Are Discretionary strategies any better than Mechanical systems ?

victor123

Junior member
Messages
34
Likes
0
Just like mechanical systems work for a limited period, is the same not true of discretionary strategies using price action and chart set-ups ? What I mean is that could the chart setups that are profitable today become unprofitable in a few years from now ? And then the trader needs to start looking for new set-ups that work. Or, am I thinking in the wrong way ?

:?:
 
I think discretionary strategies generally outperform mechanical systems. Systems just allow you to cover more markets more easilly. Discretionary traders tend to use a mix of technical and overall fundamental bias in their position taking.
Overall who really cares so long as you can make a good return. I have found following a mechanical system easier and I do not believe mechanical systems work for a limited period if they are well developed anymore than discretionary strategies.
 
I think discretionary strategies generally outperform mechanical systems. Systems just allow you to cover more markets more easilly. Discretionary traders tend to use a mix of technical and overall fundamental bias in their position taking.
Overall who really cares so long as you can make a good return. I have found following a mechanical system easier and I do not believe mechanical systems work for a limited period if they are well developed anymore than discretionary strategies.

Hi TWI,

Can I ask what markets you currently trade using your Mechanical System ??? Just interested, as I am currently developing one myself to trade Stocks on the FTSE250.


Kind Regards,

Chorlton
 
A purely mechanical system will kill you during a black swan. Imagine trading interest rate swaps on Black Wednesday!
 
Just like mechanical systems work for a limited period, is the same not true of discretionary strategies using price action and chart set-ups ? What I mean is that could the chart setups that are profitable today become unprofitable in a few years from now ? And then the trader needs to start looking for new set-ups that work. Or, am I thinking in the wrong way ?

:?:


Trading is a skill, and like any skill it requires hard work, study and practice in order to attain proficiency. You are unlikely to become a proficient trader using a mechanical system but it’s the easiest route to take and that’s why most aspirants go down that path. A discretionary trader is much more likely to develop a skill that ensures consistent profits and a permanent edge.
 
You are unlikely to become a proficient trader using a mechanical system

A discretionary trader is much more likely to develop a skill that ensures consistent profits and a permanent edge.

Can I ask why this would be?

Surely, if one can find an inefficiency in the markets and carry out sufficient backtesting to ensure that it can be potentially be exploited then surely a mechanical system would be a better approach, as it would ensure that those rules are consistantly followed??

I would agree that a discretionary Trader would probably develop more in-depth skills and probably have a better "feel" for the markets but isn't there also the risk that other factors such as emotions can have an impact on his/her's trading??

Just interested....
 
I don't think that this question can really be answered because there is no way to conduct an objective analysis. I do get the impression though that discretionary traders are more successful than those using a purely mechanical approach.


Paul
 
Can I ask why this would be?

Surely, if one can find an inefficiency in the markets and carry out sufficient backtesting to ensure that it can be potentially be exploited then surely a mechanical system would be a better approach, as it would ensure that those rules are consistantly followed??

Most people don't carry out sufficient back testing, and even if they do, they will find excuses as to why the system failed to produce acceptable results during certain periods. You are right, in the VERY long term even a simple mechanical strategy will make money. So, if you were using it to build up a retirement fund over 25+ years then you might do OK. But most people want to trade full time and make a good living from it NOW.

I tested a mechanical system which followed ALL THE RULES a mechanical system should. I back tested it with 25 years of data and was amazed to discover it turned $5000 into over $1 million in this time. Then I looked more closely at the figures and discovered there was one period where for over 500 days (nearly 2 years!) it was in a drawdown. That is to say, it was below its previous equity peak and took over 500 TRADING days just to get back to it's previous level. Now remember, this was tested on a computer which took a few seconds to run the back test. During this time, the computer didn't get sick or go on holiday or have a girlfriend who demanded to be taken to a romantic restaurant and it never lost its internet connection or any one of a number of things that might cause it to miss a crucial trading day. So, I ran a "What If" scenario and needless to say, I don't trade that or any other mechanical system.


I would agree that a discretionary Trader would probably develop more in-depth skills and probably have a better "feel" for the markets but isn't there also the risk that other factors such as emotions can have an impact on his/her's trading??

Just interested....

Yes, and this is what frightens most people away.
 
