Mechanical systems a route to the poor house.

[tyrant]

The fact is this.....when all the rules are 100% defined, mechanical systems seldom withstand the test of time. Thus, most successful traders are "discretionary" when it comes to the implementation of their mostly mechanical or rule-based trading......BUT, the performance of such traders cannot be objectively verified, and hence, there is always the possibility that even highly successful traders with years of track record are a result of RANDOMNESS...

So, it does not matter whether you are mechanical or discretionary....you still need TIME to prove that you are successful going forward...the longer the time, the less you are a result of RANDOMNESS....

such is the paradox of trading...

 

[toc]

'I get the signals, which are mechanically generated, then make a discretionary decision as to whether to actually take the trade.'

Can work if discretionary decisions are based on another set of mechanical type rules.

 

[FanOfFridays]

Quote from me:I get the signals, which are mechanically generated, then make a discretionary decision as to whether to actually take the trade.

Quote from RoughTrader:
A very good example of something NOT to do.
RoughTrader

IMO, this is a very good example of the kind of thinking traders must avoid at all costs. There are as many ways to trade successfully as there traders. Never let anyone tell you that something you want to try is 'wrong' or 'stupid'. It would be pointless for me to start to list all the great traders who have traded exactly this way - the list is way too long. It is also pointless to name all the traders who feel that any discretionary component in the decision making process is a mistake.

The hallmark of a good trader is adaptability and an ability to admit that one is wrong. Any rigidity will soon be dealt with, harshly, by the markets.

Take an idea, try it with sensible risk management rules in place, and if it works for you, trade it until it no longer works. Leave the judgement of other people's trading to the paper traders.

 

[RoughTrader]

Quote from traderNik:

IMO, this is a very good example of the kind of thinking traders must avoid at all costs. There are as many ways to trade successfully as there traders. Never let anyone tell you that something you want to try is 'wrong' or 'stupid'. It would be pointless for me to start to list all the great traders who have traded exactly this way - the list is way too long. It is also pointless to name all the traders who feel that any discretionary component in the decision making process is a mistake.

The hallmark of a good trader is adaptability and an ability to admit that one is wrong. Any rigidity will soon be dealt with, harshly, by the markets.

Take an idea, try it with sensible risk management rules in place, and if it works for you, trade it until it no longer works. Leave the judgement of other people's trading to the paper traders.

I have seen traders take a perfectly profitable system and turn the real results into a long-term loser by trying to outsmart it like the way you are describing. If you are still profitable by applying your "opinion-based" filter, more power to you. I would never be able to pull money out of the markets by doing this.

RoughTrader

 

[FanOfFridays]

Quote from RoughTrader:

I have seen traders take a perfectly profitable system and turn the real results into a long-term loser by trying to outsmart it like the way you are describing. If you are still profitable by applying your "opinion-based" filter, more power to you. I would never be able to pull money out of the markets by doing this.

RoughTrader

Fine. I thought it strange that you said watching the markets and making the decision to enter them is a good example of something not to do, since traders have been doing this for what, 1000 years? (When were those first open outcry rice markets anyway?).

The fact that I run variables through an algorithm on a computer, take the output and make a decision is no different than a tape reader reading the tape (taking variables) analyzing them using his brain (running them through a computer) and making a decision to enter or exit; how many traders have traded this way? I will say it again - I do not trade intraday.

If you need to eliminate the discretionary component in your trading to optimize your results, fine. If you could never pull money out of the markets my way, better for me. Just don't claim that my way is an example of what 'not to do'. It's a) misleading for beginners reading this and b) dead wrong.

 

[RoughTrader]

Quote from traderNik:

Fine. I thought it strange that you said watching the markets and making the decision to enter them is a good example of something not to do, since traders have been doing this for what, 1000 years? (When were those first open outcry rice markets anyway?).

The fact that I run variables through an algorithm on a computer, take the output and make a decision is no different than a tape reader reading the tape (taking variables) analyzing them using his brain (running them through a computer) and making a decision to enter or exit; how many traders have traded this way? I will say it again - I do not trade intraday.

