/jp/ - 2D/Random

hmmm ok

Was trying to prove that in a series of coin flips, the more you consecutively land on one side the greater the chance the next flip will land on the other, due to it eventually balancing out to 50/50.

I wrote this but the computer didn't validate this thought, at the end of the day it's just running another instance `random.randint(0, 1)`

Doesn't account for metaphysical luck attached to the coin flipper.

>>506221
thanks for all your hard work

>>506221
use a light theme nerdo

>>506221
>due to it eventually balancing out to 50/50
That's not how it works. It's not being drawn toward some true outcome, the discrepancies just become less and less noticeable with larger samples because of the actual odds.

>>506227
So what you're telling me is that's exactly how it works.

>>506223
Light themes are for norms

>>506228
I'm assuming if you're intelligent enough to try to script this that you're already aware of the gambler's fallacy. Maybe I was wrong, but you seem to be implying that there's some magnetic fate that will draw your results to an exact 1:1 ratio. It's just that when you're looking at the mass of results from a distance, all of the individual discrepancies disappear. You just see purple instead of blue and red.

>>506231
your first mistake was assuming buttmin the creationist was intelligent

>>506221
> the more you consecutively land on one side the greater the chance the next flip will land on the other, due to it eventually balancing out to 50/50
I flip a coin twenty times but don't tell you what it lands on. If I flip the same coin one more time, what are the chances it lands on heads? Would your answer change if I revealed all twenty of the previous flips were heads? Why would those flips have any impact on the final flip? How would the previous flips influence the trajectory of the coin? If I take a break for a week and flip again, do those flips still have an impact? What if I use a different coin?

> Would your answer change if I revealed all twenty of the previous flips were heads?
Yes I'd choose tails.

>Why would those flips have any impact on the final flip?
Because if you flip a coin long enough it will result in about 50/50 heads and tails, and a significant amount of only one side was flipped in a row meaning that you should start betting on the other side.

I do not care that mathematically on paper HHHT and HHHH have the same chance.

>If I take a break for a week and flip again, do those flips still have an impact?
Doesn't matter.

>What if I use a different coin
Doesn't matter, but switching the person flipping the coin would, as the probability/chance is metaphysically attached to the person.

>>506240
i'd actually bet heads because its probably an unbalanced coin if it does 20 heads in a row

>Was trying to prove that in a series of coin flips, the more you consecutively land on one side the greater the chance the next flip will land on the other
I am an incomparable homo classic cslets

>>506240
>Because if you flip a coin long enough it will result in about 50/50 heads and tails
You are just looking at this incorrectly. If I flip 20 heads in 20 tosses, I am more likely to still have flipped 20 more heads after 100,000 tosses than to have flipped an equal amount of heads and tails. In this same theoretical situation, after 1 gorillion tosses I am more likely to have flipped 30 more heads total than an equal amount of heads and tails. The absolute difference just affects the "odds" less as the sample grows.

>>506250
Look bro, math and computation in this case does not reflect reality or metaphysical probability/chance attached to a person.

>>506251
Okay, here's how you prove this and get your Nobel prize. Go flip a coin until you have your desired streak. Record the next coin flip. Repeat this until you have a usable sample and analyze the results.

>>506251
can you describe in as much detail as possible the mechanism by which previous flips affect the trajectory of the next flip.

>>506253
after you get three heads in a row the goddess of fate spins the fourth flip into a tails(about 50% of the time, maybe a little more)

>>506240
>Doesn't matter, but switching the person flipping the coin would, as the probability/chance is metaphysically attached to the person
Ok what if there is a line of 10 people with different coins in the same room. The first 9 flip and get heads. Do their flips effect the 10th? What if they are in different rooms? What if they are on different continents? What if they each flip their coin a decade apart from anyone else in the group?

>>506255
its simply the central limit theorem. you take a lot of random coin flips and add them all up together an them form a normal distribution. but when you get a lucky metaphysical streak that breaks your normal distribution then the next few flips you add will have to help you approach the center and so they have to be the balancing type of coin flip(heads or tails).

>>506257
You're confusing the distribution (the sum of heads and the sum of tails) with the outcome of a particular sequence (the exact order in which H and T appears).

Say for example, 4 coin flips, there are 16 possible sequences. Each sequence has a 1 in 16 chance of happening. HHHH is just as likely as HHHT, but if you looked at the distribution from all possible sequences, you'll have:
1 outcome of 4 H total,
4 outcomes of 3 H,
6 outcomes of 2 H,
4 outcomes of 1 H, and
1 outcome of 0 H

You can see that distributions don't care about the order of occurrence, they don't care which one came up first or last, they don't give a shit about that. Distributions *only* look at the total number of occurrences. So you can see with a distribution curve there are many more sequences with a mix of H and T which will gravitate toward a 50:50 ratio. However, every particular sequence (every particular order in which H and T is thrown) is still as likely as any other.

These are related but distinct concepts. They are congruent, they coexist, but you need to be careful where you apply each concept - you flipping a number of coins and guessing the very next one is an examination of sequence outcomes, not of distributions. If you didn't care about the order in which they appeared and were only guessing the total number of H and T, then that'd be an examination of distribution.

I hope that helps.

(you're not dumb or anything btw, i'm not trying to be rude)

>>506257
>but when you get a lucky metaphysical streak that breaks your normal distribution then the next few flips you add will have to help you approach the center
Why does it need to be the next few flips? Why can't it be a dozen flips later?

>>506262
The goddess of luck manifests herself in mysterious ways

>>506257
central limit theorem only applies to sums of independent random variables.

surely you must see the irony here.