Any discretionary traders left?

[bullet21]

i always thought they were the majority, but the more i read the more it seems they're few and far between. Why is it that it seems the authors of these books seem to opt more for the Mechanical route?

 

[explod]

i always thought they were the majority, but the more i read the more it seems they're few and far between. Why is it that it seems the authors of these books seem to opt more for the Mechanical route?

Charts are an image in which we can see many things. The old saying "a picture is worth a thousand words" is worth thinking about.

I went down the tips, then the chart/analysis/computer identification method and found it worked well overall but takes away your life. You live with the screen and market action.

Fundamentals is hard work, it is endless but if you can put it in, in my view it gives the rewards. Find a switched on group developing a new company and getting on for the ride at the right time has been fantastic for me the last two years.

But to sell books, no one wants to put the mundane work into something that is also mundane to read.

No pain no gain, but we do not like to hurt these days.

just my

 

[professor_frink]

i always thought they were the majority, but the more i read the more it seems they're few and far between. Why is it that it seems the authors of these books seem to opt more for the Mechanical route?

Hi bullet,

which books are you referring to here? To be honest I always thought discretionary traders were the majority too.

Personally I think that trying to incorporate as much testing as is possible into even the most discretionary set of methods is a good thing though. If it can be tested then it should be. If it can't be tested then sim trade it until you can get some numbers to work with.

 

[So_Cynical]

i always thought they were the majority, but the more i read the more it seems they're few and far between. Why is it that it seems the authors of these books seem to opt more for the Mechanical route?

I'm a discretionary investor i figure You can sell a system but you cant really sell discretion...therefore there's alot of books about systems.

 

[tech/a]

I'm a discretionary investor i figure You can sell a system but you cant really sell discretion...therefore there's alot of books about systems.

Discretionary traders can and should find out the exact same numbers on their discretionary trading method as those who use mechanical systems.

If I could provide 5000 trades in a discretionary manner over 20 yrs which returned me 25% compounding over that time Ill bet I could fill a hall at $300 a head no problems and sell a heap of books to boot!

I cant---yet so no hall has been booked-- and the ink is still being penned.

 

[It's Snake Pliskin]

i always thought they were the majority, but the more i read the more it seems they're few and far between. Why is it that it seems the authors of these books seem to opt more for the Mechanical route?

Perhaps it's not easy for true discretionary traders to write how they operate when they rely on nuance and particular circumstances. That would be a tome of immense proportions.

 

[Grep]

I tried for years to successfully trade stocks via discretionary methods.
I gave up because the best I could manage is break even.

I have done much better with 100% mechanical stock systems using weekly bars. I also trade a mechanical daily bar system on stocks, but it too has pretty ordinary performance, i.e about the same as the overall index.

On the other hand, FOREX and commodities have been much nicer to me over the years on both mechanical and discretionary trading.

Both my mechanical and discretionary trading methods are trend following in nature.Stocks just dont seem to trend as well as other tradable instruments unless you look really long term such as weekly or even monthly.

 

[bullet21]

Hi bullet,

which books are you referring to here? To be honest I always thought discretionary traders were the majority too.

Personally I think that trying to incorporate as much testing as is possible into even the most discretionary set of methods is a good thing though. If it can be tested then it should be. If it can't be tested then sim trade it until you can get some numbers to work with.

Penfold, Radge and Tharp's books seem too all at least lean towards the mechanical side. Penfold definately is mechanical.

But the "sim trading a discretionary system" part is what i dont understand. If you do 30 sim trades then calculate your expectancy its not really accurate is it? because you may have altered your system in each trade ie: by using your discretion. so all the variables you get to calculate your expectancy, risk of ruin etc aren't as concrete.

garbage in, garbage out?

sorry im still learning and just need to get my head around this. Im leaning towards mechanical at the moment, but the only programming i know is some Visual Basic from high school : . just trying to decide which avenue to pursue.

 

[professor_frink]

Penfold, Radge and Tharp's books seem too all at least lean towards the mechanical side. Penfold definately is mechanical.

But the "sim trading a discretionary system" part is what i dont understand. If you do 30 sim trades then calculate your expectancy its not really accurate is it? because you may have altered your system in each trade ie: by using your discretion. so all the variables you get to calculate your expectancy, risk of ruin etc aren't as concrete.

garbage in, garbage out?

sorry im still learning and just need to get my head around this. Im leaning towards mechanical at the moment, but the only programming i know is some Visual Basic from high school : . just trying to decide which avenue to pursue.

I'm talking of a little more than 30 trades! IMO the process is similar to what the mechanical trader goes through when trying to find a profitable system to trade. You start playing around for awhile, look at various ideas until you find one your happy with. From there test it as much as you possibly can.

A lot of software programs will have some kind of replay feature on them, you can pick a stock code, a start date, and it will start printing bars at whatever speed you choose. I use amibroker, and can set up a running equity curve below the chart as it's replaying, and with the ability to enter orders at the end of each bar. Once I'm done, it will spit out all of the same stats that I would get from a coded test.

There is always the risk of garbage coming out at the end of a test. But then the same can be said about a mechanical test too. There is always the risk of over optimising or some inappropriate code that peeks into the future.

 

[bullet21]

I'm talking of a little more than 30 trades! IMO the process is similar to what the mechanical trader goes through when trying to find a profitable system to trade. You start playing around for awhile, look at various ideas until you find one your happy with. From there test it as much as you possibly can.

A lot of software programs will have some kind of replay feature on them, you can pick a stock code, a start date, and it will start printing bars at whatever speed you choose. I use amibroker, and can set up a running equity curve below the chart as it's replaying, and with the ability to enter orders at the end of each bar. Once I'm done, it will spit out all of the same stats that I would get from a coded test.

There is always the risk of garbage coming out at the end of a test. But then the same can be said about a mechanical test too. There is always the risk of over optimising or some inappropriate code that peeks into the future.

thank you for your informed input, im wish to purchase Amibroker. do you know if its possible to buy somewhere in Aus or do you have to get it off their website?

 

[professor_frink]

thank you for your informed input, im wish to purchase Amibroker. do you know if its possible to buy somewhere in Aus or do you have to get it off their website?

No worries. I got it off their website, as far as I know that's the only place to grab it.