TurboGenius Gambler's Fallacy (absurd ?) Proof.

Turbo, Ka2's point is that a random selection will head towards 1/37 at the same rate as the sleepers, so it's not a contradiction. It doesn't matter that the sleepers apparently have a head start; they won't approach 1 in 37 any faster. That's where your horse race analogy falls down. I can also make an argument on your terms (ie past numbers DO matter), because when you start counting how do you know that the last X cycles (that you haven't seen) weren't great for the current sleepers? If they were, it's not time for them to start getting hot again soon, is it?

 

This is where most of you fail....
Your session (past & future spins) will only start once you start tracking, that's when your session produces a history (spin 3 of your session, is the future of spin 1 and 2 and spin 1 and 2 is the past) whatever happend an hour at the casino before you sat down isn't important to you, because your session hasn't begun yet. I know turbo will say the same thing, but because he is offline i responded. Again, once your session starts (when you start playing) you are creating your own historyboard so to speak.

 

doesn't work but thanks.
ps i'm on android

 

I really want to believe that one can gain an advantage by playing hot or cold numbers but I don't see it. I ran another couple of 1000 tests where I compared ONE set of 18 random numbers (green) with 2 previous sets of 37 random numbers (yellow, pink). All appearing un-hits (which were not part of the previous sets of 37) were marked with one (grey), all others with zero. And of course - again no difference, the two sums are almost identical.

 

Just pick one number at a time and try to stay with the hottest numbers. Hottest numbers hit 2 or 3 times per every 37 spin cycle. A hot streak tends to last 4 to 5 intervals. I'm suggesting that you learn and can see the nature of the hottest numbers. Don't just look for conditions of wishful expectations. Look to see what is happening in activity. Forget the math. The landscape will tell you where the action is.

 

i'm not into these charts and things, but i agree there is no fallacy. the mathematicians predict down to decimal points, what the house edge is in each casino game. the play will eventually zone towards those predicted odds. so how can mathematicians say that play cannot be predicted? do people seriously believe heads could keep coming up ad infinitum on a two-sided coin, just that it hasn't happened yet? well mathematicians, keep believing in your fallacy that you prove to yourselves doesn't exist!

 

karumba,

It's a common mistake among system players to confuse the Law of Large Numbers with Statistical dependence. What you're talking about is the Law of Large Numbers. Yes, outcomes are "predictable" in the sense that they will converge to the expected probability of the distribution. This is basically Turbo's argument (although it doesn't apply to a small sequence - there is no "Law of Small Numbers"). However, that doesn't mean that outcomes are dependent. Look at the definitions:

The Law of Large Numbers
In probability theory, the law of large numbers (LLN) is a theorem that describes the result of performing the same experiment a large number of times. According to the law, the average of the results obtained from a large number of trials should be close to the expected value and will tend to become closer to the expected value as more trials are performed.
Importantly, the law only applies (as the name indicates) when a large number of observations is considered. There is no principle that a small number of observations will coincide with the expected value or that a streak of one value will immediately be "balanced" by the others (see the gambler's fallacy).

Statistical Independence
Two events are independent, statistically independent, or stochastically independent if the occurrence of one does not affect the probability of occurrence of the other.

Notice that statistical independence is about at least TWO events and their relation to each other, or in this case, the lack of relation. The law of large numbers isn't about events, but about the size of the sample (e.g. number of spins). It means that as the number of spins increases, the average result gets closer and closer to the expectation. So in 37 spins #23 might not show at all, or it might show 5 times, neither of which is the expected number of times (ie, once). But in 10 * 37 = 370 spins it's likely that the number of hits for #23 will be closer to the expected number of shows (10). After 37000 spins, the proportion of #23s will be much closer to 1/37 than in the previous two samples, and after 3700000 it will be closer still.

