Author Topic: A Common Error in Probability  (Read 49185 times)

0 Members and 2 Guests are viewing this topic.

Real

  • Fighting the war on absurdity one foolish idea at a time.
  • Hero Member
  • ******
  • Posts: 1416
  • Thanked: 211 times
  • Gender: Male
Re: A Common Error in Probability
« Reply #30 on: April 29, 2015, 11:27:53 PM »
Quote
For your info, I don't know what losing is (never mind losing the entire bankroll).  I The worst thing that has happened is come back from the casino with just $20 over my bankroll. I haven't experience a losing session yet, because I simply don't allow it to happen. Patience has been proven to be the player's greater asset. Nowhere in my play,  I find your theories to be valid. So I don't know what you are talking about. Your advice  probably better serves a new gambler.-Palestis

Oh my!  So you've never even lost before!  Bedazzling!



Quote
According to Real, NOBODY can win consistently,  unless you take the physics route. If you played that many spins and you are ahead of the game it's definitely not a coincidence. I doubt if Real will believe you, but I have no reason to doubt you because I'm at the same level.-Palestis

Palestis,

That is correct! Nobody can win consistently, unless they're exploiting the gaming device.  And it's not just my opinion, it's history's as well. Furthermore, you can win some serious money if you can prove that it can be done! 
$30,000 Betting System Challenge
http://vegasclick.com/gambling/betting-system-challenge.html

A positive variance after 300k spins?  Maybe, depends on how many numbers you bet, on how many spins bets are actually placed, and whether or not you're flat betting or running a progression.  An internet friend, named Kelly, had one survive over a million trials.   I've also seen it done when people have played in free mode via online sites and using an rng source from the older versions of Excel.  However, it's much like playing slots in the free mode online.  Everyone seems to win on that free app as well.  (By the way, I really doubt Bayes is playing in free mode.  I'm sure he knows the difference, but many of "the holy grail searchers" on the forums aren't aware of the difference.) 

 If someone repeated the test and won a relevant amount after one million trials from random.org, flat betting, then I'd be more impressed.  Unfortunately, too many people are working from some of the Wiesbaden and other public board data samples, where you'll likely find board misreads and or repeats.  Corrupt data samples are sometimes the source of such anomalies.   

  Furthermore, surviving after thousands of spins and winning 20, 30, or even 100 units also isn't a break through.  It's usually nothing more than variance, or corrupt data.  If someone had to play that way for a living, then they'd certainly starve to death.


Palestis, I have to say, you're on an entirely different level!  You're above and beyond the rest of us... having never lost and all.  I mean, you're definitely better than Bayes and me, because I sometimes lose.  I suspect he's lost, like me, before as well. ( Yes, really, I lose sometimes.)  Perhaps it's because the rest of us play in the real world though, ya think???

Cheers,

Really

-
« Last Edit: April 30, 2015, 05:50:48 AM by kav »
 

Bayes

  • Veteran Member
  • ****
  • Posts: 688
  • Thanked: 557 times
  • roulettician.com
Re: A Common Error in Probability
« Reply #31 on: April 30, 2015, 07:49:02 AM »
Slacker,

How many units were won over the 100k bets?
A live wheel or an RNG wheel online in fun mode?

I mostly play a no-zero RNG, even chances. With no house edge I shouldn't be making ANY profit at all over time, but I'm way up. I'd rather not say exactly how much, but I'll create some vids of playing sessions with running commentary to give you an idea of the techniques I use and the win rate. I'll upload the vids to youtube and post the links here.
« Last Edit: April 30, 2015, 02:30:45 PM by Slacker »
 

palestis

  • Veteran Member
  • ****
  • Posts: 666
  • Thanked: 496 times
Re: A Common Error in Probability
« Reply #32 on: April 30, 2015, 12:43:46 PM »
Quote
For your info, I don't know what losing is (never mind losing the entire bankroll).  I The worst thing that has happened is come back from the casino with just $20 over my bankroll. I haven't experience a losing session yet, because I simply don't allow it to happen. Patience has been proven to be the player's greater asset. Nowhere in my play,  I find your theories to be valid. So I don't know what you are talking about. Your advice  probably better serves a new gambler.-Palestis

Oh my!  So you've never even lost before!  Bedazzling!

