Moderated Coin Flipper

[Leumas]

... I am inexpertly confident ...

Now... that is a keeper... well done... QED!!!

 

[jsfisher]

No you did not... I don't believe you.

That bothers me not at all. My statement was true, and I am confident others can perform the exact same feat without any instruction from me.

 

[Leumas]

That bothers me not at all. My statement was true, and I am confident others can perform the exact same feat without any instruction from me.

No it was not... and as you said others can prove you wrong every single time.

Do you know how?

By running Coin Flipper V4.1 and seeing for themselves how it will prove you wrong about the "convergence" handwaving and wishful thinking as can be irrefragably demonstrated with the EMPIRICAL DATA similar to the ones in

• this post
• and this one
• and this one

 

[jsfisher]

Wudang... just like jimbob... and you and others... is concerned about the Math.random() being used to shuffle and pick the "cards" from the "deck of cards".

I do have "concerns", if that is the right word, for some of the more inane aspects of your code. Attempting to maintain the "true random number" nature of your static dataset by a Math.random()-based shuffle is a fools errand. Adding four "crypto random" 32-bit integers does not give you a 1-in-10,000 of getting a zero result.

The questions you continually harp upon do not address my "concerns" in any way. Instead, they attempt to misdirect the discussion towards irrelevant issues.

 

[jsfisher]

Now... that is a keeper... well done... QED!!!

You are welcome. The use of the word, inexpertly, was an admission on my part that I am not an expert on the Mathematics and computer science of random number generation. It is a topic I have studied, but I am not an expert.

I am also not an expert in wine making, analog circuitry, advanced statistics, nor the culinary arts, if that makes a difference.

 

[jsfisher]

No it was not... and as you said others can prove you wrong every single time.

Do you know how?

By running Coin Flipper V4.1 and seeing for themselves how it will prove you wrong about the "convergence" handwaving and wishful thinking as can be irrefragably demonstrated with the EMPIRICAL DATA

My claim was, pure and simple, "I did correctly predict the approximate results before peeking. It was easy to do." I also claimed others are capable of the same feat without any help from me.

The "'convergence' handwaving and wishful thinking" is straw you provided. Please stick to the actual argument rather than substitute ones of your own.

 

[Leumas]

I do have "concerns", if that is the right word, for some of the more inane aspects of your code.

Are you inexpertly confident though?

Maybe you can use some of that confident inexpertise to hazard some answers for the questions in the card post... it might make you see confidently how your inexpertly confident concerns are baseless and unwarranted.

Let's say that one picks the 20th card from the right side of a shuffled and spread out deck of cards every time.

That is not just deterministic... it is not even random

But the deck is shuffled before one picks the 20th card from the right.

Now the deck is not even random numbers... it is the same deck of cards... no?

Do you think anyone can determine whether a red (diamonds/hearts) or black (clubs/spades) card will be drawn??

So

• We have the same deck of cards... no random numbers or changing at all.
• We have the same card position picked every single time... not random or not even unknown... fully determined
• But the cards deck is shuffled before every pick... from the same deck... from the same position

Is the resulting pick (red or black) random?
Is it deterministic? If you say yes... then by whom or what?

 

[jsfisher]

Yes I did... many times... I did not say I gave the code at your command... I said I described the code... do you know the difference.

Nobody commanded you to "give" the code. You were asked about the method you chose for modeling the edge case. You ignored all such queries.

Then, after your actual source code was posted (not by you), you are asked some direct questions about it. You ignored the queries or responded with ridicule. If I recall correctly, it got worse as your TRNG implementation drew fire.

Be that as it may, some time later we get to the post you cited. In it, you make a vacuous claim then recite what was obvious from anyone looking at the source code.


If asked to describe A Tale of Two Cities, I hope you can do better than reading us the first line of the book.

 

[jsfisher]

Are you inexpertly confident though?

Maybe you can use some of that confident inexpertise to hazard some answers for the questions in the card post... it might make you see confidently how your inexpertly confident concerns are baseless and unwarranted.

With respect to the inane aspects of your code, I am confident, and expertly so, that some aspects of your code are inane.

I am also expertly confident your continually harped upon questions are irrelevant to anything I have posted in this thread.

