The Gambler's Fallacy

[psimoes]

TG, so you wait for the numbers that hit three times (threepeaters ) in 37 spins and bet for them to hit a fourth time?

[Priyanka]

El arroz es moho y rancio . Muchos platos de duchas nos saludo a todos

That translated to "Rice is moldy and stale. Many dishes showers greeting us all". I am sure thats not what you meant

[Steve]

Actually Priyanka, its almost exactly what I said.
I was just messing with peoples confidence with Google translate.

[RouletteGhost]

which is something the General forgets (or doesn't realize). That's why, again and again, he accuses forum members of being "trapped in the box that is the gambler's fallacy".

But only in very few cases do people seem to be committing GF. What they're doing is challenging the premise of "random". Isn't that what Priyanka is trying to show in her thread - that there may be ways of looking at outcomes which may reveal that they are not quite so random as we think?

To then come along and say something along the lines of 1 + 1 = 2 is just obnoxious. More to the point, it begs the question (assumes that the thing under investigation just cannot be even worth investigating). And therefore anyone pursuing it must be "illogical".

THANK YOU

Also i think the general just gets the jollies by not reading a thread and just interjecting the same cookie cutter BS he usually does

Good post.

No matter WHAT the purpose is it moet likely is not read. And all we get is this:

[RouletteGhost]

Steve, that's not the point. The AP player challenges the assumption of randomness. Why can't this be done in ways other than just focussing on the physics?

The General will repeat his mantras which amount to saying "it just can't".

Exploiting rare patterns by creating a particular bet selection

I think it can be done

[Steve]

Yes but you are forgetting to properly test to see if your theories actually work. Proper testing involves a lot of spins and realistically you need automated software.

Also many of caleb's comments come from the knowledge of what has already been tested. He can be a prick, but he is not the troll you think he is. Any serious player should investigate what he says.

[RouletteGhost]

Yes but you are forgetting to properly test to see if your theories actually work. Proper testing involves a lot of spins and realistically you need automated software.

Also many of caleb's comments come from the knowledge of what has already been tested. He can be a prick, but he is not the troll you think he is. Any serious player should investigate what he says.

Never said he was wrong

Just most people dont have time to monitor 10 thousand spins on a wheel

Noone wants something shoved down their throat either

He has explained to me via PM how to VB and its time consuming. I have a life to live.....

I believe that there is a way to create a bet selection where betting against a pattern can win more then lose. That is something i believe can be uncovered

Could be 3 groups of 12 Random numbers betting against something that hapens less frequently.

When a thread has zero to do with fallacy and he comes in with the same cookie cutter stuff he becomes delegitamized and hated

Your a little guilty to steve. You assumed pri thread was along the lines of rbbr

[Steve]

Vb isn't time consuming. Actually what's time consuming is a 9-5 job for 50 years.

Again if you play for fun, anything goes.  But anything serious need proper testing.

[BellagioOwner]

TG, so you wait for the numbers that hit three times (threepeaters  ) in 37 spins and bet for them to hit a fourth time?

The hot number approach was always interested. Why do you say that there should be threepeater(3 times+ same in 37 spins) in 37 cycle? Is thereany statistics/mathematics backing it up that i miss?

If we jsut wait for any number to be ABOVE expected (once every 37 spins) then you can bet any number that has been shown 2 times within 37 spins, any number that is shown 3 times within 74 spins (2*37 spin cycle), any number that is shown 4 times withinh 111 (37*3)etc.

Testing out of curiosity this type of counting i have ended so far after 185 spins to be +255 Units. Not that much testing considering long runs but still interesting

[buffalowizard]

Bellagio

How have you been playing it exactly? Bet all numbers that show twice in 37 spins? How long do you bet them for?

Thanks

[BellagioOwner]

i start betting and stop betting the numbers whether they are hot or cold. once they are hit more than expected i bet them, once they are hit less than exepected i stop betting them.
so yes. starting in the first 37 spins i will bet the numbers that has shown 2 or more times so far. If i pass 37 (i'm at 259 spins now) i calculate to see how many spins it needs to be shown to be considered a hot number. for example in 260 spins now it needs a number to be hit 9 times or more to be counted above expected. so every number that has shown so far 9+ times i bet it.

now i'm at 259 spins and +334 units. to be fair though in the beginning it went down to -150 -190 units as well...

