An effort to help some members get out of the worst gambing fallacy

[albertojonas]

Nothing is guaranteed, but past spins do indicate future spins, and that's not a contradiction of GF.

And I disagree with MOP that hit & run is the worst fallacy. In fact it's relatively harmless.

[ddarko]

***Nothing is guaranteed, but past spins do indicate future spins, and that's not a contradiction of GF.***

Well, there's some Voodoo for ya!

Sam

Quite the statement there 2Cats !!! Care to expand on what you mean please?

[ginger]

***Nothing is guaranteed, but past spins do indicate future spins, and that's not a contradiction of GF.***

Well, there's some Voodoo for ya!

Sam

Hello,

I'm very pleased to read this ....you are right 2.C.S.

Your luck in Roulette is fixed in the past hits , otherwise it would be impossible to play the system Dealers Signature.

Have a nice weekend

Cheers


John            Rotterdam

[Ralph]

One thing is out of doubt, the past spins affect the bankroll size.

[Bayes]

Well, there's some Voodoo for ya!

Sam, why is it voodoo?

What I said was:

independent trials follow distributions and exhibit regular patterns as defined by the laws of probability, which means to that extent they're not independent

Of course the odds don't change in the way that they do in Blackjack when cards are removed from the deck, that's not what I meant.

An example of voodoo is hit & run - expecting to miss the bad sequences by keeping sessions short. No-one can give a plausible reason why that should work, but there are mathematically sound reasons why using past spins as a guide can help with making bets. If you understand that outcomes are constrained by the laws of probability, this is obvious. Anyone who disagrees is just signing up to "scientific" dogma. 

I've proved it to myself many times. Sometimes I deliberately bet randomly, just to see if I'm kidding myself about bet selections - do they really make a difference? What always happens is that I get into trouble pretty quickly, it then becomes a matter of pure money management to try to recover losses. However, when I'm not betting "cold", but use past spins and probability in order to select bets, it's a MUCH easier ride. It's hard work because I don't use a rigid system, only probability and statistics applied to past spins. To be honest, I find it strange that people are looking for a simple, mechanical system which will let them play like a robot - boring! 

[Robeenhuut]

Math-boys* say "nothing is due at roulette".
They also say "you will be a loser in the long run". But this second claim is based on the assumption that something is due and that will happen given enough time. If you talk about the "long run", you effectively admit something is due. Either you say "in the long run Reds will catch with Blacks" or "in the long run you will loser", makes no difference. Both statements require the belief that something is due and given enough time it will occur.



________
*math-boys = persons that dismiss any creative thinking about roulette, on the basis that "no system can win, because the payout is not fair" (referring to the 2,7% house edge).


Some hope that GF is "You will be a loser in the long run" 

[Ralph]

Math-boys* say "nothing is due at roulette".
They also say "you will be a loser in the long run". But this second claim is based on the assumption that something is due and that will happen given enough time. If you talk about the "long run", you effectively admit something is due. Either you say "in the long run Reds will catch with Blacks" or "in the long run you will loser", makes no difference. Both statements require the belief that something is due and given enough time it will occur.







________
*math-boys = persons that dismiss any creative thinking about roulette, on the basis that "no system can win, because the payout is not fair" (referring to the 2,7% house edge).


Some hope that GF is "You will be a loser in the long run" 

I found this statement at roulette30. And it is easy to understand every number will show in the long run. We have some problem to know when. In that sense it is due, but statistical trials say they even out.
The even out is problematic, it can take millions of spins to a cross, and still they even out by % but not numbers of occurense.  45/55 is not strange at 100 runs, but  450000/500000 is.
At a longer run they tight up i % count but not sure by number count.

[Stepkevh]

Yes thats correct.

I've made an excel tracker for % on d/c an ec's.
They never even out, always around 31->33% for d/c and 41->43% for ec's

Stephan

[Robeenhuut]

I found this statement at roulette30. And it is easy to understand every number will show in the long run. We have some problem to know when. In that sense it is due, but statistical trials say they even out.
The even out is problematic, it can take millions of spins to a cross, and still they even out by % but not numbers of occurense.  45/55 is not strange at 100 runs, but  450000/500000 is.
At a longer run they tight up i % count but not sure by number count.

The author meant rather that it is impossible to know for sure that a player will always lose in the long run meaning that even when its proven that no system can survive gazilions number of spins its possible to be in profit playing relatively large number of separate sessions.
For me GF would mean that its possible to have a bet selection thats better than others by looking at past spins and even if spins are not independent (Marigny theories) then its possible to construct a winning strategy based on this fact without exceeding table limits.

