S/R and TL on Randomized Price Movement | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Price movement as a function of Supply and Demand 43 replies

Spot Fx Trades based on futures volume and price movement 22 replies

How to differentiate pro and amateur price movement? 6 replies

Random Price Movement 177 replies

GBP/USD daily pip movement vs. EUR/USD daily pip movement 3 replies

Attachments: S/R and TL on Randomized Price Movement
Exit Attachments

S/R and TL on Randomized Price Movement

  • #61
  • Mar 26, 2010 2:39pm Mar 26, 2010 2:39pm
Quoting Bytebodger
Some people have posted in this thread that you can't profit from a random system. If this were the case, there would be no lottery. Random systems make people money - sometimes, boat loads of money - every single day. The key is that there is a huge difference between saying that "this system is profitable" rather than saying "I am profitable".
is the lottery really random.. you are limited between certain balls.. and the weight of the balls im sure matter and the smoothness of the exterior. and the speed of the wind that hits the ball and the trajectory that the ball enters the holding unit.. also the speed of the circulator in the ball chamber.. shake the container and wouldnt the heavier balls move to the bottom ? wouldnt the lightest balls be shot first.. so wouldnt over X amount of chosen balls the lightest balls be more repeatative? wouldnt the ones with bends in the ball catch more air and wouldnt the smoother ones just spin?
  • #62
  • Mar 26, 2010 2:54pm Mar 26, 2010 2:54pm
Quoting domino
is the lottery really random.. you are limited between certain balls.. and the weight of the balls im sure matter and the smoothness of the exterior. and the speed of the wind that hits the ball and the trajectory that the ball enters the holding unit.. also the speed of the circulator in the ball chamber.. shake the container and wouldnt the heavier balls move to the bottom ? wouldnt the lightest balls be shot first.. so wouldnt over X amount of chosen balls the lightest balls be more repeatative? wouldnt the ones with bends in the ball catch more...
Yes, you are correct. Referring back to my second post in this thread, I've said that randomness is relative. In other words, to the participants in the lottery the results ARE random. They have no way to determine/calculate any of the millions of tiny factors that go into the actual selection of the lottery numbers. From mankind's perspective, it is a random game.

On a larger philosophical scale, I am not sure that anything is truly random. Again, referring to one of my earlier posts, "random" is just another way of saying "I don't have the ability to compute all the factors involved"
  • #63
  • Mar 26, 2010 3:00pm Mar 26, 2010 3:00pm
Quoting Bytebodger
On a larger philosophical scale, I am not sure that anything is truly random. Again, referring to one of my earlier posts, "random" is just another way of saying "I don't have the ability to compute all the factors involved"
i agree but your statement i dont have the ability to compute all the factors involved.. what about in trading .. we dont does that make it fall under random?
  • #64
  • Mar 26, 2010 3:01pm Mar 26, 2010 3:01pm
It's also important to note that randomness is usually expressed in probabilities, and probabilities infer an event that has not yet taken place. So even though I say that something like a coin flip is not truly random, what I really mean is that it is not truly random at the moment it takes place. Once the coin flip is initiated, in theory, all of the factors affecting the result could be calculated and used to predict the outcome.

But if we assume that the coin flip occurs in the future (as opposed to right now) and that certain conditions of that coin flip are unknowable at this point in time (like the exact humidity, or the velocity of any surrounding air currents), then the future event is "random" and can only be expressed in terms of probabilities.

Of course, if you could know with absolute certainty all of the factors that would go into a future event, then you could theoretically predict that event with absolute certainty.
  • #65
  • Mar 26, 2010 3:05pm Mar 26, 2010 3:05pm
Quoting Bytebodger
It's also important to note that randomness is usually expressed in probabilities, and probabilities infer an event that has not yet taken place. So even though I say that something like a coin flip is not truly random, what I really mean is that it is not truly random at the moment it takes place. Once the coin flip is initiated, in theory, all of the factors affecting the result could be calculated and used to predict the outcome.

But if we assume that the coin flip occurs in the future (as opposed to right now) and that certain conditions...
this will probably keep me awake tonight.
  • #66
  • Mar 26, 2010 3:12pm Mar 26, 2010 3:12pm
Quoting domino
i agree but your statement i dont have the ability to compute all the factors involved.. what about in trading .. we dont does that make it fall under random?
First, if you do not know or understand ANY of the factors that go into the price action, then yes, from your perspective, trading would be random. This does not mean that the markets are random. It just means that the markets are random by your frame of reference. This is why I said in a previous post that randomness is relative.

