Does probability really exist?

I was thinking about probabilities the other day, especially the coin example where a coin has a fifty/fifty probability of landing on either heads or tails; But then it dawned upon me that this probability is hampered right from the beginning. The fifty/fifty probability only takes into account two variables, that being the heads side and the tails side. This ignores all other variables, such as force of the flip, wind speed and direction, gravitational differences if flipped at various heights on the planet. I can think of a bunch of different variables that play a huge effect on the so called probability of a coin landing on one of two sides.

This was just a quick thought, perhaps I just don't understand probabilities as well as I thought I did. If not, then can someone explain it to me a little better so I can learn more.

I think probability is a really flawed concept. Anything can happen at any time. But it is a really hard topic to explain, so hopefully some fellow ATSers can help us out.

Originally posted by CaninE.G
I think probability is a really flawed concept. Anything can happen at any time. But it is a really hard topic to explain, so hopefully some fellow ATSers can help us out.

I hope someone can explain it a little better. If probability doesn't take into account all variables that would have an effect on the outcome of a result then the experiment is flawed right away. The lack of real probabilistic occurrences might be a big problem for quantum mechanics.

If you knew the force applied, wind, specific gravity, initial position, and the state of all surfaces that the coin could interact with you'd have a near 100% certainty of knowing how the coin would land.

This actually gets to a point I've been trying to make for some time now. Nothing is random.

There are two actions:

1. The initial state of the universe
   and,
   
   
2. Choice

On the macro-scale objects do not deviate. If I throw a ball up in the air I can reliably track the motion both forward and backwards in time. The only thing that causes deviation is choice. I can choose to let the ball fall to the ground or attempt to do something else like swat it away, but without my involvement the state can be perfectly predicted.

Put another way things can be 100% deterministic inside of a specific domain using explicit constraints. For instance your computer works day in and day out because the MOV, ADD, SUB operations on the processor behave exactly the same way, every time. The only time this isn't true is when the processor starts to fail due to age, heat, or other external stimuli like an electrical surge. Thus this caveat represents the constraint.

Likewise in math we can build on past knowledge using corollaries because abstract proofs have been demonstrated to continue to hold in a particular domain (human knowledge) despite being built on core axioms that have no proof and are simply self-evident. No one contests the identity axiom B = B.

As a simple test get 6 people together and ask them, "Does B = B?"

The assertion that it's physically impossible for 100% certainty should be viewed as a mathematical limit. Sometimes we approach so close the two are virtually the same despite their still being an underlying constraint that makes it hold 100% of the time. Removing the constraint it might instead represent 99.9999999999999999...% accuracy rounded up, which still presents enough accuracy to say it's reliably true.

So the explicit constraint that makes a coin flip represent 50/50 odds is based on the notion that it only has two faces to land on irregardless of other information known about the system. As humans become more advanced we solve more of these systems (i.e. checkers is a solvable game).

This implicitly changes the odds because we shift the constraints.

[edit on 11-12-2009 by Xtraeme]

reply to post by sirnex


If you knew every variable there was to know, could you predict the numbers to be generated by a random number generator before they were generated?

This is an interesting question. I can think of two possible solutions:

1. Probabiliy does exist, because nothing is certain in the world.
2 Probably does not exist, when all variables are known

Theoretically if you could know infinite variables at once, you could predict the future with certainty and then there would be no such thing as probability. However, this would work in only a purely deterministic universe. What Quantum Mechanics shows, that the universe is not purely deterministic, that there is a certain element of choice involved as well and those choices cannot be determined. If you try to predict human events for example, you will invariably fail, because you cannot account for the human choice factor or human unpredictability. Similarily, even particles are exhibiting what seems to be choice type behaviour as demonstrated in experiments like the doubt-slit. This suggests that the universe is not entirely deterministic and probability does exist.

Originally posted by CHA0S
reply to post by sirnex

 

If you knew every variable there was to know, could you predict the numbers to be generated by a random number generator before they were generated?

If you're talking about a pseudo-random number generator they can be easily predicted. Real random number generators are usually based on systems that aren't fully calculated and involve non-observation (ie/ balls blowing around in a closed container with random human choice employed to determine when to make a selection).

So the randomness in this case is based on sentient choice.