[Ralph]

Mathematical the house brings in around 2.7% of the betting turnover, its easy to show. Its from large number of bets, and that's even out to the HE. A table can even have a loss some days.
That's not so as many calculate, it takes 2.7% from every bet. The bet is losing or winning at 100%.
The same goes for sessions and players, the are not the large number in the sense as the all casino bets. If the wheel do not have memory, it will not know who is winning, and bookkeeping to distribute the HE even among sessions and players at a even rate in short or longer time.

I do not exactly know how many bets needed to for sure reach the negative EV. That's still something expected, the expected it not allways the outcome.

[Bayes]

Math-boys* say "nothing is due at roulette".
They also say "you will be a loser in the long run". But this second claim is based on the assumption that something is due and that will happen given enough time. If you talk about the "long run", you effectively admit something is due. Either you say "in the long run Reds will catch with Blacks" or "in the long run you will loser", makes no difference. Both statements require the belief that something is due and given enough time it will occur.

There's a bit of sophistry here because "due" is used in two different senses. The GF sense of "due" is that the outcome is expected after a definite number of spins (usually no longer than the length of your progression). You may not know exactly how many, but at least you'll know when the expected event has arrived.  But "due" in the sense of "being a loser in the long run" isn't expecting any particular event - how do you know when you're a loser? there is no event which, once past, defines you as a loser. Being a loser in the long run just means that over time, the house edge will erode your bankroll, so it's more of a statement about the effects of the house edge.

In my experience, the way that the "math boys" operate is to say that anything can happen, they say things like "you could get 100 reds in a row", or "there are no rare events, all events are equally rare". The implication is that no matter what you try, it's a waste of time and effort because you'll be destroyed by the randomness, and if the randomness doesn't get you, over time the HA will.

This is nonsense and shows a complete lack of understanding of probability and statistics. It can't be the case that all events are equally rare, because "rare" is a relative term: rare compared to what? If all events were rare, the entire field of inferential statistics would be useless; you wouldn't be able to determine whether an event was random or followed some regular pattern or law, and advantage play would be impossible, even in principle.

[Turner]

To be honest, I find it strange that people are looking for a simple, mechanical system which will let them play like a robot - boring! 

I said similar in my Brett Morton post

It just reminds me of the way i used to approach roulette like i had found some magic formula that works for all roulette scenarios.
I've done far better once i left the magic world of turbogenius roulette systems to acting on very recent information.
Like in chess, pushing a pawn or controlling the centre may be fundamentally correct actions....but not in the particular position you see before you. I am starting to realise the similarities between chess and roulette
The current position you see on the maque denotes how you play.
Static systems like this don't let you afapt to the changing position
Hope that explains where i am at
Turner

[iggiv]

by the way i have never said that using hit-n-run is a guarantee to win with any method. What i wanted to say is that certain methods may perform better with it. The reason for this is as i said already. Unless it is a sophisticated advantage play, any method will usually sink with roulette after playing it for long time. That's a nature of a game with negative expectation. Actually absolutely most people with knowledge and experience admit it. So don't expect to win always, try to win from time to time, and if You can-- then it is good enough. It has nothing to do with statistics or "high school math". Those who say that a winning non-advantage-play method is supposed to win thousands and thousands and thousands spins in a row (and if it can't then it can't win at all)  -- to say a least -- have a bit unrealistic expectations. "I want it all and i want it now" sounds good, but in Queen's song only, not in the game of roulette.

and someone said "u have to know when to hit and when to run", sure, u have to chose your entrance and exit points carefully, otherwise it does not make sense. Otherwise it is just random betting and losing by house edge.

[Robeenhuut]

by the way i have never said that using hit-n-run is a guarantee to win with any method. What i wanted to say is that certain methods may perform better with it. The reason for this is as i said already. Unless it is a sophisticated advantage play, any method will usually sink with roulette after playing it for long time. That's a nature of a game with negative expectation. Actually absolutely most people with knowledge and experience admit it. So don't expect to win always, try to win from time to time, and if You can-- then it is good enough. It has nothing to do with statistics or "high school math". Those who say that a winning non-advantage-play method is supposed to win thousands and thousands and thousands spins in a row (and if it can't then it can't win at all)  -- to say a least -- have a bit unrealistic expectations. "I want it all and i want it now" sounds good, but in Queen's song only, not in the game of roulette.

I agree  but if you play just one game involving 60 step progression can we still call it hit n' run 

[iggiv]

call whatever u like it Rob. i won't play this kind of game. I am not setting terms here. I am just giving my opinion how it makes sense to play. I don't believe in progressions. And if someone uses them they would better be short. Some say "golden progression" with regression helps. But it is short as well.

I agree  but if you play just one game involving 60 step progression can we still call it hit n' run