articles about VP are simply passing on information that is well known by thousands of expert VP players. John Robison, Bob Dancer or Jean Scott do not have any extra inside information. They are simply a few of the best known experts.
 
Finally, since you are essentially claiming you can predict your own future. That is equivalent to stating you are psychic. There have been tests run that seem to indicate that certain people can make predictions better than statistical expectation. If true, you may be one of those folks and I wouldn't change anything if it's working for you.

[optimalf]

I read the article  mentioned in the last post and that math is correct.   The fact that is it correct, however, doesn't mean it applies to the real world.   That is sadly often the case with mathematics, it is a wonderful tool and even sometimes beautiful fantasy, but I need to operate in the real world and that involves, since I choose to do these things, making money and having cash flow daily or something close to that.  I respect the  approach as what I would call an advantage player method.   I saw quite early on that that approach wouldn't work for me, so I found something else that does.   Ive seen analogs in trading systems where the math, too, looks wonderful however in reality the application of that method loses money. I am not saying that the AP method couldn't be profitable, I am just saying I wont' do it that way, and from the math espoused for me it is profitable potentially in the distant future to functionally never.    In other probabilistic fields I have succeeded in finding ways that work profitably and that's the perspective I operate from.  However, its so fascinating to me that in these endeavors, as in trading, being correct and making money can be two vastly different endeavors.  If someone makes money  with any method over time then fantastic.   I am just a simple guy despite my advanced degrees in these fields.   My applied math in all of these endeavors did nothing to really help me achieve profitability.  In otherwords, most mathematics at these levels tends to not match the real results.   I quoted that "expert" mainly because what he said about non randomness was correct, not because I have any notion that he derives his income from profitably playing video poker.  The same is true of the others mentioned.   In trading, there are plenty of pundits who write books and are on cnbc, and don't make a dime from extracting money directly from the markets, and also are paid mainly to promote strategies and investment ideas that benefit investment firms  or to promote trading activity at the expense and demise of the individual investor or trader.   In my opinion, you can make this same analogy in many cases to the gambling and casino industry.  (see the movie, if you want entertainment and interesting ideas:  "broke, the new american dream"  you can find it on itunes and similar sites.  I don't derive any income from this, although one of my mentors is featured in this film.)   The same is true of video poker in my honest opinion.   In addition, the idea of playing video poker any more than 5 to 10 hours per week doesn't appeal to me for reasons of health and boredom.   Good trading and good video poker for me, are, relatively boring.   I also don't want to end up with esophageal cancer or at least try to limit that by the time I am 60.For those of you that play 40 hours per week, have any of you had your vo2 max measured at a doctors office?   The results may cause you to think differently about the time you spend in this endeavor, and adjust your play accordingly.   If you do, then you might want a different method.  My purpose was to show that there are alternative methods.  I think playing a long term math oriented strategy could work, however that is not my experience.  Even though the math is correct.  As I said for many years I have been profitable and that could all end today, nobody can predict that.Some boards like this tends to be a contest of who can spit the furthest.   I don't engage them anymore.     Some of the things  commented about my posts are valid, but not the significant areas.   As I mentioned before I leave that up to the individuals who read these things to do their own discovery.One thing I will leave this board with is one thing that has been true for me during my successful years.   I began to consistently succeed when my need to make money became more important than my need to be right.So I am comfortable with anyone saying that I am wrong, or any number of people responding like that.   As I alluded to before, my writing was to pass on my ideas to probably 100 other people, of those 100, 10 might think its worthwhile, of those 10, maybe 1 person will be able to find a method of profitably playing video poker that works for them if they so choose.    If its true that I am psychic, then I am sure others out there are as well.  I don't posses anything that others do not.My purpose in writing was to spark thinking.  I hope it has done that.  Thank you. 

[shadowman]

Claiming math somehow doesn't "operate in the real world" is nonsense. You can play anyway you want, but don't believe for a second that you have somehow rewritten the field of statistics. It's just plain silly.
 
Once again you bring up stocks. Sorry, totally different situations. The fact you confuse the two is very telling in and of itself. You really need to try and understand what you are talking about.
 
If your desire in posting here is to convince people to ignore reality then I personally hope you fail miserably.

[cddenver]

If you do, then you might want a different method.  My purpose was to show that there are alternative methods.. 

 
Well, I'm sorry, but you haven't "shown" any alternate methods.  Not one!  You've just said that there are some.  The statements you've made about unrandomness, work experience, and an advanced degree imply that you've used those in combination to develop a method (set of rules) for play that lead to positive bottom-line dollars when looking at all play over time. 
 
In an earlier post you said that you could post informatioin about your play good enough to model, but that you wouldn't because the personal views of people checking the information would affect the results.  That just doesn't make sense; rules for play are either implemented correctly in a model or not, feelings and views don't enter into it. 

[rolanddude]

Randomness...Optimalf vs. Shadowman/Archimedes  Some perspective on the two doing battle here!! Click to watch. (3 minutes)

[cddenver]

Great.  Another video NOT to watch.   That's not a comment about the contents but the video style of presenting views.  RD, as I just wrote in my previous post optimalf has not yet "shown" anything about unconventional play leading to long-term winning dollars.  So I think you're kind of jumping the gun on the issue.  Let optimalf catch up to SM by putting some "meat on the table" first.   That would be the details on how his his method of play leads to better long-term dollars than AP, when all sessions are taken into account.


 
You're flattering optimalf by calling his posts with shadowman a "battle".  He hasn't said anything of substance to impress me.  Looks like you're just trying to drum up some interest for another round of short-term strategy posts, changing the names, faces, and format (video) to present it as something new.  I'm sure the arguments will be the same, though.

[royal flush]

in response to stocks the way i play to an edge over time is writing naked puts a little under current price of large cap s&p stocks and if assigned writing call options against that stock (example i wrote in may after jpm 40 puts just expired out stock closed 40.05 wrote june jpm 37 puts for 1.08 if assigned i will write july 39 or 40 calls on that stock. the key to the game is trying to avoid stocks that could go out of business which is not always easy but over all i am playing to a big edge