Is random real?

[sfs]

Edit: btw, I think your use of the word "pretending" is presumptuous based on information given. Do you have information to suggest otherwise?

You have offered no reason to expect to find a particular relationship involving these numbers. In the absence of any reason to expect it, I conclude that someone noticed the pattern without looking for it particularly. This is a standard blunder in science; it's why, in a any well-designed study, you have to specify in advance just what it is you're looking for. Otherwise you can always find something odd in the data and declare a significant finding.

 

[oi_antz]

You have offered no reason to expect to find a particular relationship involving these numbers. In the absence of any reason to expect it, I conclude that someone noticed the pattern without looking for it particularly. This is a standard blunder in science; it's why, in a any well-designed study, you have to specify in advance just what it is you're looking for. Otherwise you can always find something odd in the data and declare a significant finding.

Is that a problem? Does it make the finding any less significant or something? (I really don't understand why you have made a point of it, please be patient when you explain it to me). Thanks.

 

[Loudmouth]

Is that a problem? Does it make the finding any less significant or something? (I really don't understand why you have made a point of it, please be patient when you explain it to me). Thanks.

It does make it less significant. If I took a sample of 10,000 people and collected data without looking for anything in particular, I could find correlations like red hair is caused by living in odd numbered houses or having dogs causes leaky foundations.

When you can predict a specific pattern BEFORE you collect the data and have a mechanism in place, then you have something significant.

 

[oi_antz]

It does make it less significant. If I took a sample of 10,000 people and collected data without looking for anything in particular, I could find correlations like red hair is caused by living in odd numbered houses or having dogs causes leaky foundations.

When you can predict a specific pattern BEFORE you collect the data and have a mechanism in place, then you have something significant.

Why is that? Is the data not the same?

 

[Loudmouth]

Why is that? Is the data not the same?

A random sampling will always have false correlations if you dig deep enough. The focus of any scientific study is to limit the effects of these false correlations as much as possible. If you don't have those safeguards in place before you gather and analyse the data, then you will find many false correlations.

For your example, you had no reason to expect to see those relationships between those values. You have an extremely small sample base, and you are only using a tiny, tiny portion of the data that has been singled out only because you find it significant after you look at it. Such an approach is destined to find false correlations.

 

[sfs]

Is that a problem? Does it make the finding any less significant or something? (I really don't understand why you have made a point of it, please be patient when you explain it to me). Thanks.

In science, by "significant" we usually mean that the finding is unlikely to have occurred by chance. If you're only testing one possibility, then it's easy to tell whether the result was significant. If I announce that I'm going to draw a ten of hearts, and then draw a single card from an ordinary deck and it really is the ten of hearts, that's surprising result: there was only 1 chance in 52 that I would randomly draw that card, and you'll suspect that something else was going on (e.g. I cheated, or was doing a card trick of some kind).

But when you don't specify in advance what you're looking for, then it's a lot harder to tell whether your finding is really surprising, because there are an enormous number of possible things that might look interesting. If you draw three cards, and then note that they're all red, is that surprising? What if they had all been black instead? Or had added up to a number divisible by 7? And if, instead of drawing cards from a deck, you're picking numbers from the universe, then the number of possibly interesting combinations is very, very, very large. It's really not surprising, after all, that some of them turn out to happen.

 

[oi_antz]

A random sampling will always have false correlations if you dig deep enough. The focus of any scientific study is to limit the effects of these false correlations as much as possible. If you don't have those safeguards in place before you gather and analyse the data, then you will find many false correlations.

For your example, you had no reason to expect to see those relationships between those values. You have an extremely small sample base, and you are only using a tiny, tiny portion of the data that has been singled out only because you find it significant after you look at it. Such an approach is destined to find false correlations.

I am afraid I don't get it, and perhaps because you are making general statements without giving specific details. You said that I have an extremely small sample base, and I am using a tiny, tiny portion of the data that is singled out because it is found to appear significant. I can accept that. What I don't understand is why you are saying that it is not significant on one hand, but on the other hand you are saying it has been singled out because it appears to be significant.

