The Flaw in Trading System Reasoning...?

If a tree falls in a forest and no one is around to hear it, and if it causes a butterfly to flap its wings, will there be a tornado in Brazil?

Or what if a tree falls in a forest and there is a machine recording air vibration but no-one ever see the results?

Seriously I never understood the point of this. Firstly as long as there's not a vaccuum there will be sound. Secondly there is always an answer, whether or not anyone knows about it.

With the rubics cube thingy, if you're looking for an example to express how many ways there are to trade the markets then a more appropriate example would be to imagine Rubic sitting at a desk with a blank piece of paper and a desire to create something.
 
Quote from Remiraz:

If a tree falls in a forest and no one is around to hear it, does it make a sound?



Okay, for example :

You restrict your backtest data to only 10 days and only the opening and closing price of these 10 days.

Lets say in these 10 days, the open is always lower than the close (uptrend).

The number of systems you can get to show profit within these 10 days would be finite.



A Rubik's Cube can have (8! × 38−1) × (12! × 212−1)/2 = 43,252,003,274,489,856,000 different positions (~4.3 × 1019), about 43 quintillion.

A huge number, but still a finite number.
More...

First, regarding that finite number...what is it? I substantiated my point about the infinite number using asymtotic limits as they are applied in calculus. Maybe you haven't had calculus so that didn't really click with you. But, thats one reason why the calculus evolved, was to explain and work with concepts like "while the number of variables between any two systems increases arithmatically (that is from 1 to 2, from 2 to 3), the number of different systems created from that arithmatic increase increases geometrically (from 1 squared to 2 squared, 2 squared to 3 squared). This it becomes infinite at some finite variable input.

Okay...so you don't understand the math.

Lets just say that there are more variations in between the numbers 1 and 2 than there are on the rubics cube. In fact, an infinite number. (1.001, 1.00002, 1.00000005, etc). So a rubics cube is not a true example of the infinite. Even still..AS YOU ADMIT...the number of combinations is quite high for a Rubic's cube....CERTAINLY INFINITE COMPARED TO THE 5000 systems you used for your hypothetical example of profitable systems.

And, most basically, yes. If you say there is a finite number, you need to be able to either name that number, or at least, give the order of magnitude of that number. That would be an example of a rigorous proof. That proof you have not provided.




How many ways are there to reach a P/L of $100.00 on any given day? Each way represents a different system. Maybe to you all systems that go long are the same type of system. So thats why you appear stuck in the finite.

Fundamentally, everyone who trades follows a DIFFERENT SYSTEM. Every trader who makes a profit in a given timeframe has been profitable at slightly different entry, exit and slippage points even if they are trying to follow EXACTLY THE SAME RULES>

Have you ever traded in a trading room with others? Ever been with guys who are supposedly trading the same system? Guess what, each guy trades it slightly differently. IF both guys are profitable, in reality, they trade different systems.

I think our definition of what a system is is just plain different. Yes, I think mine is better and more indicative of the kind of mentality one needs in order to be a top level trader. But if your definition is more narrow, and it works for you, and you are successful, then probably the essential idea behind this talk is the same.

The idea that infinite numbers derive from using a system of infinite numbers (decimalization, for example) is the only point needed to show, mathematically, the truth of my belief. The belief that there indeed are an infinite number of profitable systems. Just because so many are not profitable has nothing to do with that fact. All the observations in the market behavior of many traders are incidentals that support that core idea.

The bottom line is, if your belief is functional (if it works for you), then its right (for you). Since my belief works for me, then its right for me. I spent time writing these ideas because they represent another core belief: the idea of abundance vs. scarcity. Since mathematically, abundance is supported given the variables inherent in trading, I thinks its a more flexible and powerful belief to hold. I hope that my ideas reach someone who can use them to their benefit and improvement.

I used to belief what you do, and the changes in my beliefs seemed to coincide with my explosion in my success as a trader.
But that doesn't mean it was the sole reason for my success.