If you need to eliminate the discretionary component in your trading to optimize your results, fine. If you could never pull money out of the markets my way, better for me. Just don't claim that my way is an example of what not to do. It's a) misleading for beginners reading this and b) dead wrong.

Wow, someone needs their morning coffee. I think I misunderstood your approach. I thought you were advocating the use of opinion to override a numerically computed entry signal of a system. It seems you are talking about something else entirely. You are using a rule-based approach to determine whether to take the trade or not, if I understood correctly, rather than basing the decision on whim.

If my understanding is correct, you are actually still applying a mechanical approach to entry, and use a defined method to filter the trades. There have been traders on this board who have posted about not having the discipline to follow the orders of their system in the strictest sense, and have suffered the consequences as a result. My point was only to emphasize that deviating from one's plan is something that should never be done.

Peace.

RoughTrader

 

[FanOfFridays]

Quote from RoughTrader:

Wow, someone needs their morning coffee. I think I misunderstood your approach. I thought you were advocating the use of opinion to override a numerically computed entry signal of a system. It seems you are talking about something else entirely. You are using a rule-based approach to determine whether to take the trade or not, if I understood correctly, rather than basing the decision on whim.

If my understanding is correct, you are actually still applying a mechanical approach to entry, and use a defined method to filter the trades. There have been traders on this board who have posted about not having the discipline to follow the orders of their system in the strictest sense, and have suffered the consequences as a result. My point was only to emphasize that deviating from one's plan is something that should never be done.

Peace.

RoughTrader

I see - no, I was not advocating taking a fixed system that you have backtested successfully and then trying to deploy it by taking its outputs and making a snap decision as to whether to take a particular trade - nothing of the kind. This would of course ruin the system and most likely, the one trade that you decided to pass would be the exact one that would run big and make up for the losing trades generated by the system. Of course this is a primary reason why purchased systems fail so often - the buyer does not understand the logic of the system. If you write your own systems you understand why they trigger and can take every signal with confidence.

I simply take variables, process them and then come up with a universe of possible trades, and choose the ones which I want to enter.

I think we understand each other. Have a good one.

 

[tyrant]

Quote from traderNik:

I see - no, I was not advocating taking a fixed system that you have backtested successfully and then trying to deploy it by taking its outputs and making a snap decision as to whether to take a particular trade - nothing of the kind. This would of course ruin the system and most likely, the one trade that you decided to pass would be the exact one that would run big and make up for the losing trades generated by the system. Of course this is a primary reason why purchased systems fail so often - the buyer does not understand the logic of the system. If you write your own systems you understand why they trigger and can take every signal with confidence.

I simply take variables, process them and then come up with a universe of possible trades, and choose the ones which I want to enter.

I think we understand each other. Have a good one.

Is your selection out of the universe of possible trades 100% rule based? Why?

 

[RoughTrader]

Quote from tyrant:

Is your selection out of the universe of possible trades 100% rule based? Why?

Yes. It allows me to remove any thought or analysis from the decisions to buy or sell. That is, by definition, stress-free trading for me.

RoughTrader

 

[FanOfFridays]

Quote from tyrant:

Is your selection out of the universe of possible trades 100% rule based? Why?

If you were asking me, let me first quote something you wrote earlier in this thread.

"Thus, most successful traders are "discretionary" when it comes to the implementation of their mostly mechanical or rule-based trading......BUT, the performance of such traders cannot be objectively verified, and hence, there is always the possibility that even highly successful traders with years of track record are a result of RANDOMNESS..."

I am not sure how to understand the phrase '100% rule-based'. I suppose it is... decisions are made according to certain rules I have about my trading which pertain to diversification and risk management, rules which are not part of the computer code through which I run the variables I am interested in. I don't want my trades to be too highly correlated and I want to make sure my risk management rules are being followed.

I am not sure why you say that the performance traders who use a discretionary implementation of their 'mostly mechanical or rule-based' strategies cannot be verified. Why can't it? If a trader shows positive returns for a period of 15 years, what standard deviation is that at? It is highly academic to say that you could never show that their performance wasn't random. At least it seems that way to me. Maybe I'm not understanding what you are saying here.