Do you think it is possible that I am being overly pragmatic and you are not? The fact still exists that 3:11 = 27.3% = 27.3 rotations of earth to one orbit of moon, of which a mile fits the ratio when multiplied by 360, and the odds of these numbers forming a pattern is amplified by the odds of this (attempted expression):

1 : (total population of human race)*(number of days each person lived)*(number of decisions each person makes in a day).

(A number that I don't know can be expressed or sufficiently imagined). It does seem significant.

Here is an idea: Is the 360 a variable which is causing you to doubt the significance of this fact? If a kilometre was used instead, it would require ~579 instead of 360.

 

[oi_antz]

In science, by "significant" we usually mean that the finding is unlikely to have occurred by chance. If you're only testing one possibility, then it's easy to tell whether the result was significant. If I announce that I'm going to draw a ten of hearts, and then draw a single card from an ordinary deck and it really is the ten of hearts, that's surprising result: there was only 1 chance in 52 that I would randomly draw that card, and you'll suspect that something else was going on (e.g. I cheated, or was doing a card trick of some kind).

But when you don't specify in advance what you're looking for, then it's a lot harder to tell whether your finding is really surprising, because there are an enormous number of possible things that might look interesting. If you draw three cards, and then note that they're all red, is that surprising? What if they had all been black instead? Or had added up to a number divisible by 7? And if, instead of drawing cards from a deck, you're picking numbers from the universe, then the number of possibly interesting combinations is very, very, very large. It's really not surprising, after all, that some of them turn out to happen.

Ok, good explanation, this is something I can follow. So this began by my objection to having been told I was pretending that an outcome was an intended outcome. I objected to the word "pretend". I still object to that word. If anything, I am making an assumption that the outcome was destined to be that way due to some absolute force of influence (a very vague phrase on purpose). If I was pretending this, then you would expect to see some evidence that I might be wrong. But, I have not seen such evidence and I haven't noticed anyone attempting to provide it.

Therefore, I feel it is right to sustain my objection to that word being used (so far as I understand the word to mean).

 

[Elendur]

I am afraid I don't get it, and perhaps because you are making general statements without giving specific details. You said that I have an extremely small sample base, and I am using a tiny, tiny portion of the data that is singled out because it is found to appear significant. I can accept that. What I don't understand is why you are saying that it is not significant on one hand, but on the other hand you are saying it has been singled out because it appears to be significant.

Do you think it is possible that I am being overly pragmatic and you are not? The fact still exists that 3:11 = 27.3% = 27.3 rotations of earth to one orbit of moon, of which a mile fits the ratio when multiplied by 360, and the odds of these numbers forming a pattern is amplified by the odds of this (attempted expression):

1 : (total population of human race)*(number of days each person lived)*(number of decisions each person makes in a day).

(A number that I don't know can be expressed or sufficiently imagined). It does seem significant.

Here is an idea: Is the 360 a variable which is causing you to doubt the significance of this fact? If a kilometre was used instead, it would require ~579 instead of 360.

Why did you round off the distances?

 

[oi_antz]

Why did you round off the distances?

With the 579? I thought it was beside the point I was making, and I had not calculated to a high degree of accuracy anyway. I think I see your point, and I think it reinforces my question. And, I think I know the answer to my question.

What do you think?

 

[Elendur]

With the 579? I thought it was beside the point I was making, and I had not calculated to a high degree of accuracy anyway. I think I see your point, and I think it reinforces my question. And, I think I know the answer to my question.

What do you think?

This also applies to the 360 ratio. Every calculation you've done is an approximation.

Why did you choose to round off at those values?

 

[oi_antz]

This also applies to the 360 ratio. Every calculation you've done is an approximation.

Why did you choose to round off at those values?

I did not make the decision initially, but I did accept it when I discovered it. The reason I found it acceptable to do so is because I found the discrepancy sufficiently tolerable, as it does not impact the overall observation. Just as I can make an observation that in nature, there is symmetry everywhere, yet no two sides are identical mirror images. The discrepancy is sufficiently tolerable and does not impact the overall observation that most things in nature demonstrate symmetry.

 

[Elendur]

I did not make the decision initially, but I did accept it when I discovered it. The reason I found it acceptable to do so is because I found the discrepancy sufficiently tolerable, as it does not impact the overall observation. Just as I can make an observation that in nature, there is symmetry everywhere, yet no two sides are identical mirror images. The discrepancy is sufficiently tolerable and does not impact the overall observation that most things in nature demonstrate symmetry.

