TurboGenius Gambler's Fallacy (absurd ?) Proof.

In the other section we are 23 pages into arguments about Gambler's Fallacy.
I decided to run off spins - that anyone can do to show why there is no fallacy.
If the naysayers want to debate this data and show their own - have at it.
I played each number until they "caught up" or came close to. Not every number
will catch up - but that isn't what the "fallacy" is about. It states that past spins
and results don't effect future spins/predictions.
So here are the results.


I have highlighted every time a number appeared better than 1 in 37.
See anything strange ?
For the initial 37 spins...24 out of 37 produced better than expected results. Normal.
For the 2x group spins, of the 13 number left in play - 8 of 13 did better than expected.
For the 3x group spins, of the 9 numbers left in play - 5 of 9 did better than expected.
For the 4x group.... of the 8 numbers left in play - 5 of 8 did better than expected.
For the 5x group.... of the 7 numbers left in play - 4 of 7 did better than expected.
For the 6x group.... of the 7 numbers left in play - 6 of 7 did better than expected.
For the 7x group.... of the 7 numbers left in play - 4 of 7 did better than expected.
For the 8x group.... of the 6 numbers left in play - 3 of 6 did better.
For the 9x group.... 4 of 6 did better.
For the 10x group...4 of 5 did better.

The player always has the advantage - winning more times than losing.
This happens because numbers that have appeared less than they are expected -
will hit multiple times and more often then expected as they "catch up" to the
1 in 37 we all agree they will average at.
It's not rocket science, but supposedly the heated debate is real.
Even the coldest number (0) after being a two time long term sleeper -
repeated 16,13,9,36,14 spins. 5 wins to try to make up for the slow start.
#4 ? 72 spins to show and then 6 spins later repeated.
#7 ? 81 spins to show and then 9 spins later repeated.
#13 ? 88 spins to show and then 4 spins later.
#21 ? 133 spins to show and then 2 spins later !!

Sure - if there is a fallacy here, it is invisible to the naked eye and to the common
sense of anyone taking data and looking at it.

 

Far out. Now that's telemetry that I can understand.

I hate reading explanations of what intentions means.

I can just spend time on it and learn.

 

Hey Turbo?

What do you think about only going after one number at a time?

You can change numbers in the middle of a session but you must bet every spin for around 300 spins bet in that session.

Still just one number. It's all about flat betting and hottest number hunting.

 

This is all nice and dandy but this is (as always) an extreme example. I mean 24 numbers havent hit a 2x in all these spins???

And could you answer me this. According to your math if we have 13 numbers that have showed 2x they give you better odds because you are not picking from 37 numbers right but from 13.

3 spins later one of the 13 numbers showed. Now what were the odds. 1:14 or 1:39?

 

should be 1:13 or 1:39 of course. I cant edit post yes turbo typos do happen...

 

That some data that works for that system i posted. Play them once they appear and bet enough to cover the losses. Put the right mm on it and You'll have a Nice casino visit

 

Odds, 1:37 , ....
It's more like what random create almost every time we play...
Does those 13 have more odds to appear ? Not really. But random doesn't really care about that .

 

Those numbers weren’t tracked past the first cycle as they were already “caught up”.

Only numbers that were “behind” after the first cycle were tracked to see how long it would take them to return to the expected hit rates…which they all did.

 

A test done just now ...all kinds of variance was present....yet it won

 

Here is something to think about Ka2 -

Let's say you play the game as intended and pick whatever you are betting on and the wheel spins.
If you're playing numbers, you have to win better than 1 in 35 because of the payout and there
are 37 numbers. This is the only point the naysayers have and keep coming back to.

Is having to win 1 in 35 out of 37 numbers my only option when it comes to how to play ?
They never really consider tracking or spins where no bets are placed.
If they do consider this, they say it doesn't matter.
So what if I track 37 spins and now have 12 that haven't shown. If I begin now, I
only need a number to appear 1 in 74 to win. We know the average is 1 in 37
I only have 12 possible numbers that could achieve this, not 37.
If I wait 2 cycles and begin, I only need a number to appear 1 in 111 spins to win.
After 2 cycles there are usually 6 numbers that haven't appeared yet.
So I have 6 numbers and only need one of them to appear 1 in 111.
And if the math is correct (and it is) I'm expecting these numbers to appear
3 times because they are behind what is expected. The all won't - that's not
how it works, but a few of them will and I don't need them to appear 1 in 37
anyway - I need them to appear 1 in 111 which is completely logical and not a fallacy.

So I'm running off spins, 74 have gone by and I have some sleepers. Here is what happens.
(doing it as I type, as I always do)

#0
#10
#18
#23
#29
#32

on spin 81, #10 appears and I bet on it. It's not appearing 1 in 37, it's 1 in 81
on spin 88, #32 appears. It's showing 1 in 88, not 1 in 37. I'm betting it.
on spin 99, #29 shows. It's 1 in 99, not 1 in 37. Adding it.
on the next spin it repeats. It's 1 in 50 now and hasn't caught up but it doesn't matter.
I'm done playing it even though it never "caught up" and I won on it. The math isn't broken
and I don't need all the numbers to catch up in order to win because I never played #29 until
spin #99. I remove the bet and carry on with +4 units to this point.