Second, if we assume that you're not an idiot and that you can at least learn even the most basic principles of the market, then this knowledge tends to diminish the randomness of the markets.

As referenced in my previous post, this is not a Boolean equation. It's not like we can say, "Trader X knows nothing and so the markets are random for him but Trader Y knows much, and thus the markets are nonrandom for him."

Let's assume that you know Everything There Is To Know About The Markets. Does this mean that there is no randomness in your perception of the market? No. It means that you understand the probabilities involved in the market. But even if you absolutely know all of the true probabilities associated with the market, this does not mean that there is no randomness involved in the market.

For example, having "perfect knowledge" of ALL the factors influencing GBPUSD would not tell you whether GBPUSD will rise or fall. Having perfect knowledge would tell you the probability that GBPUSD would rise or fall.

In other words, if you had perfect knowledge of all the factors involved in GBPUSD, you might realize that over the next 24 hours, there is a 50% chance that GBPUSD will rise, and a 50% chance that it will fall.

So if the "true" probability of GBPUSD rising/falling is 50%/50%, then what is it that actually determines whether it rises or falls? The answer: randomness.

Of course, you wouldn't normally want to make a trade when you think the odds are 50/50. So you try to buy when you think the odds are greater than 50% that the pair will rise and you try to sell when you think the odds are greater than 50% that the pair will fall.
  • #67
  • Mar 26, 2010 6:15pm Mar 26, 2010 6:15pm
Quoting mikkom
Here is one relevant post I posted some time ago,
You are right, but this particular example is flawed.Think about it for a while: in a slowly but constantly downtrending or uptrending but otherwise completely random market, at which hours would most likely be the daily high or low?

A better statistic that would not suffer from such edge effects would be the high of the surrounding 24 hours (+-12h) for each hour.
  • #68
  • Mar 26, 2010 6:48pm Mar 26, 2010 6:48pm
Let's dispense with the word 'randomness', since it seems to mean different things to different people (although Bytebodger's thesis is pretty much correct, IMO).

Technical analysis is the study of price. To the purely technical trader, price is effectively nothing more than a sequence of numbers; how the numbers were generated is irrelevant.

Consequently, the question, as I see it, is: Are there any discernible patterns or phenomena existing in these numbers (the 'cause'), that on balance of probability, create some kind of bias in the way in which the next part of the sequence (the 'effect') will be generated?

If so, then TA is potentially capable of providing an 'edge'. If not, then our price charts are worthless, hence we might as well dispense with them, and look elsewhere (fundamentals, orderflow, astrology,…??).

An imaginary chart, created using a random number generator, will exhibit trends and patterns as a matter of course, but because the next part of the sequence is always generated in such a way that there is no causal link to the immediately prior numbers, there is no subsequent bias that can be exploited.

I agree that it's possible to profit from a random event, but that it is fortuitous. We can have a 'lucky' day at the casino, for example. But the casino has an edge, and the law of large numbers guarantees that its edge will eventually prevail. Similarly, a trader must somehow find a real, enduring edge that's large enough to overcome the negative expectancy imposed by the spread. If that is not possible, then the spread will ensure that every trader will eventually lose. 'Luck' is not sustainable indefinitely.
  • #69
  • Mar 26, 2010 10:02pm Mar 26, 2010 10:02pm
Quoting Bytebodger
First, if you do not know or understand ANY of the factors that go into the price action, then yes, from your perspective, trading would be random. This does not mean that the markets are random. It just means that the markets are random by your frame of reference. This is why I said in a previous post that randomness is relative.

Second, if we assume that you're not an idiot and that you can at least learn even the most basic principles of the market, then this knowledge tends to diminish the randomness of the markets.
dunno if I should be offended by this or not seems directed im just pondering randomness and rules of random chart making.
  • #70
  • Mar 26, 2010 10:16pm Mar 26, 2010 10:16pm
Quoting domino
dunno if I should be offended by this or not seems directed im just pondering randomness and rules of random chart making.
No offense intended. I'm not implying that YOU are an idiot, I was just making the argument regarding you (the generic "you", as in, a nameless person playing a role) for the sake of illustrating a point. Sorry for any confusion.
  • #71
  • Mar 26, 2010 10:41pm Mar 26, 2010 10:41pm
Quoting hanover
If so, then TA is potentially capable of providing an 'edge'. If not, then our price charts are worthless, hence we might as well dispense with them, and look elsewhere (fundamentals, orderflow, astrology,…??).