[BellagioOwner]

+152 Units in 333 spins. It stopped being so profitable. It was logical but still in front. I will stop it for now since i don't know how to program it in Java or Roulette Xtreme bots etc forlong run and many more spins. If anyone is intrested or knows how to do it feel free to reply

[Bayes]

I was going to post this in the other thread (randomer thoughts) but it's more appropriate for this one, and I didn't want to disrupt the thread (again).

Priyanka said in that thread:

My sincere advice is come to reality where the fact is spins are independent and outcomes are equally likely. 

To which my reply would be: If you take these as "givens" then any "exploration" of the concepts which you've introduced will be tantamount to committing the gambler's fallacy - "outcomes are random, now let's explore the possibility that they're not". You'd be falling straight into the general's hands.   

The only way to avoid inconsistency is to challenge the assumption of randomness. I don't know why people are so squeamish about questioning independence. Maybe it's because they assume the only way outcomes can be independent is by a "with replacement" setup - which is a fact about roulette. Therefore it seems obvious that outcomes really are independent in every possible way - which isn't a fact about roulette as actually encountered.

It's a fact about "the random game", sure. If "random" means outcomes are unbiased and independent, and therefore unpredictable, then it's a simple oxymoron to say "let's explore the possibility that unpredictable outcomes are predictable!".

Wouldn't it be better to say "roulette seems to be unpredictable, but let's explore the possibility that it's not".

There is no "generic" random game that anyone ever actually plays, there's always a particular wheel (or RNG). The general says that every wheel has some degree of bias. So what are we to make of his oft-repeated "random game"? It seems more like a stick to beat system players over the head with!

In reply he might say "That's Correct. In reality there's no actual random game which anyone plays, but forum members assume that there is a generic non-random game, because they're not looking at individual wheels like the AP does, but looking for ways to beat all wheels, which assumes that the non-randomness is built-in to the game, not just specific wheels".

That would be a good point, and if I'm honest, one for which I don't have a glib answer. It may indeed be a foolish enterprise to try to find some generic non-randomness in the "game", at least of a kind that can be proven mathematically.

However, there's another alternative, and I suspect that it might be the way Turbo plays, or Gizmo.

[psimoes]

However, there's another alternative, and I suspect that it might be the way Turbo plays, or Gizmo.

And if that doesn´t work it´s got to be G.U.T. Seriously! I´ll explain later. BTW Bayes, excellent post. Thumbs up.

[psimoes]

Why do you say that there should be threepeater(3 times+ same in 37 spins) in 37 cycle?

Thought the General hinted at that (at the count of three, attack).

A la cuenta de tres, vamos a enjambre, y atacarlo ( Turbo ). 

Numbers that "will" hit three times have to hit two times first. Numbers that hit four times have to hit three times. Some methods are based on this. It all depends on how the trot seems to go, however. There are times when the stream of uniques never seems to end, with the lone repeater finally hitting after say 14 spins. Another twelve uniques come along and you´re there betting for that number to hit a third time and losing money. There are times when six, seven, eight repeaters hit in a few spins and then nada, no third hit. These are some rather extreme situations that I have encountered.

My "serious" method goes like this: start by betting all the numbers that just hit in the last 9 spins. 1u per unique, 2u per repeater. If in the last 9 spins a number hit three times, bet 3u on that number. Every new number that hits gets 1u bet. So on the 10th spin you bet 10 numbers, on the 11th spins you bet 11 numbers and so on. Everytime one of your numbers wins you add 1u to it. Reward the winners. Sometimes there´ll be like 4 units on a number and if that number wins it will either make a generous amount of money or it will save the session when it´s going bad.
Playing like this, one notices some behaviour in the stream and by the 18th spin you´ll find numbers that hit only once are just sucking up your bankroll; even when they finally hit a second time the payout isn´t making a profit, so you´d better give up on them, leaving just the numbers with more than 1u bets on carpet. Hot numbers won´t stay hot forever, though and around the 24th spin the cold numbers start to wake up, and irritatingly so. So much that you´d rather now bet on all the unhit numbers. From what I think I know about G.U.T. , it must be really helpful as a tool to determine when to keep betting on repeaters or unhitters.