Is this discussion assuming a 'mechanical system' is indicator based?
 
Most people don't carry out sufficient back testing, and even if they do, they will find excuses as to why the system failed to produce acceptable results during certain periods. You are right, in the VERY long term even a simple mechanical strategy will make money. So, if you were using it to build up a retirement fund over 25+ years then you might do OK. But most people want to trade full time and make a good living from it NOW.

I tested a mechanical system which followed ALL THE RULES a mechanical system should. I back tested it with 25 years of data and was amazed to discover it turned $5000 into over $1 million in this time. Then I looked more closely at the figures and discovered there was one period where for over 500 days (nearly 2 years!) it was in a drawdown. That is to say, it was below its previous equity peak and took over 500 TRADING days just to get back to it's previous level. Now remember, this was tested on a computer which took a few seconds to run the back test. During this time, the computer didn't get sick or go on holiday or have a girlfriend who demanded to be taken to a romantic restaurant and it never lost its internet connection or any one of a number of things that might cause it to miss a crucial trading day. So, I ran a "What If" scenario and needless to say, I don't trade that or any other mechanical system.




Yes, and this is what frightens most people away.



NewTrader,

Many Thanks for your comments.

I do agree with most of what you say but would you agree that if the Mechanical system being developed fell within acceptable constraints (as defined by the system developer) for such things are MaxDD, Profit % Return, Winning Trade %, etc, etc, then trading such a system could offer a highly level of confidence than trading the same set of rules in a discretionary manner, due to the points mentioned in my other post.

I do personally believe that if a mechanical system is designed correctly then it should be able to make money in ALL time-frames (rather than only LONG term) but obviously the system needs to return a positive expectancy in the first place.

I agree that in the example you mentioned a drawdown period of 2 years is not acceptable. Personally, I wouldn't trade a system if the MaxDD exceeded 10% but thats just my opinion.

In answer to your point about missing trading days due to other commitments this could be resolved by focusing on a longer timeframe for trading. ie a weekly system as opposed to a EOD.

At the end of the day, I guess the main point I am making is that it is down to the individual to decide what kind of personality they have and use what method they feel more comfortable and "in-tune" with.........

Chorlton
 
You are the weakest link. Hello.

Any strategy is only as good as its weakest point. In knowing this we have the opportunity to build strength.

You are the weakest link. Hello.
 
which way round is the risk

Can I ask why this would be?

Surely, if one can find an inefficiency in the markets and carry out sufficient backtesting to ensure that it can be potentially be exploited then surely a mechanical system would be a better approach, as it would ensure that those rules are consistantly followed??

I would agree that a discretionary Trader would probably develop more in-depth skills and probably have a better "feel" for the markets but isn't there also the risk that other factors such as emotions can have an impact on his/her's trading??

Just interested....

Hi Chorlton

I think the risk is the other way round :?:

Would you allow an air line pilot discretion to land the plane (switch over from Auto) if he thought it was justified

Have you ever run a red light late at night when you can see its clear both ways and the lights have got stuck

Switched off the auto wipers on your car because they squeek the windscreen (not enough rain:LOL:)

Backtesting = logic in it :?:

benefit to build a method, but an edge for all conditions

I trade with absolute discretion in a system like way, I am required to work within a % of normal I guess

Emotion creeps in, its part of life, you learn to recognise this quicker and act accordingly or you will be wiped out.

Your on your 3 rd trade of the day

STOP THINK = I only average 1:5 per day = whats going on, am I ....... whatever = experience.

If you rely on a system and I guess there are plenty who do and who are far better traders (money won) who have retired rich and not been really tested, so what they took the risk IMO and it payed off for them, same as the hoards who bought shares in the early eighties and sold them just before the crash in 87 because they were going on holiday or some other random reason.

The point = you are handing control over important decisions to a rule based method, so the rules had better be good. An ordinary engineer is a very skilled man and can make a good living anywhere in the world.

A toolmaker is the next division up :eek: and we do not produce them, not everybody can be one, it takes ....................the whole nine yards of effort and some.

My granddad was such a man, he tried to enlist at the age of 13, he was already a skilled wireless operator (self taught)(world war 1) :LOL: lucky:clover: for him they turned him down. later in life, some of the machines he set up and maintained and kept running built and kept in the air the avro Lancaster and then the Lancaster.