I agree that it's possible to profit from a random event, but that it is fortuitous. We can have a 'lucky' day at the casino, for example. But the casino has an edge, and the law of large numbers guarantees that its edge will eventually prevail. Similarly, a trader must somehow find a real, enduring edge that's large enough to overcome...
Yes, absolutely agreed. Despite my ever-so-deep-and-insightful ramblings about probabilities and randomness, the central question is still quite simple:

Does any aspect of previous price action have an effect - ANY effect - on future price action???

If previous price action has absolutely positively no effect on future price action, then almost all technical analysis is completely worthless. All of our indicators and charts are a monumental waste of time.

Although I feel entirely comfortable speaking in the general realm of randomness and probabilities, I must admit that I am not equipped to answer (prove) the deeper question about the overall validity of technical analysis.
  • #72
  • Edited 1:36am Mar 27, 2010 1:22am | Edited 1:36am
Quoting Bytebodger
Does any aspect of previous price action have an effect - ANY effect - on future price action???
A VERY BIG YES.
In other words, you ask if there's some function which takes as arguments some old prices and give you a better than random entry. It is not a prediction I'm referring to but an increase of the probability of winning. The old prices put the odds in your favor on the long run even the next outcome is meaningless.

One example is DIBS, the other Donchian Channels. They worked for all the pairs in the last 10 years, if you wait long enough to see the profits and you are satisfied with small rates of return. An EA for them is trivial and the concepts behind them too. Accordingly, they fail on the random charts.
  • #73
  • Mar 27, 2010 5:44am Mar 27, 2010 5:44am
Quoting Sauron
A VERY BIG YES.
In other words, you ask if there's some function which takes as arguments some old prices and give you a better than random entry. It is not a prediction I'm referring to but an increase of the probability of winning. The old prices put the odds in your favor on the long run even the next outcome is meaningless.

One example is DIBS, the other Donchian Channels. They worked for all the pairs in the last 10 years, if you wait long enough to see the profits and you are satisfied with small rates of return. An EA for them...
How do you know this for sure? Are you sure the result was not noise? Any evident easy to present here? Have you tested 1000x1000 at least?
  • #74
  • Mar 27, 2010 6:14am Mar 27, 2010 6:14am
Quoting kk007
How do you know this for sure? Are you sure the result was not noise? Any evident easy to present here? Have you tested 1000x1000 at least?
LOL
Well, my answer is actually my last phrase, which you forgot to quote.
  • #75
  • Mar 27, 2010 6:16am Mar 27, 2010 6:16am
Quoting Sauron
LOL
Well, my answer is actually my last phrase, which you forgot to quote.
LOL. Well, you mean this "They worked for all the pairs in the last 10 years"? It is a claim without support.
  • #76
  • Mar 27, 2010 6:27am Mar 27, 2010 6:27am
I meant the last sentence.
Is it clear now?
  • #77
  • Mar 27, 2010 7:42am Mar 27, 2010 7:42am
Quoting Sauron
I meant the last sentence.
Is it clear now?
Your last sentence on the post is this "Accordingly, they fail on the random charts.". Or, are you talking about your signature?
  • #78
  • Mar 27, 2010 11:54am Mar 27, 2010 11:54am
Quoting Sauron
A VERY BIG YES.
In other words, you ask if there's some function which takes as arguments some old prices and give you a better than random entry. It is not a prediction I'm referring to but an increase of the probability of winning. The old prices put the odds in your favor on the long run even the next outcome is meaningless.

One example is DIBS, the other Donchian Channels. They worked for all the pairs in the last 10 years, if you wait long enough to see the profits and you are satisfied with small rates of return. An EA for them...
I'd like to test DIBS and Donchian channels on random charts to see if this is true. If so, that's pretty powerful since I've had really good success with DIBS and know from other traders that DIBS and Donchian have worked in many markets for as long as they've been tried.
  • #79
  • Edited 7:09pm Mar 30, 2010 6:53pm | Edited 7:09pm
Quoting kk007
Any Ideas?
Had some fun:

Attached Image (click to enlarge)
Click to Enlarge Name: RandSim05.jpg Size: 41 KB
  • #80
  • Mar 30, 2010 9:05pm Mar 30, 2010 9:05pm
There is absolutely nothing random about FOREX. An experienced eye can see uniformity whereas an inexperienced eye will see randomness. If a new trader looks at a FOREX chart for the first time....everything will look completely random. Basically, the more inexperienced you are as a trader, the more random the chart will appear to you.
1 trader viewing now