Quote
According to Real, NOBODY can win consistently,  unless you take the physics route. If you played that many spins and you are ahead of the game it's definitely not a coincidence. I doubt if Real will believe you, but I have no reason to doubt you because I'm at the same level.-Palestis

Palestis,

That is correct! Nobody can win consistently, unless they're exploiting the gaming device.  And it's not just my opinion, it's history's as well. Furthermore, you can win some serious money if you can prove that it can be done! 
$30,000 Betting System Challenge
http://vegasclick.com/gambling/betting-system-challenge.html

I lost before as I was  gaining experience. Now days I almost never lose. That doesn't mean I always win.
I avoid losing by stopping on time. I simply exercise the roulette player's greatest advantage, You can stop anytime and walk away. Whether winning at this point or losing very  little compared to the average session winnings. There is no CONTRACT in roulette playing, to wage your bets indefinitely. And neither a contract to bet a specified amount. The player has absolute freedom to come and go as he pleases. Let's not forget that.
What history and experts are you talking about?
Show me the work of an expert, where I can read an in depth study that should've
 taken years to complete, and then I might pay some attention. With specific approaches and details. There is none. It's all hearsay. 
As far as the Vegas click  challenge you have to be na├»ve to take it seriously. Part of the conditions are to play for 200,000 spins. Impossible in live roulette. Possible  in simulation, but simulation does not allow for instant (spurr of the moment) player's input. Preprogrammed plans of action, do not resemble  real life situations. Like quitting. And waiting for conditions to change. Simulation has to  have all those factors preplanned before it can proceed. On top of it, it  doesn't allow for a player's 6th sense. Which is very much active during the game. So the possibility that someone will take the challenge does not exist.
The only way you can resolve the roulette challenge is for a team of players to get together and spend 1 year in a casino.  Then they can lay out their systems and methods and test them out in a real environment and see the results.
Since that's unlikely to happen, the debate will go on for ever.
At the end what counts is the player's own experience. Only he knows where he stands with the game of roulette.
For you all those who are winning consistently are liars. Because according to your ghost experts they shouldn't be winning. Beating the device is easy to say.
Y do you avoid posting its details, so others can comment?

 

Real

  • Fighting the war on absurdity one foolish idea at a time.
  • Hero Member
  • ******
  • Posts: 1416
  • Thanked: 211 times
  • Gender: Male
Re: A Common Error in Probability
« Reply #33 on: April 30, 2015, 03:32:00 PM »
Palestis,

There's no difference between playing 200k consecutively or off and on over several weeks.
 

Bayes

  • Veteran Member
  • ****
  • Posts: 688
  • Thanked: 557 times
  • roulettician.com
Re: A Common Error in Probability
« Reply #34 on: April 30, 2015, 03:53:00 PM »
Beating the device is easy to say.
Y do you avoid posting its details, so others can comment?


That wouldn't be such a great idea on a public forum. Casinos don't care about mathematical systems, but they are interested in methods which target the wheel, dealer etc, and would take steps if they were made aware of any weaknesses. Real obviously doesn't want to share trade secrets, and who can blame him?

The terms of that challenge are actually pretty good, but I doubt whether anyone will ever beat it. Ideally, a system should be as mechanical as possible, but it would be a lot of work programming it, and assuming it did "work", where would the fun be in playing any more? I play as much for fun as the money, but then, it's not much fun losing.
 

Bayes

  • Veteran Member
  • ****
  • Posts: 688
  • Thanked: 557 times
  • roulettician.com
Re: A Common Error in Probability
« Reply #35 on: April 30, 2015, 04:00:01 PM »
FWIW, an interesting article by Al Krigman (casinocitytimes.com)

Betting Systems: Verity or Voodoo?

A primary gambling theorem holds that betting systems can't alter expectation - expected percentage gain or loss - in independent-trial games of chance. For instance, raising or lowering bets in some pattern from hand to hand at baccarat doesn't change house advantage. Likewise, big bucks on pass with full odds at craps has the same house edge as an equal amount on pass and one or more come bets with comparable odds.

Yet, gamblers galore believe strongly that how they bet affects their chances to win. And a surfeit of "systems" can be found claiming to exploit secrets about simultaneous and sequential wagering schemes the casino bosses don't want anyone to know.

Who's right: the cerebral statisticians or the superstitious solid citizens? Surprise! Both are correct. The apparent conflict only arises because they're talking about different things.

Expectation, which betting systems don't change, is an objective criterion. It's what the casino earns by virtue of its edge. And, it's a long-term average which stabilizes after tens or hundreds of thousands of decisions - many more than a single bettor would encounter even after protracted play.

Individual success, which may be influenced by systems, is a subjective goal. It may involve a host of non-monetary factors such as obtaining free meals or maximizing playing time, along with targets ranging from breaking even or showing a small profit to doubling a bankroll or hitting a lifestyle-changing jackpot. Moreover, it's a short-term effect which may be dominated by statistical properties of the game other than expectation.