The discrepancy is tolerable?

You've arbitrarily chosen a scale and deemed it better than another, in this case your rejection of ~579 instead of 360.

Why didn't you round the kilometer estimate to 580? That's a much cooler number you know. Way better than 360. Not to mention that 580 > 360, which really sums things up nicely.

 

[oi_antz]

The discrepancy is tolerable?

You've arbitrarily chosen a scale and deemed it better than another, in this case your rejection of ~579 instead of 360.

Why didn't you round the kilometer estimate to 580? That's a much cooler number you know. Way better than 360. Not to mention that 580 > 360, which really sums things up nicely.

Rounding from 579.3 etc to 580 is not rounding to nearest whole. My purpose for rounding was to simplify the number for readability so as to not distract from my point by requiring the reader to make sense of detail. You have shown me here that it was inefficient for you. I don't think that is the point you are trying to make though, can you let us know what that is?

 

[oi_antz]

Rounding from 579.3 etc to 580 is not rounding to nearest whole. My purpose for rounding was to simplify the number for readability so as to not distract from my point by requiring the reader to make sense of detail. You have shown me here that it was inefficient for you. I don't think that is the point you are trying to make though, can you let us know what that is?

Oh, and 579 is a cooler number than 580, IMO

 

[Elendur]

Rounding from 579.3 etc to 580 is not rounding to nearest whole. My purpose for rounding was to simplify the number for readability so as to not distract from my point by requiring the reader to make sense of detail. You have shown me here that it was inefficient for you. I don't think that is the point you are trying to make though, can you let us know what that is?

My point is that if you go into a situation looking, you'll find.
I'd be more surprised if we wouldn't find any kind of patterns in nature. The patterns which matter are those who suspect exists and are able to show, without knowledge in advance.

This is especially important to keep in mind as we humans are demonstrably good at finding patterns, often where there are none (you could read a fun demonstration here). Another example, here you have a random string of 1's and 0's:
11110001
10011001
01111101
00111000
10111001
11000011
01000101
01000101
10001000
00110101
I see a river, the 0's, surrounding a small island of 1's. Coincidence? Yes.
You'll also notice that there's at least two 0's and 1' per row. Conclusion? None.

Oh, and 579 is a cooler number than 580, IMO

 

[JimFit]

The physical laws predict randomness. That is what you keep missing.

Random doesn't mean "anything can happen", in case you were under that illusion.

Randomness MEANS NOTHING CAN HAPPEN, if you don't have any laws, constants how do you expect to have something?
Everything random you describe is part of a Deterministic Universe, either the causes are too many to predict it or it isn't known yet.

 

[Loudmouth]

Randomness MEANS NOTHING CAN HAPPEN,

That is your made up definition.

if you don't have any laws, constants how do you expect to have something?
Everything random you describe is part of a Deterministic Universe, either the causes are too many to predict it or it isn't known yet.

Empty assertions.

 

[Loudmouth]

Ok, good explanation, this is something I can follow. So this began by my objection to having been told I was pretending that an outcome was an intended outcome. I objected to the word "pretend". I still object to that word. If anything, I am making an assumption that the outcome was destined to be that way due to some absolute force of influence (a very vague phrase on purpose).

I wasn't implying evil intent on your part. "Pretending" in this context is equivalent to "a mistaken assumption".

If I was pretending this, then you would expect to see some evidence that I might be wrong. But, I have not seen such evidence and I haven't noticed anyone attempting to provide it.

What are you trying to imply with these measurements, and what evidence do you have to support that conclusion?

sfs' example of the cards is a really good one. It would have been impressive if you had predicted these relationships between the Earth and Moon BEFORE they were made just as it is impressive to predict the first 5 cards pulled from a deck. If you predict them only after they have been observed, then it really doesn't hold up.

 

[sfs]

Randomness MEANS NOTHING CAN HAPPEN, if you don't have any laws, constants how do you expect to have something?
Everything random you describe is part of a Deterministic Universe, either the causes are too many to predict it or it isn't known yet.

That's not what quantum physics says. Do you have some reason for rejecting it?