 

[Leumas]

If asked to describe A Tale of Two Cities, I hope you can do better than reading us the first line of the book.

Thanks so very very much. Now I feel really good for the rest of the day...

WOW...

Comparing my code to the Tale Of Two Cities is beyond praise... it is accolades indeed... I just wish it were not all out of confident inexpertise though.

But hey I'll take it...

I thank you so much... you are a discerning fellow!!

Note: And as it happens... Charles Dickens is one of my all time favorite authors in the whole world.... so that comparison although out of confident inexpertise is still touching my heart. Thank you!!!

 

[jimbob]

That bothers me not at all. My statement was true, and I am confident others can perform the exact same feat without any instruction from me.

Indeed,

using something like this, for example.

https://stattrek.com/online-calculator/binomial?utm_content=cmp-true

 

[jsfisher]

Indeed,

using something like this, for example.

https://stattrek.com/online-calculator/binomial?utm_content=cmp-true

I took an even simpler approach. "Approximate result" left me with the latitude of choosing the result range.

 

[jsfisher]

Thanks so very very much. Now I feel really good for the rest of the day...

WOW...

Comparing my code to the Tale Of Two Cities is beyond praise... it is accolades indeed...

Had I done that, then I suppose it would have been high praise. I didn't; it wasn't.

You really need to work on your reading comprehension and your logic and reasoning skills.

 

[psionl0]

[IMGW=700]http://godisadeadbeatdad.com/CoinFlipperImages/TRNG_Graph_RunData.png[/IMGW]

[IMGW=700]http://godisadeadbeatdad.com/CoinFlipperImages/Crypto_Graph_RunData.png[/IMGW]

These graphs show that as the number of flips per run increases, the relative frequency of heads indeed converges to 50%. In the second chart where there are 1,000,000 flips per run, the deviation is less than 0.1%. The first chart shows a deviation as high as 1% when there are only 10,000 tosses per run.

 

[Leumas]

These graphs show that as the number of flips per run increases, the relative frequency of heads indeed converges to 50%. In the second chart where there are 1,000,000 flips per run, the deviation is less than 0.1%. The first chart shows a deviation as high as 1% when there are only 10,000 tosses per run.

Also see the one for the 10,000,000 flips... also notice that the error is not 1% (0.1%)... it is FLUCTUATING between maximum of 1% (0.1%) to minimum of -1% (-0.1%)... and anywhere in between... i.e. RANDOM.

1% of 10,000 = 100 heads of deviation (50% = 5,000 i.e. error δ of 2%)
0.15% of 1,000,000 = 1,500 heads of deviation (50% = 500,000 i.e. error δ of 0.3%)
0.06% of 10,000,000 = 6,000 heads of deviation (50% = 5,000,000 i.e. error δ of 0.12%)

100 times more coin tosses lowered δ from 2% to 0.3%... and 1000 times lowered it from 2% to 0.12%.

So what number of coin tosses will result in δ being 0?... infinity?

δ=0 is required for a deterministic non random "convergence" to 50%... and as you have seen above, 10⁷ is still not 0.

So what 10^? will result in δ=0 for a DETERMINISTIC "convergence" to 50%.... infinity??

Another question... How would you answer this question?

A single coin toss produces an unpredictable result. But we can predict the approximate results of ten thousand coin tosses. Now, is this random?

Is 10⁴ coin tosses determinable? Is it random?

what about 10⁶ or 10⁷?

What 10^? will be determinable and not random?.... infinity??

Remember anything other than δ=0 means there is a CHANCE of the GUESS to be wrong... i.e. RANDOM.

ETA: and FLUCTUATING up and down from 0.06% to -0.06% is not a determinable error... it is a RANDOM error.

[IMGW=700]http://godisadeadbeatdad.com/CoinFlipperImages/PRNG_RunsData_10M.png[/IMGW]
​

 

[Leumas]

Empirical Data vs Bare Assertions

In my reckoning Empirical Data will always trump any hand waving bare assertions.

Having faith in mathematical 300 years old theorems being what one CONSTRUES them to mean when they do not mean anything of the sort is fine and dandy.

