Is fully automated trading a fantasy? Your opinions here

[TraDaToR]

I think( but not sure ) that profitable discretionnary traders are close to a mechanical approach, even if they don't acknowledge it... They are not changing their setups every day.

They have found a method that suited their personnality, and the "unmechanical" improvements are quite thin...So if you code the closest system, you can apply it to a large panel of markets. That's why autotrading is superior in my opinion, just because you can be extremely diversified and handling 50 systems every day...

Personnally, I stop my systems manually when DD has been too deep( based on predetermined levels ), so I'm not "fully automated", but totally mechanical at every level.

But living off 10000 $??? You may know systems I'm definitly not aware of

 

[MarkBrown]

if you can make one point a day on the emini sp you will will own the world. take into consideration when i say that - that includes blowing up drawdowns etc.

btw i have gone on short vacations - been at conferences - given lectures - and all the while my systems were trading away. i have been doing this a long time too. software, hardware, datafeeds, redundancy out the yang yang - my hp ipaq with remote vnc - all allow me to do what i want where i want. i have servers in chicago, texas and i am typing you this from florida setting in a motor coach with a wireless evdo network using a junxion box. in april i will type you from my md explorer flying back from a night in paris heading back home to geneva.

when someone says i made so much this week - i ignore it - when someone ask what did your systems do yesterday - i ignore it - see to me its about the Self-discipline of applying what i know in mechanical terms. if i get wrapped up in every trade i will never succeed. you cant drive the freeway looking down 10 feet in front of you. as far as tweaking a system - then you didn't have anything in the first place.

tweaking should be done prior to implementation. testing a system should be done over thousands and thousands of trades to make sure you have something real. just about no one does that and so many have short run success stories.

 

[Wayne Gibbous]

Takes a while to put it all together and gain the confidence to let it run.

The code and interfaces necessary are simple enough. The hard part is letting go of the throttle and steering wheel and letting it fly.

Just takes time.

Good trading to all.

 

[TrueStory]

It is quite possible to earn a substantial living trading automated systems.

That is not to say there are great ones for sale. However, I have seen publicly available systems that make money in real-time. Assuming it's impossible is simply naive. Some system developers do not have the capital to trade their own systems, so they lease them out or sell them...

 

[MarkBrown]

Quote from TrueStory:

Some system developers do not have the capital to trade their own systems, so they lease them out or sell them...

you know thats a good point - most people fly on commercial aircraft and YET they did not build the craft or fly the craft - but they got to where they were going just the same.

i was quoted in a book one time as saying i didn't have to make the toilet paper in order to use it.

 

[asap]

Quote from TrueStory:

It is quite possible to earn a substantial living trading automated systems.

That is not to say there are great ones for sale. However, I have seen publicly available systems that make money in real-time. Assuming it's impossible is simply naive. Some system developers do not have the capital to trade their own systems, so they lease them out or sell them...

no.

they sell them because it is wiser from a risk benefit standpoint. that is, leasing the software for $50 is better deal for them than putting their own money at risk with their automated system.

i am not saying it is impossible or even difficult to make money with those systems, just arguing that charging $50 per head and month is a better deal.

obviously there are no holy grails, so any automated system is a straightforward representation of risk and reward. After comparing each and every risk reward available, the subject picks the best one.

there are systems that have exceptional risk reward profiles, they have been customized and tailored to benefit from market discrepancies or non random patterns. those wont be for sale for $50 or even for any amount at all. just bare in mind that you get is what you pay for.

 

[TrueStory]

Quote from markbrown:

if you can make one point a day on the emini sp you will will own the world. take into consideration when i say that - that includes blowing up drawdowns etc.

btw i have gone on short vacations - been at conferences - given lectures - and all the while my systems were trading away. i have been doing this a long time too. software, hardware, datafeeds, redundancy out the yang yang - my hp ipaq with remote vnc - all allow me to do what i want where i want. i have servers in chicago, texas and i am typing you this from florida setting in a motor coach with a wireless evdo network using a junxion box. in april i will type you from my md explorer flying back from a night in paris heading back home to geneva.

when someone says i made so much this week - i ignore it - when someone ask what did your systems do yesterday - i ignore it - see to me its about the Self-discipline of applying what i know in mechanical terms. if i get wrapped up in every trade i will never succeed. you cant drive the freeway looking down 10 feet in front of you. as far as tweaking a system - then you didn't have anything in the first place.

tweaking should be done prior to implementation. testing a system should be done over thousands and thousands of trades to make sure you have something real. just about no one does that and so many have short run success stories.

Well said...

 

[TrueStory]

Quote from asap:

no.

they sell them because it is wiser from a risk benefit standpoint. that is, leasing the software for $50 is better deal for them than putting their own money at risk with their automated system.

i am not saying it is impossible or even difficult to make money with those systems, just arguing that charging $50 per head and month is a better deal.

obviously there are no holy grails, so any automated system is a straightforward representation of risk and reward. After comparing each and every risk reward available, the subject picks the best one.

there are systems that have exceptional risk reward profiles, they have been customized and tailored to benefit from market discrepancies or non random patterns. those wont be for sale for $50 or even for any amount at all. just bare in mind that you get is what you pay for.

You're right about it being a better deal...

 

[Dustin]

Why are you so focused on just having $10k when there are prop shops everywhere offering guys like you 10:1 or more, with good platforms and API's included?

If you are really considering this as an occupation then it's time to get serious and stop worrying about funding issues. Nobody makes 5-10% per week consistently, but I do know guys making a great living off fully automated systems.

 

[MarkBrown]

Quote from TrueStory:

You're right about it being a better deal...

i fully agree, one of my partners and i set back and look at larry connors web site and we think darn man - why didn't we do that. it would be like the guys who pit trade in chi town reading larry williams books (prior to his exposure as having not even enough money to bail himself out of jail) and thinking to themselves man now thats the life. sell books yada yada yada.

i would have made more money more easily selling systems than trading i think. but sooner or latter all these guys get found out and its over. when you can trade and make killing then after taxes and train wrecks you too can be like me and just make a living.

so thats my motto now "you have to make a killing to make a living".

trademark 2007 markbrown