Just as example of how it's never a 1 in 37 game unless you sit down and play from spin #1
and randomly pick a number to play.
And contrary to what many anti system people think, a number doesn't have to be hot to
win on flat betting. *yes, read that again*
I'll run off another example.



These are ALL wins on cold numbers flat betting.
I skipped 2 cycles and flat bet only - and only on cold numbers.
I removed any number that "caught up" which all but 2 did and that's not
even the shocking part.
The 2 coldest numbers are the ones that made all of the wins in the balance
spike at the end (spin 305 to the end)
Impossible, right ?
The last two I'm betting on are #14 with 9 wins out of expected 11.38 and -0.71 standard
deviation and #32 with only 8 wins of expected 11.38 and -1.02 standard deviation.
Neither of these are "hot" in the 421 spins.... UNLESS you also see that I didn't bet on them
for the first 74 spins. I didn't lose 148+ units on them because I never bet on them for
2 cycles. #32 went 214 spins before it appeared and this is what happened flat betting
on it. It's still a cold number - it's trying to catch up but isn't and probably won't.
I don't need it to. I only need the 1 in 37 average to be a real thing - which it is.
The other number went 133 spins without appearing and those two numbers made
the spike in the balance. Not because they were hot overall, but cold and catching up.

I wish more people would understand this before they say there is a Fallacy or that
past spins don't matter. And yes someone will say I hand picked this chart and I didn't.
I typed this up as I ran the spins. Below are the spins, I assume they are time stamped
and should almost match the time this post was made.

 

If continued on, #32 "caught up" at spin 571.
#14 "caught up" at spin 891
But again, neither of these are "hot" numbers when looking at the entirety of the 891
spins - at best they are "average". And this is flat betting only.
Feel free to tell me about Fallacy and how absurd it is and complete nonsense.


Clearly flat bet - the summary shows this.
$25.00 bets per number, max was 5 numbers played so $125
and player's Net is +23.74% and NOT the house edge -2.7%

 

Ok I programmed the above (see screenshot)





I used your rules so betting after a number was "behind" and stopped with that number when it has "caught" up. After a few 100 cycles what do you think the odds of winning were?

 

Denzie, two questions

how did you proceed from 37 -- remove the number - after a hit, once in positive, once picked up for a few hits correlated to overall exposition; when have you restarted the whole thing, after all numbers got the near-average 1/37 removed on-by-one along the way, once the exposition in positive or near-zero
.. something else entirely ?

 

@Denzie, also

you began betting on all unhit from 38th spin straight-away or once shown; then the above ..

 

Ok I had some time left this morning and adjusted the first test. Now it chooses 37 random numbers each and every spin! It then goes to step 2 if the average is below 37 it stops otherwise it continues, just like turbo's idea.

Here is a screenshot from a random play.



What is the difference? Nothing it is exactly the same, Why? Because thats how random works! So @turbo there is no proof! It IS a FALLACY!

Play this random system for 100's cycles and it also shows 1:37...

 

By the way dont take my word for it. Pick a random number each spin play it until a hit. Note down that number go to step 2 if the average is above 37. Stop if it is below or at 1:37.

Compare your results with mine and turbos, according to turbo there must be a difference now you can see for yourself!

 

Why people are continuing to engage (directly or indirectly) in this un-winnable argument is beyond me.

 

@Ka2 now tell me why did you cut the numbers eg. #6,19,20 so quick even though they did not show profit yet (although your ratio avg touches on or below 1/37) .. & ran numbers eg. #26 for so long despite profiting already

eg.

26#140,78,21,4,12,25,(11,51,78,9,58,112,6,34,2,28,28,15,14,33)
(+4)-78+15+32+24+11

19#45,48,13
(-25) -48+23

20#44,45,15
(-24) -45+21

 

I see that by spin120 there's more than enough profit to close, notwithstanding the potential hits on numbers you haven't ran further.

Out of that, 9 unhits gets resolved with a single hit, 3 still show a slight negative balance, 2 are out of equation since first appearance is way beyond ≈120th spin; the only one that adds substantially to the negative is #9 (-75), but even with the addition of #13 (-21) ---

the profits overweigh.


Further I haven't looked as you haven't provided sufficient data.


So I don't know what the big deal is about it as you probably opted to display (one of the) worst examples/occurences.

 

All math will do is confirm math. It has nothing to do with winning or prediction. Math is not magical. All you have is luck and targeting good luck streaks. The pit bosses always know when you are in a lucky streak. You should know too. Big deal if from pretzel logic you can prove fallacy. You can take the same twists and turns and prove $kill.