I have asked repeatedly for anyone to determine this

n = f(p,ε) for p=1 and ε=0
Which are the values necessary for the claim that a coin toss of n coins will result in a determinable guess of the number of heads with no random error.... i.e. deterministic.
​

So far not a single person has been able to do so... neither specify what f() is in the first place.... nor give a value for n.

The claim that tossing 10,000 coin tosses will result in a determinable GUESS of the number of heads being 5,000 and tails being 5,000 and that guess being right... is... manifestly MEANINGLESS.

Why?

Because GUESSING is not determining... it is guessing.

Moreover.... and therein lies the major rub... it is blatantly debunked by EMPIRICAL DATA.

This is not to say that the 300 years old mathematical theorem is disproven... far from it.

Rather what is debunked is the fundamental and deep misunderstanding of what that mathematical theorem is saying.

It is not saying that one can determine anything nor guess with certainty anything... it is just that the deep misunderstanding of what it says is the reason that one can MISCONSTRUE the mathematical theorem to mean DETERMINING WITH CERTAINTY when it says nothing of the sort.

p being specified as < 1 but arbitrarily close to 1 and ε > 0 but arbitrarily close to 0... does not mean p=1 and ε=0 at all ... it means it will always be p< 1 and ε> 0.

And if ε> 0 and p< 1 that means heads-tails will never be 50%-50% with a 100% probability.

In other words 50-50 is a probability spread... not a determinable outcome.

In layman terms... the probability of one being right when one guesses at the number of heads being 5,000 after 10,000 coin tosses is not 100%.... i.e. it is a GUESS.

It also means that if one guesses that the heads will be 5,010 or 4,980 or whatever... it is still a GUESS with a probability of being right less than 100%.

So the deep and fundamental misunderstanding here is that Guessing with a probability of less than 100% being right, is determining and NOT random.

Guessing with a probability of being wrong... is called GAMBLING on a RANDOM event.

Not a certain determining of an outcome.

So when one says

"But we can predict the approximate results of ten thousand coin tosses. Now, is this random?"

The answer to the question is

• Predict here means GUESS
• Approximate here means not exact... and how approximate is approximate
• And yes... it is indeed random... because GUESSING at a result... even when allowed a fudge factor of error... that does not have a 100% probability of being correct... is indeed RANDOM.

The very definition of probability is randomness.... not determining.

And Random ⇔ indeterministic

Guessing an approximate value with a probability < 100% of being right is not determining the result... it is gambling on a random possibility of being right.



So the answer to the above question.... "is that random?".... is most definitively YES it is random!!!

Now... given the arrant scarcity of any forthcoming defining of f(p,ε) above let alone calculating it for p=1 and ε=0 (which is what is required for determining with no randomness of error)... what can be done?

Well... SCIENCE to the rescue.... EMPIRICAL EXPERIMENTATION is what is needed.

So... get 10,000 coins and toss them up and start tallying.

And for a good experiment to be valid it has to be repeated many times to assure repeatability and reliability... so you will have to collect the coins again and do the whole thing again and again say for 10 or 100 or even 1,000 times to be assured of the results' validity or even 1,000,000 for even better confidence.

And what are the parameters of the experiment here

• 10,000 coins tossed
• Guessing a number for the heads H_g (tails will of course be 10,000 - H_g)
• Specifying what margin of error is allowed (e.g. H_g +/- δ) where δ has to be specified.
• If after tallying the results, the number of heads (H_r) is H_g-δ <= H_r <= H_g+δ then SUCCESSFUL GUESS (S_g).... if not then FAILED GUESS (F_g).

Repeat the above say 10 or 100 or 1,000 times and tally up S_g.

If Sg ≠ 100% then the answer to the question "is it random?" is YES it is.

And of course δ has to be specified... if it is anything other than Zero then there is no need to do any of this since of course the answer to the question... "is this random?" becomes YES by default anyways.

So

come on... get the 10,000 coins and start tossing.​

Or

get cracking on calculating n= f(p=1,ε=0)... or get a math priest to divine it for you.​

Verify for yourself using 300 years old math that no one here can specify so far (let's hope one day soon)... or use EMPIRICAL EXPERIMENTATION to show that it is indeed random (i.e. probability < 100%) that guessing the H_g for 10,000 coin tosses will be determinable with no random error.

