Our members are dedicated to PASSION and PURPOSE without drama!

Different way towards the regression towards the mean.

Started by Sputnik, December 20, 2013, 02:40:21 PM

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

Sputnik


Now assume that one red is one event and have the value of 1.
Also black is one event and have the value of 1.

Then 14 reds and two blacks is 3.0 STD or 14 blacks and two reds are 3.0 STD.

Assume the random flow comes in waves of STD.
Small, middle and large STD waves.

Now to the twist.

I can pick 8 random trails and match or miss-match them with the future 8 random bits.
If i get 14 matches and 2 miss-matches i would have 3.0 STD.

So when i pick the random bits and compare so will one side get ahead with 3 miss-matches or matches.
Then i play opposite as its the underrepresented regression towards the mean.
To prevent getting 14/2 events or window of 3.0 STD ...

The bets is 3 to 4 attempts depending on how the random bits unfold.
Quote

2
1
1
2
2
1
2
2

2 S 2
2 O 1
2 0 1
2 S 2
1 O 2
2 O 1 L
2 S 2 W
1 0 2

1
2
1
1
2
1
2
1

1 S 1       
1 O 2
1 S 1
1 S 1
1 O 2 W
1 S 1
1 O 2
2 O 1

2
1
2
1
1
1
2
2

1 O 2
1 S 1
2 S 2
1 S 1
2 O 1 W
2 O 1
2 S 2
2 S 2

1
2
2
1
1
2
2
1

1 S 1
2 O 2
1 O 2
2 O 1
1 S 1 W
1 O 2
1 O 2
2 O 1


Xander

Regarding regression towards the mean:

"Regression to the mean" is very often misunderstood. It does not mean that if you have a sample that is skewed one way or the other, that you will even out in the future by skewing the other way. Past random events have no effect on future random events.



There is no expectation that a current offset from expectation will ever even out. If you currently have 20 more reds than black, then your future expectation is that you will be 20 reds over EV forever. Your expectation from this point forward is always just the mean no matter what already happened. The future random walk is about the point where you are now, not about zero.

The average absolute offset (in dollars, flips, red,black or whatever) from expectation tends to get larger over time, not smaller. It's only the proportion of the offset to the total expectation that tends to get smaller. This is because the standard deviation grows slower than the sample size, as a square root function. 



-Xander.


Sputnik


But you missing one important part ...
The hovering state when it stop growing and not getting weaker just stopping growing is part of correction and draw-down.
So you have 2 against 1 or 2 in 3 probability.

The states of correction is not 1 in 2 possibility as you state above.
Is just show me your misunderstanding.

Xander

No, that's where you're wrong.  There is no reason to expect a correction in a game of independent trials.

If red has hit 10 more times than black, then the expectation is that it will forever remain above expectation by 10 more hits.

You're not alone in your misinterpretation.  Many people have made the same mistake.  I highly recommend that you read more on basic probability, gaming math, and most importantly the gambler's fallacy.  There's some quality information available out there, written by gaming math experts like Mike Shakleford (Wizardofodds),  Dr. Edward Thorp, and other mathematicians.


Best of luck,


Xander

Sputnik


I know gamblers fallacy and i know red can stay ahead for one year at one table or wheel.
You are not telling me anything new that i did not know.

You're not alone in your misinterpretation the real fact that the hovering state and the draw-down state is part of correction.
Then come in tiny, middle and large waves, no matter what experts or what you say.

I have simulate at least 1 million trails, so i know for a fact that does two states show after 3.0 STD.
The world record is 5.24 STD and guess what happen after that, yes it either start hovering or getting a draw-down.

When you say even out it has not have to do that, nothing even out 100%.
Correction comes in various random bits, that is reality.

You can read all nonsense you want, i base my assumptions on significant statistical research.
I do that as hobby and would not be fool by assumption taken from wikipedia.

Bayes

Xander,


If I read another misrepresentation of RTM I'm going to scream.  :stress:
From Wikipedia:


QuoteRegression toward the mean simply says that, following an extreme random event, the next random event is likely to be less extreme. In no sense does the future event "compensate for" or "even out" the previous event, though this is assumed in the gambler's fallacy (and variant law of averages). Similarly, the law of large numbers states that in the long term, the average will tend towards the expected value, but makes no statement about individual trials. For example, following a run of 10 heads on a flip of a fair coin (a rare, extreme event), regression to the mean states that the next run of heads will likely be less than 10, while the law of large numbers states that in the long term, this event will likely average out, and the average fraction of heads will tend to 1/2. By contrast, the gambler's fallacy incorrectly assumes that the coin is now "due" for a run of tails, to balance out.


But no need to believe anyone; like Sputnik says, do your own research and you'll know the truth.

Xander

I
Quoteknow gamblers fallacy and i know red can stay ahead for one year at one table or wheel.
You are not telling me anything new that i did not know.-Sputnik

You're not alone in your misinterpretation the real fact that the hovering state and the draw-down state is part of correction.
Then come in tiny, middle and large waves, no matter what experts or what you say.

So what you're saying is that Dr. Thorp, Mike Shackleform -professor of gaming math Univerity of LV, other mathematicians and myself are all wrong?  Ok.   I must ask though, why do you feel that you're more qualified than the people that I have listed above?  Smile emoticon
QuoteI have simulate at least 1 million trails, so i know for a fact that does two states show after 3.0 STD.
The world record is 5.24 STD and guess what happen after that, yes it either start hovering or getting a draw-down.-Sputnik
FYI, there are no world records for this.  Furthermore, if you were to observe a sufficient number of trials, you'd eventually observe a standard deviation that exceeded even 10.  Regarding the simulations, I suspect that I've simulated for more, but that's really irrelevant.  There's no real reason to run such simulations on EC's when you can easily do the arithmetic.

QuoteWhen you say even out it has not have to do that, nothing even out 100%.
Correction comes in various random bits, that is reality.-Sputnik

I didn't say that it would even out. 

QuoteYou can read all nonsense you want, i base my assumptions on significant statistical research.
I do that as hobby and would not be fool by assumption taken from wikipedia.-Sputnik

I'm sorry if you don't believe that the experts that I have listed above are up to your standards.  I'm sorry if I've upset you.   :)

But the facts are what they are. 

@Bayes,

I agree with what is written in the Wiki article.  I don't know why you believe that I do not agree.
Perhaps you should take the time to more carefully read what I have written.  ;)

Buffster

You do realize Wikipedia information should be taken with a grain of salt .... Might be accurate ... might not...




B

TwoCatSam

Quote from: Buffster on December 20, 2013, 06:39:37 PM
You do realize Wikipedia information should be taken with a grain of salt .... Might be accurate ... might not...




B

Get outta here!!
If dogs don't go to heaven, when I die I want to go where dogs go.   ...Will Rogers

Bayes

Buffster,


Wikipedia is pretty reliable on science/math, but it doesn't matter, you will find similar definitions elsewhere. Or better still, try it out for yourself.


@ Xander, is that you Snowman/Herb? I recognize the style.  ;)


QuoteThere is no reason to expect a correction in a game of independent trials.


But it always happens, nevertheless. Sputnik's strategy is to be ready when it does, that's all.