To see the implications, picture seven players, all of whom gamble two weeks in a row with budgets first of $240 then of $480. Their goals are to double their bankrolls at roulette each week or lose the money trying. They'll all bet $12 a pop, but will use different betting systems. These are summarized below, with associated probabilities of winning and payoffs.

Player System
Al $12 on a single number (2.63 percent, pays $420)
Betty $6 on each of two numbers (5.26 percent pays $204)
Carl $4 on each of three numbers (7.89 percent, pays $132)
Dee $3 on each of four numbers (10.52 percent, pays $96)
Ed $2 on each of six numbers (15.79 percent, pays $60)
Fran $1 on each of 12 numbers (31.58 percent, pays $24)
Gail $6 on 1-12 and $6 on 13-24 (63.16 percent, pays $6)

The players all have the same expectation - a theoretical loss of 5.26 percent or $0.6312 per round on their $12 bets. The casino, which takes the long view, would rate them equally. Their systems are irrelevant in this respect.

Chances the players will double their money each week before tapping out differ, however, owing to single-round risk and reward characteristics and the bet-to-bankroll sizes. The probabilities of success are shown below.

Player     $240     $480
Al     48%     47%
Betty     47%     44%
Carl     45%     40%
Dee     43%     37%
Ed     39%     29%
Fran     33%     20%
Gail     11%     2%

These probabilities suggest that to double a stake before losing it, chances of success improve as bets become a) greater longshots with larger payoffs but steeper odds, and b) higher fractions of the starting bankroll. Players having other criteria - say, extending a streak of winning games without regard to amount, minimizing the chance of going belly-up, or testing the air in a high-limit pit - might find this strategy disastrous.

Betting systems can be optimized for any specified gambling goals. High likelihood is no guarantee, of course. And, don't forget the insidious law of unintended consequences. Raising the chance of meeting a specified set of goals may have an unpleasant downside, like excessive loss when things go wrong, too long a required playing time, sacrifice of a desirable fall-back position, or sneers from dealers you're trying to impress.
 

dobbelsteen

  • Hero Member
  • ******
  • Posts: 1499
  • Thanked: 479 times
Re: A Common Error in Probability
« Reply #36 on: May 04, 2015, 11:21:38 AM »
A session of two weeks  says nothing.The final result depends on the number of spins.This example is easy to symulate on the computer. I predict, together they will lose and one or two will have a small profit. Nobody shall double his start amount.
 

becker

  • New
  • **
  • Posts: 29
Re: A Common Error in Probability
« Reply #37 on: May 24, 2015, 01:26:39 PM »
I'll upload the vids to youtube and post the links here.

Can we expect this anytime soon?
« Last Edit: May 24, 2015, 01:31:44 PM by becker »
 

scepticus

  • Hero Member
  • ******
  • Posts: 2078
  • Thanked: 433 times
Re: A Common Error in Probability
« Reply #38 on: May 24, 2015, 07:36:32 PM »
The $30,000 Challenge doesn't actually dismiss the possibility of a winning system. He says that no one who had a winning system would sell it . His target is scammers who GUARANTEE that their system wins all the time.
 

BlueAngel

  • I always express my opinion
  • Hero Member
  • ******
  • Posts: 1571
  • Thanked: 239 times
  • Gender: Male
  • Do you want truth? You cannot handle the truth!
Re: A Common Error in Probability
« Reply #39 on: May 24, 2015, 09:46:56 PM »
The $30,000 Challenge doesn't actually dismiss the possibility of a winning system. He says that no one who had a winning system would sell it . His target is scammers who GUARANTEE that their system wins all the time.

If there was a HG then would be any interest at all for its creator to reveal,publish,sell it???
I think the common sense says a big NO.
BUT,this doesn't mean that someone out there winning consistently while staying under the "radars"...
 

Reyth

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3964
  • Thanked: 1276 times
Re: A Common Error in Probability
« Reply #40 on: May 25, 2015, 02:57:56 AM »
LOL.  The players are like robots in his example.  Why?  Because he wants to have an easy time writing an article while sounding intelligent.
 

scepticus

  • Hero Member
  • ******
  • Posts: 2078
  • Thanked: 433 times
Re: A Common Error in Probability
« Reply #41 on: June 16, 2015, 08:15:51 PM »
Slacker. Just an explanation. When I read or hear about an unusual comment made by a mathematician I ask myself " Can I use that idea in  roulette ? " You  read my Birthday Problem Method " but failed  to point out that it was derived from the birthday problem supposition that if  23 people were  in a room then the probability of two of them sharing the SAME birth date was just over 50 %. I merely applied  that argument to roulette by reducing  the 356 to 37 and adjusting the maths. So , IN THEORY , there  NEEDS to be 23 already in the room and so the NEED for 8 numbers to have already  been spun . So your view that you need to start from a base of one is wrong.
It is  just an idea I put on the table  and  , like my others , put there on a take it or leave it basis.
 