What is that you say... doing all that is prohibitive in time and cost and physical efforts?

Ok... you can use Coin Flipper V4.1 instead to do all that tossing and tallying and displaying of the results for you... you can now easily obtain EMPIRICAL DATA for yourself.

What is that you say.... you are concerned about PRNG and TRNG and whatnot and whatchamacallit... then go do it physically or figure out the 300 years old math formula above....

Failing to do any of the above 3 options but then handwaving this or that is nothing but an Ipse dixit fallacy.


Here are the results for 10 runs of 10,000 coin tosses.... can you see how RANDOM it is???






.

 

[Leumas]

Empirical Experimentation

Here are some graphs and screen shots and the raw data

[IMGW=450]http://godisadeadbeatdad.com/CoinFlipperImages/Empirical_Crypto_Data.png[/IMGW] [IMGW=440]http://godisadeadbeatdad.com/CoinFlipperImages/Empirical_Crypto_Averages.png[/IMGW]

[IMGW=700]http://godisadeadbeatdad.com/CoinFlipperImages/Empirical_Crypto_Graph_Data.png[/IMGW]

[IMGW=700]http://godisadeadbeatdad.com/CoinFlipperImages/Empirical_Crypto_Graph_Averages.png[/IMGW]​

​

Here is the raw data

Crypto Data for 10 Runs @ 10,000 flips/run & Edge Probability 0%
==========================================
Running Averages
--------------------------------
#	Heads%	Tails%	Edges%
--------------------------------
1	0.07000000	-0.07000000	0.0000
2	0.20500000	-0.20500000	0.0000
3	0.28333333	-0.28333333	0.0000
4	0.26000000	-0.26000000	0.0000
5	0.15600000	-0.15600000	0.0000
6	0.05666667	-0.05666667	0.0000
7	0.00285714	-0.00285714	0.0000
8	0.06375000	-0.06375000	0.0000
9	0.08666667	-0.08666667	0.0000
10	0.09100000	-0.09100000	0.0000
================================================

Runs Data
--------------------------------
#	Heads%	Tails%	Edges%
--------------------------------
1	50.0700	49.9300	0.0000
2	50.3400	49.6600	0.0000
3	50.4400	49.5600	0.0000
4	50.1900	49.8100	0.0000
5	49.7400	50.2600	0.0000
6	49.5600	50.4400	0.0000
7	49.6800	50.3200	0.0000
8	50.4900	49.5100	0.0000
9	50.2700	49.7300	0.0000
10	50.1300	49.8700	0.0000
================================================

 

[jsfisher]

I have asked repeatedly for anyone to determine this

n = f(p,ε) for p=1 and ε=0​

Repeating the same bit of nonsense does not make your statement any more profound. It is still nonsense. You do not get to customize definitions to your own personal liking. Doing so doesn't prove you right; it shows you to be desperately wrong. p<1 and 0<ε. The inequality constraints are important.

Which are the values necessary for the claim that a coin toss of n coins will result in a determinable guess of the number of heads with no random error.... i.e. deterministic.
​

The claim made in the opening post was, "we can predict the approximate results of ten thousand coin tosses." Stop trying to move the goal posts.

 

[jimbob]

Repeating the same bit of nonsense does not make your statement any more profound. It is still nonsense. You do not get to customize definitions to your own personal liking. Doing so doesn't prove you right; it shows you to be desperately wrong. p<1 and 0<ε. The inequality constraints are important.



The claim made in the opening post was, "we can predict the approximate results of ten thousand coin tosses." Stop trying to move the goal posts.

We can predict the approximate results with known confidence, or state how many coin tosses we would need to better an arbitrary confidence and accuracy.

And when you take this to actual physical systems relying on very large numbers, say pressure gauges, or even computer circuitry, we have very good accuracy as percentages.

 

[jsfisher]

See, Leumas? jimbob can do it, too.

We can predict the approximate results with known confidence, or state how many coin tosses we would need to better an arbitrary confidence and accuracy.

And when you take this to actual physical systems relying on very large numbers, say pressure gauges, or even computer circuitry, we have very good accuracy as percentages.