Each to their own but its not for me, I no I am not clever enough to build such a system and let it trade its self. I no there are filters and extreme models, back testing every possible outcome etc etc

At the end of the day I think

ITS ONLY AS GOOD AS THE MAN THAT BUILT IT

For what its worth :LOL: I think to build a really good system you need a full discretion trader to work with an individual who is or does excell in the field of errrr setting it in a program

I am not sure the two go together or exist in any one individual, I like to trade, I like the percieved skill and control I think I exercise, I think or no from results I can keep drawdown to a minimum because of the selective/discretion I exercise.

I am sure the reverse is true for strict method only mek systems.

You can build a pretty good table or car from a simple manufacturing process (fully auto & bored workers)

I do not think you will build anything that will last or stand the test of time and have true worth.

Any really good system traders (average returns / not loss) want to work with a full discretion trader and see what happens :?: it might be :clap::clap:

Teamwork, thats how we got this far, maybe we can go a little further :p

Hey ! whats that :?: "the moon"

Lets go :eek:(y):clap::cry::mad:
 

Attachments

  • moon.jpg
    moon.jpg
    2.1 KB · Views: 762
  • man on moon.jpg
    man on moon.jpg
    4.6 KB · Views: 802
Last edited:
You are unlikely to become a proficient trader using a mechanical system but it’s the easiest route to take and that’s why most aspirants go down that path.

A discretionary trader is much more likely to develop a skill that ensures consistent profits and a permanent edge.

I can see an element of truth in your statement.

From my experience to develop an above average (80%+) mechanical system your rules will have to be derived from some kind of discretionary style. Programming them is a little more complex but nevertheless they are based on a deeper understanding of trading.
 
NewTrader,

Many Thanks for your comments.

I do agree with most of what you say but would you agree that if the Mechanical system being developed fell within acceptable constraints (as defined by the system developer) for such things are MaxDD, Profit % Return, Winning Trade %, etc, etc, then trading such a system could offer a highly level of confidence than trading the same set of rules in a discretionary manner, due to the points mentioned in my other post.

I do personally believe that if a mechanical system is designed correctly then it should be able to make money in ALL time-frames (rather than only LONG term) but obviously the system needs to return a positive expectancy in the first place.

I agree that in the example you mentioned a drawdown period of 2 years is not acceptable. Personally, I wouldn't trade a system if the MaxDD exceeded 10% but thats just my opinion.

In answer to your point about missing trading days due to other commitments this could be resolved by focusing on a longer timeframe for trading. ie a weekly system as opposed to a EOD.

At the end of the day, I guess the main point I am making is that it is down to the individual to decide what kind of personality they have and use what method they feel more comfortable and "in-tune" with.........

Chorlton

The main problem is this: People who want to be mechanical system traders for the reasons you mentioned, invariably end up being quasi-mechanical system traders for the reasons I mentioned.
 
One and the same

Is this discussion assuming a 'mechanical system' is indicator based?

Is a mechanical system not simply a dicretionary system formalised? The only exception I can think of is the trader who only trades on news or instinct, but even then its likely that the money managment side will have some form of mechanisation - apart it would seem from the one used at SocGen.
 
Is a mechanical system not simply a dicretionary system formalised? The only exception I can think of is the trader who only trades on news or instinct, but even then its likely that the money managment side will have some form of mechanisation - apart it would seem from the one used at SocGen.

If one has a permanent edge and it is derived from charts alone, surely in order to perform that edge on a consistent regular basis takes discipline and repetitiveness akin to a performance of a computer program.

Assuming a discretionary method has accounted for 99% of all variables from volatility to range bound periods, and that method has certain criteria, as it must do on both counts, in order to be called a 'permanent edge', then whether it is performed by a trader, computer program or monkey should make no difference.

As you state, only the trader going with instinct or news must be physically present to trade. Though I wouldn't define that as a permanent edge, more an ingrained skill from years of experience. 90% of traders have a set of rules and criteria and even if they choose to be discretionary in the trades they take, that is too something that can also be accounted for in a program.

Personally, if I were to go with a program for my current trading methods, nothing would change except my tolerance for a quality gin Martini.

Just my 2 cents.
 