[edit on 11-12-2009 by Xtraeme]

reply to post by Xtraeme


But haven't they developed a formula which they use in "true" random number generators in which they claim the results are absolutely random?

EDIT: Ummm...doesn't chaos theory state that even when an ordered system becomes so complex, even though the results should be predictable, it breaks into chaos and we get totally random results? With chaos comes order, and with order comes chaos?

[edit on 11/12/09 by CHA0S]

Originally posted by CHA0S
reply to post by Xtraeme

 

But haven't they developed a formula which they use in "true" random number generators in which they claim the results are absolutely random?

Every attempt at a machine-run random generator, whether it be hooked up through a serial port using a dongle or any other instrument has inevitably proven to be horribly weak because it's inevitably based on a symmetric input source. For the random input generator to be strong it has to be based on some asymmetric system and we're the only things that we currently know of in creation capable of this.

Originally posted by CHA0S
reply to post by Xtraeme

 

Ummm...doesn't chaos theory state that even when an ordered system becomes so complex, even though the results should be predictable, it breaks into chaos and we get totally random results? With chaos comes order, and with order comes chaos?

Chaos is based on the notion of bifurcation. The Feigenbaum constant has shown the ratio of convergence at which successive period-doubling bifurcations occurs tends to a constant of around 4.6692. This result was the foundation that allowed mathematicians to start to unravel the "random" behavior of chaotic systems. What was somewhat surprising about this is the constant also had direct applications to fractals. Which gives the hint that basically bifurcation (or chaos) follows an order based on a general overriding algorithm or set of algorithms.

Ok.. I as well as you perhaps do not know much about probability. The only thing I know would be viable for this topic is a more.. "smart $## remark" so I apologize...

Death is ... wait.. that is definite, not probable. Is death probable or does that not count when it comes to this? .. guess it isn't that off-topic as I thought.

reply to post by Xtraeme


ok...so basically a chaotic system isn't as random as we first thought? Hmmm...because I had a theory on why we do in fact have free will, and no matter how much a person knows about us, even if they have all the variables involved, they will not know what we will do next...we will never be 100% predictable...

Something hit me when you said, "there's simply too much to take into account - too much to weigh", I thought about chaos theory, and it's assumption that when a system reaches a certain level of complexity, it breaks into total, probabilistic, uncertain and undeterministic chaos. The results of a system are chaotic, even if we do know all the variables involved in the events, actions or decisions, we simply can't predict the outcome. Now, I postulate, that we do in fact have free-will, and it is because the amount of variables involved in our decisions, make it so complex that it is extremely chaotic, undeterministic, and unpredictable, therefor, there is no causality and no underlying factor that can be used to state we have no free-will. We are simply so complex, our decisions are our own, and no one, no matter how much they know about us, will be able to predict exactly what we will do next.

[edit on 12/12/09 by CHA0S]

reply to post by sirnex


It does not matter if you knew the outcome.

Experimentation with coin tosses will ALWAYS result in 50/50 probability, whether or not you knew the outcomes in advance. If you really could predict the future, you will see that it averages out to 50/50 probability.

If there were some outside forces influencing the coin tosses, then those forces would have to be included in the probability, see?

But, I get what you are saying. But probability is about experimentation, experience, and logic.

A god may see all the outcomes, but it will always follow probability. You have your gaussian distribution, coin tosses, etc.

Probability is one of the inevitable consequences of the mind operating within a conceptual relativistic view of the self and the universe ...

That's partly why it struggles a tidbit the the whole infinity thingie.

Originally posted by Deaf Alien
reply to post by sirnex

 

It does not matter if you knew the outcome.

Experimentation with coin tosses will ALWAYS result in 50/50 probability, whether or not you knew the outcomes in advance. If you really could predict the future, you will see that it averages out to 50/50 probability.

If there were some outside forces influencing the coin tosses, then those forces would have to be included in the probability, see?

Nice and succinct! The explicit constraint is that we have two faces. That forces the scenario of 50:50. Knowing the outcome doesn't make it 100% because the point is there are two possible states. Factoring in advance knowledge of the outcome is, as I said, shifting the constraint.

[edit on 12-12-2009 by Xtraeme]

reply to post by Indigo_Child


I believe the word your looking for is chance. The coin example rate of flip and force you use could never be repeated exactly the same twice. There are incidents in the universe that the odds where against but they happen altering projected outcomes.This is why probability is just the act of guessing most likely outcome.