Mike

  • Veteran Member
  • ****
  • Posts: 623
  • Thanked: 49 times
Re: A Common Error in Probability
« Reply #42 on: June 22, 2015, 08:45:34 AM »
I merely applied  that argument to roulette by reducing  the 356 to 37 and adjusting the maths. So , IN THEORY , there  NEEDS to be 23 already in the room and so the NEED for 8 numbers to have already  been spun . So your view that you need to start from a base of one is wrong.

I think the point trying to be made was not that you should start from a base of one but that if you calculate in advance the probability of a series you cannot then wait for the series to be partially completed and then calculate the chance of the remaining spins given what has already happened. To do that assumes that outcomes are not independent. It's just the gambler's fallacy rearing its ugly head again.
 

scepticus

  • Hero Member
  • ******
  • Posts: 2078
  • Thanked: 433 times
Re: A Common Error in Probability
« Reply #43 on: June 22, 2015, 02:18:04 PM »
No Mike, the point Slacker was making was that I WAS indulging in the Gamblers ' Fallacy  and you are doing the same  here.
I was merely taking the maths of  the Birthday Problem adjusting it from 365 Days to 37 spins. and finding that 8 was   the resultant number.  Waiting does NOT necessarily apply here . If  23 people are already  in the room then the  calculation can be applied.So, like Slacker , you misunderstand the nature of the problem.
Incidentally,it should be noted that the Birthday Problem refers only to the Probability of two people sharing the same birthday - not the certainty that they will do so.
What I was doing is what I said I did - take something a mathematician said that seemed unusual and tried to apply it to roulette.
 

palestis

  • Veteran Member
  • ****
  • Posts: 666
  • Thanked: 496 times
Re: A Common Error in Probability
« Reply #44 on: June 22, 2015, 03:05:25 PM »
I merely applied  that argument to roulette by reducing  the 356 to 37 and adjusting the maths. So , IN THEORY , there  NEEDS to be 23 already in the room and so the NEED for 8 numbers to have already  been spun . So your view that you need to start from a base of one is wrong.

I think the point trying to be made was not that you should start from a base of one but that if you calculate in advance the probability of a series you cannot then wait for the series to be partially completed and then calculate the chance of the remaining spins given what has already happened. To do that assumes that outcomes are not independent. It's just the gambler's fallacy rearing its ugly head again.
The probability of series has one value and one value only. And that value remains the same from the beginning of the series till the end, because we don't know when the expected result will happen. Of course looking for 2 results that already happened (partially completed as you say),  and they are against the expected outcome of that series (virtual loss), technically cancel the series and a new probability of series is needed for the remainder of the series. But as I have described  many times, that's not the case the way I look at it.
I actually bet the entire series. And I'm there on top of the roulette when the actual guessing/betting starts from spin 1 in the series till the end or until  the expected outcome occurs. .
I just chose to stick with the  situation where my first 2 bets for example lost and then carry the series till the end. I stop the series bet  if one of my 2 first guesses won, and then look again for a new series where my first 2 guesses lost (cost free of course). Just because I wait for that particular situation it doesn't mean that the probability of series changes. As by definition probability of series is  at least one desired outcome at any point in the series (which is unknown), and then the series ends. Whether the series ended after a LW result or a LLLLW doesn't change its predetermined value.  I arbitrarily chose a situation where after guessing AND BETTING my first 2 results were LL.
But I actually placed bets in those 2 results. it just happened that the betting amount in those 2 bets to be $0.001. Don't forget that I didn't look to find those 2 results ready made. I actually placed bets.
So instead of betting a planned $5-10-20-40-80 amount I chose to bet $0.01 - 0.01-5-10-20. For obvious reasons. The question is:  does the probability of series change if the bet amounts are of the first kind or  they are of the second kind? No it doesn't.
To make things simpler I only bet profit yielding amounts in the most frequent WINNING RANGE of the series. A fact that can only be determined after a long time empirical research.
But as long as I bet all the 5 spins in a 5 event series the probability value is always the same.
That's not the same as looking around for an LL situation first. Because it then becomes a 3 series event, and it has a different (lesser ) value.
The $ amount in each bet doesn't change the value of the series.
That is what you are failing to understand.
« Last Edit: June 22, 2015, 03:08:56 PM by palestis »