If one has a permanent edge and it is derived from charts alone, surely in order to perform that edge on a consistent regular basis takes discipline and repetitiveness akin to a performance of a computer program.

Assuming a discretionary method has accounted for 99% of all variables from volatility to range bound periods, and that method has certain criteria, as it must do on both counts, in order to be called a 'permanent edge', then whether it is performed by a trader, computer program or monkey should make no difference.

As you state, only the trader going with instinct or news must be physically present to trade. Though I wouldn't define that as a permanent edge, more an ingrained skill from years of experience. 90% of traders have a set of rules and criteria and even if they choose to be discretionary in the trades they take, that is too something that can also be accounted for in a program.

Personally, if I were to go with a program for my current trading methods, nothing would change except my tolerance for a quality gin Martini.

Just my 2 cents.

It's very easy just to say these things but it is very different putting it into practice. I've said it before and I will say it again, people overestimate the power of computers. Computers don't learn from experience and they do not have realizations.

Imagine the simple task of driving a car from Point A to Point B. Sounds easy enough to program a computer to do this, especially now with GPS guidance. But now think of all the things you would have to account for, things that you have gained through driving experience. Traffic jams, accidents and accident avoidance, unexpected road closures, kids running across the street, animals dashing across the street etc..etc..Think about it deeply and carefully. If you honestly have any sort of programming experience you would know that this would be a near impossible challenge.

I recall one time I was driving along a quiet suburban street when a tennis ball came out of knowhere and started bouncing across the street. The car infront of me didn't slow down, but I did. I knew that there will be a kid running after that ball who wont check to see if any cars are coming. I was right. The car infront had to skid to a halt and blew his horn :rolleyes:

I know that driving a car is very different to trading but it highlights how easy it is for people to take what we do for granted. Especially when I read things like "that is too something that can also be accounted for in a program."

Remember computers don't see in the way humans do. 'Seeing' or pattern recognition would consume a phenomenal amount of processing power. Do you think it's just a simple matter of programming: If pattern = head and shoulders Then ?
 
It's very easy just to say these things but it is very different putting it into practice. I've said it before and I will say it again, people overestimate the power of computers. Computers don't learn from experience and they do not have realizations.

Imagine the simple task of driving a car from Point A to Point B. Sounds easy enough to program a computer to do this, especially now with GPS guidance. But now think of all the things you would have to account for, things that you have gained through driving experience. Traffic jams, accidents and accident avoidance, unexpected road closures, kids running across the street, animals dashing across the street etc..etc..Think about it deeply and carefully. If you honestly have any sort of programming experience you would know that this would be a near impossible challenge.

I recall one time I was driving along a quiet suburban street when a tennis ball came out of knowhere and started bouncing across the street. The car infront of me didn't slow down, but I did. I knew that there will be a kid running after that ball who wont check to see if any cars are coming. I was right. The car infront had to skid to a halt and blew his horn :rolleyes:

I know that driving a car is very different to trading but it highlights how easy it is for people to take what we do for granted. Especially when I read things like "that is too something that can also be accounted for in a program."

Remember computers don't see in the way humans do. 'Seeing' or pattern recognition would consume a phenomenal amount of processing power. Do you think it's just a simple matter of programming: If pattern = head and shoulders Then ?

I know very little about programming and how such a thing would work but 70 years ago a computer looked like this and now you can do everything on this......
 

Attachments

  • 1939_ABC_large.jpg
    1939_ABC_large.jpg
    237.6 KB · Views: 616
  • 1941_zuse_z3_large.jpg
    1941_zuse_z3_large.jpg
    204.1 KB · Views: 1,236
  • img_20781_nokia_n95_8gb.jpg
    img_20781_nokia_n95_8gb.jpg
    41.6 KB · Views: 1,419
Just wanted to add: Think of what a child can EASILY do by the age of 5 and compare it to a multi-billiondollar corporation: ASIMO - Wikipedia, the free encyclopedia

Doctor Robert Zubrin who is an advocate of a manned mission to Mars as opposed to robotic missions proposed the following challenge to NASA. An Easter Egg hunt where he would have one 10 year old child and NASA could have 1000 rovers and see who ends up with the most eggs.
 
  • Like
Reactions: TWI
Top