But the great thing about the universe is yes the unexpected can and will happen.

Probability depends on the method used for experimentation too. I remember an example from a statistics class I had once... (Ugh)

If a frog is surrounded by lilly pads and it makes 4 jumps, which one will it be on. There will be some probability it's back where it started, some percentage chance it's 3 pads away, etc.

Probability isn't a perfect science that can ultimately predict everything.

It's super helpful in counting cards though! Hmm... only five 10-value cards are gone, than means the probability of me drawing a 10 is (some number I don't wanna calculate lol).

also if u study physics you have to use a lot of it in electron location probability n stuff

ok...so basically a chaotic system isn't as random as we first thought? Hmmm...because I had a theory on why we do in fact have free will, and no matter how much a person knows about us, even if they have all the variables involved, they will not know what we will do next...we will never be 100% predictable...

Over the last year I've found myself in many breathless debates trying to explain to friends in academia that science has the bad habit of "dividing and discarding." I usually find myself saying, "The goal of science shouldn't be to ignore data it should be to understand it in its own right." However I admit there's a catch-22 here because ontologically "to name something is to divide it."

For instance, if I go up to a person and ask them, "What does 'a few' mean?" Some will say 3. Others will claim it means several. Another group of people will claim it means no more than a handful. And a few pedant literati will say, "not many but more than one!"

The phrase "a few" has many meanings. No dictionary will ever include all the personal definitions that exist.

Likewise no phenomenon, physical manifestation, no one thing can be fully described perfectly because everything is dynamic. Even in the world of computers where we can have a singular sequence that defines an object the human interpretation of it changes the definition.

For instance if I post a picture to a service like Flickr. Immediately upon submitting the photo several smaller resolution versions of the image will be generated that have a different signature from the original - instantly causing differentiation. Also the meaning of the object can be further refined by adding tags and other meta-data to the picture helping to better qualify what exists abstractly inside the image.

So even though the original picture contains the raw bits it does not possess other information external to it. While I might be able to write a program to identify if the picture has a person in it. The application will have a very hard time identifying "Who?"

My point in all this is to show we would have a scientifically recursive process where any division would then result in the study of some new side-branch. So really until all things could be summed together nothing could be understood perfectly.

Which is very likely why a big crunch would be such an extremely illuminating moment because at that point everything could be understood fully by all things. Which is very much the idea behind physicist Frank Tipler's Omega Point.

So I don't think we'll be computing all chaotic systems any time soon.

[edit on 12-12-2009 by Xtraeme]

reply to post by dragonridr


No, the word I was using was defintiely choice. This is why I mentioned the double slit experiment. In this experiment the particle will collapse into a wave even before it enters the slits, almost as if it "knows" that it must choose that form to pass through. In the single slit experiment, the particle does not collapse into the wave. Thus there is an element of choice being exhibited by the particle.

Originally posted by Deaf Alien
reply to post by sirnex

 

It does not matter if you knew the outcome.

Experimentation with coin tosses will ALWAYS result in 50/50 probability, whether or not you knew the outcomes in advance. If you really could predict the future, you will see that it averages out to 50/50 probability.

If there were some outside forces influencing the coin tosses, then those forces would have to be included in the probability, see?

But, I get what you are saying. But probability is about experimentation, experience, and logic.

A god may see all the outcomes, but it will always follow probability. You have your gaussian distribution, coin tosses, etc.

Well, if this is the case, then quantum mechanics might appear to be mostly wrong to a degree of accuracy. As in the coin toss example, not all variables *are* taken into account to determine the probabilistic outcome of it's two sided landing. It really isn't a fifty/fifty probability as the same force of toss hardly ever occurs a second time, let alone a successive amount of times in a row. That variable ignored messes up the results of probability. At least this is how it appears to me right now.

I don't understand how we can 'experiment' with probabilities when we haven't had any experiments set up that take into account all variables that will or could change the results of it's outcome. I don't know how accurate quantum mechanics is anymore as it seems to rely heavily upon a probabilistic nature of reality as being true. Perhaps the interpretations of the results are wrong because not all variables are experimented with at such a small scale of the universe.