Random trading systems - "monkey with a pin"

[ryandj2222]

Arguably, tennis does have a to b type rules. It isn't random.

Given the wind direction, speed of the ball, weight of the ball, good knowledge of physics you can make a pretty good prediction of where the ball will end up if you hit it in a certain angle at a certain speed.

Of course actually being able to hit the ball in the certain angle and speed requires a lot of skill. And measuring precisely the various speeds, directions etc is difficult, but I don't see tennis as being random.

Plus swinging the racket randomly around the court isn't going to see you doing better than, say Tim Henman, or an 8-year-old who has had a couple of lessons (I can personally vouch for the second point).

 

[new_trader]

I used to argue with mr Socco that there is more than one way of skinning a cat. It makes no difference to me if people see the benefits of random entry or not, I just point out enough info to maybe persuade people that it's an approach with merit.

I would have no problem trading off horoscopes or even cloud patterns (a few of the random generators out there are based on atmospheric noise). It's as nonsensical as technical analysis or price action, and serves the same purpose in most cases.

Some people might believe that your success was a function of your ability to analyse cloud patterns others may be more enlightened. 20 years ago I probably believed successful traders using TA where successful due to their knowledge of TA but now I know better.

Markets are not simple mechanical mechanisms that can be encapsulated into simple if a then b type rules, and attempting to do so, is pointless, but of course its relatively easy to do so, hence the popularity of TA based systems

I think there's merit in TA, and I know people who appear to trade it successfully, but I equally well think they could trade the inverse of their TA signals, because trading really has jack to do with trade entries.

This is a straw man argument.

 

[kimo'sabby]

Arguably, tennis does have a to b type rules. It isn't random.

Why do people lose?

 

[ryandj2222]

Arguably, tennis does have a to b type rules. It isn't random.

Why do people lose?

Because they didn't play as well as the other player... skill obviously.

But as I said, tennis is an example of where a honed skill would work better than random whacking of the racket around.

Unlike trading where random strategies - as shown in the papers I linked to earlier - can give you better returns than some highly skilled professionals achieve using their predictive skills.

 

[the hare]

Hahaha! Durrrr! Tennis??

Think about air to surface missiles. Most systems these days try to include a random component that makes tracking and interception more difficult. Consequentially, interception systems need to be a bit more random.

If I where designing a robotic tennis player I would certainly want to introduce a random component into offensive playing at least.

Any problem where your opponent is applying an adaptive strategy would probably benefit from random component (identifying potential terrorists at airports in random stop and search being a good example)

As for the rat in a t shaped maze, I don't know the source, of much about how the background details. As I say, its been posted about 100 thousand times here by a notorious vendor ! But if its only partially true, it does show how humans always try to complicate matters and how they quickly for associations that in many cases do not exist.

 

[the hare]

This is a straw man argument.

Which bit ?

 

[kimo'sabby]

Unlike trading where random strategies - as shown in the papers I linked to earlier


Papers?? You mean, bucketshop propoganda?

 

[new_trader]

Which bit ?

Markets are not simple mechanical mechanisms that can be encapsulated into simple if a then b type rules, and attempting to do so, is pointless, but of course its relatively easy to do so, hence the popularity of TA based systems

I think there's merit in TA, and I know people who appear to trade it successfully, but I equally well think they could trade the inverse of their TA signals, because trading really has jack to do with trade entries.

Who says that markets are simple mechanical mechanisms that can be encapsulated into simple if a then b type rules?

 

[new_trader]

Unlike trading where random strategies - as shown in the papers I linked to earlier


Papers?? You mean, bucketshop propoganda?

Fix your quoting...

 

[ryandj2222]

Unlike trading where random strategies - as shown in the papers I linked to earlier


Papers?? You mean, bucketshop propoganda?

No.

 

[kimo'sabby]

But if its only partially true, it does show how humans always try to complicate matters and how they quickly for associations that in many cases do not exist.

Gambling, i'll think you'll find is the word that you are looking for.

It exerts an enormous amount of pressure on the brain.

Couple this with inexperience, or inadequacy, and the markets become a truly random and humbling experience.

 

[the hare]

Who says that markets are simple mechanical mechanisms that can be encapsulated into simple if a then b type rules?

I don't think there's an official registrar of names that you can consult, but there's no shortages of books, websites and system vendors who appear to either believe that's the case (because they actively engage in designing and testing simplistic systems based on that premise), or proactively market systems based on simple rules (only 89 copies left etc)

If you gave me 10-15 seconds I could probably design you a simple mechanical system that gave great historical performance, which might even have made money if traded.

There's enough people trying to find a solution on the assumption that markets are simple and predictable, and enough vendors peddling products on that basis that I think I'm justified in assuming that there are a significant number of people who hold this belief.

Perhaps I should start a poll ? (Previous polls I've seen on other sides almost unanimously show the 'traders' answering those polls believe that markets are predictable using very basic analysis techniques)

FFS just look what happens when anyone dares to suggest that you can make money using non predictive methods, all hell breaks lose, that should tell you something.

 

[kimo'sabby]

FFS just look what happens when anyone dares to suggest that you can make money using non predictive methods, all hell breaks lose, that should tell you something.

Nobody (unless they are mad) are saying that the markets are 100% predictive. So why are you lashing out in all directions like a child throwing a tantrum?

 

[ryandj2222]

Nobody (unless they are mad) are saying that the markets are 100% predictive. So why are you lashing out in all directions like a child throwing a tantrum?

If they aren't 100% predictable then that implies there is some degree of randomness involved.

Say, for argument's sake that your trading methods give you a 50% hit rate. Does that mean that the other half the time the markets are behaving randomly?

 

[kimo'sabby]

If they aren't 100% predictable then that implies there is some degree of randomness involved.


Are you ill? We've already established, and agreed on this. Christ almighty!

 

[ryandj2222]

If they aren't 100% predictable then that implies there is some degree of randomness involved.


Are you ill? We've already established, and agreed on this. Christ almighty!

I am fine thanks. I know there is some randomness involved. I am just trying to understand the idea of being both predictable and random...:-0

 

[ryandj2222]

http://www.trade2win.com/boards/trading-journals/53266-random-trading-anyone-interested.html

 

[kimo'sabby]

I am fine thanks. I know there is some randomness involved. I am just trying to understand the idea of being both predictable and random...:-0

What have you learnt so far? (don't be shy)

 

[ryandj2222]

What have you learnt so far? (don't be shy)

Mainly that I don't have a clue what's going to happen next! :cheesy:
But we can have a go at predicting. Sometimes with good results sometimes with wrong ones.

 

[new_trader]

I don't think there's an official registrar of names that you can consult, but there's no shortages of books, websites and system vendors who appear to either believe that's the case (because they actively engage in designing and testing simplistic systems based on that premise), or proactively market systems based on simple rules (only 89 copies left etc).

Yes, sorry. I was narrowly focused on this thread only.