That is very predictable, but it has NOTHING to do with dependence. Dependence means that there is some relation between event A and event B such that the probability of one is affected by the occurrence of the other. It means there is some regular, predictable pattern. If the event A is no occurrences of #23 in the last 100 spins, does this affect the probability of seeing at least one #23 in the next 100 spins (event B)? If your answer is yes, then how do you explain it? Appealing to the Law of Large Numbers doesn't help because it's the law of LARGE numbers.

 

& that's flat bet, only.

 

An army may be likened to water, for just as flowing water avoids the heights and hastens to the lowlands, so an army avoids strength and strikes weakness

Lao Tzu

 

Thanks for explaining that MJ. But where in the Gambler's Fallacy does it mention it's referring only to small samples? I'm not undermining you. I'm just asking a further q.

 

Nevermind

 

If outcomes are independent it doesn't matter how big the samples are. Players commit the fallacy because they think there is a law of small numbers, meaning that in small samples you should expect to see the "balancing out" occur just as you do in large samples. But this balancing out doesn't happen because previous spins affect future spins. That belief is what the GF is.

 

The law of large numbers is a fallacy on it's own.
Because you are only playin a small pool of spins anyway with eqch casino visit.
As soon as i reach a profitpoint, i'm goin' home.
Doin't it for the past 2.5 years and haven't faced one day where it didn't happend, so or i'm extremely lucky or i'm doin' something right after all

 

thanks MJ. I did some more reading. I understand the Falacy now. It's just not clear from the brief initial explanations, how the Falacy can be true over the long term (which it's not). But it definitely can be true over the short term. Cheers

 

karumba, but as I wrote in my previous post, it doesn't matter whether you're talking about the long-term or short-term. Spins are independent regardless of how many you look at. It's no less a fallacy if you look at the previous 10000 spins. Suppose in the last 1000 spins reds heavily underperform. Does this mean that you can expect to see a higher proportion of reds in the NEXT 1000 spins? In all probability, yes. But this is only because the expectation is 50:50 and the expectation is always the most likely result. The expectation is always the same (for any given sample size) and doesn't depend on the prior outcomes.

I admit that this can seem contradictory, but it's really not. See this question on stackexchange and read the answers & comments following it for further clarification.

 

Unlike MJ, if you can't explain things in a simple manner GG, what does it say about your intelligence?

 

Yes, I agree

 

What you posted IS the definition of contradiction.
Like everyone else in this so-called debate.. you agree that a location should appear at a given rate,
but at the same time say it's fallacy to expect a location to appear at a given rate.
So number #2 not showing for 120 spins, then 90 spins, then 49 spins means it's average
is 1 in 86.3 and even you agree it is expected to appear 1 in 37 short or long term *doesn't matter*.
So betting on it now is a fallacy.
I call BS and show how in order to achieve 1 in 37 the number WILL appear below 37 spins multiple
times, working towards the 1 in 37 we all agree with - yet I'm wrong.
It's such a powerful fallacy that even common sense and basic math can't back it up I suppose.

 

Turbo, the point I was trying to make -- and I agree it can seem confusing -- is that a location will appear at a given rate REGARDLESS of what appeared in previous spins. There's no contradiction in this. Waiting for prior losses isn't going to change the expectation, which remains constant. To make this concrete, I suggest that you and anyone else who agrees with you repeat your procedure of getting 37 spins and adding the no-shows as they come up in the next cycle, BUT, when a no-show appears, add a random number (or one from a fixed preselected list) and also bet on it. That way, the two "systems" will be identical (in that you're always betting on the same number of #s at the same time) except that in one system the numbers bet will apparently have no connection with previous cycles. Which system makes more profit? You could have a rule that you stop the cycles when there are only 3 numbers left, then start over. This will take about 2.5 cycles on average. Of course, to mean anything you will have to do this more than once. Can you code it in RX?

 

Actually I had a question for turbo, but after I posted, I found an answer.. so I wrote "nevermind". I guess finding out an answer myself make me pretty smart. Dummy