Thanks for the debate.
 
Maybe try thinking about it this way:

First, define what it means for two systems to be equivalent:
Two systems are said to be equivalent over a specified period of time if they generate exactly the same signals over that period of time. (An example of two equivalent systems would be (a) to buy/sell if the 20 SMA of closes is increasing/decreasing and (b) to buy/sell if the close today is higher/lower than the close 21 days ago. Systems (a) and (b) are equivalent over all time.)

Then, think about how many different possible signals there are:
For every tick in the day, a system can either execute a buy of N contracts, execute a sell of N contracts, or do nothing.

This would allow only a finite (though very large) set of possible signals, and therefore a finite set of equivalent systems over any finite period of time. If you believe that humans will be trading for an infinite period of time, then there are an infinite number of trading systems. If not, then there are only a finite number of equivalent trading systems.

Note that no matter how large this (or any) finite number may be, it is still not infinite. There is no such thing as "infinite compared to a smaller number". There is only infinite or finite.

Just a thought...

M
 
Quote from Mathemagician:

Maybe try thinking about it this way:

First, define what it means for two systems to be equivalent:
Two systems are said to be equivalent over a specified period of time if they generate exactly the same signals over that period of time. (An example of two equivalent systems would be (a) to buy/sell if the 20 SMA of closes is increasing/decreasing and (b) to buy/sell if the close today is higher/lower than the close 21 days ago. Systems (a) and (b) are equivalent over all time.)

Then, think about how many different possible signals there are:
For every tick in the day, a system can either execute a buy of N contracts, execute a sell of N contracts, or do nothing.

This would allow only a finite (though very large) set of possible signals, and therefore a finite set of equivalent systems over any finite period of time. If you believe that humans will be trading for an infinite period of time, then there are an infinite number of trading systems. If not, then there are only a finite number of equivalent trading systems.

Note that no matter how large this (or any) finite number may be, it is still not infinite. There is no such thing as "infinite compared to a smaller number". There is only infinite or finite.

Just a thought...

M
More...

You're wrong, because your definition of 'equivalence' isn't good. You might very well find 2 systems that give the same signals on a given time period and then different signals on another time period. Your conclusion is right, but your premices are wrong.
 
I did not define "equivalent". I defined "equivalent over a specified period of time". We can take "equivalent" to mean "equivalent for all time". If the total period of time that a trading system might operate is finite (as mentioned above), then my premises and conclusion are valid and there are only finitely many trading systems. If our time is infinite, then the conclusion is that there are infinitely many trading systems.

Perhaps you've got a better definition and can walk us through the logic that supports your conclusion?

M
 
Quote from Mathemagician:

I did not define "equivalent". I defined "equivalent over a specified period of time".
More...
I have to agree with science_trader that your definition is not correct. I would agree that the OUTPUT of the two systems are equivalent over a specified period of time but to to say that any two systems that generate the same signals are equivalent strikes me as an incorrect definition of equivalence.

As to whether there are an infinite or very large but finite number of profitable trading systems, seems to me that is distinction without a difference, e.g. you could assume either answer to address the OP original point without affecting the discussion.
 
Quote from winter:

I have to agree with science_trader that your definition is not correct. I would agree that the OUTPUT of the two systems are equivalent over a specified period of time but to to say that any two systems that generate the same signals are equivalent strikes me as an incorrect definition of equivalence.
More...
If two algorithms generate the same output when given the same input, how else could one distinguish them?

Perhaps you could improve upon the definition?

M
 
I'm not really interested in this debate, but here is a short thought.

You have to make a difference between the 'considered whole time period' and the period on which two systems are equivalent.
 
Quote from science_trader:

I'm not really interested in this debate, but here is a short thought.

You have to make a difference between the 'considered whole time period' and the period on which two systems are equivalent.
More...
Um, I did.
 
You must log in or register to reply here.
Back
Top