Discretionary Versus Mechanical System Trading

[cnms2]

Quote from xtrhvydty:
My understanding is that a purely discretionary trader should have >10yrs experience. I'll buy that.

But I believe that the best and worst part of the system is us.
...
I understand the need to acquire experience, I just don't agree with this mechanical mindless drone argument.

Why should a trader remove his best asset from his investment?

Mind will help you some times will let you down other times. Statistics is objective, but can be brought down by mind: both through data fed in, and through result interpretation. You have to use them properly and cautiously.

 

[illiquid]

Quote from stephencrowley:

Because these shapes you are drawing on your chart is just wishful thinking on your part. Your brain comes up with all sorts of "patterns" out of seeming randomness. The reason to disassociate yourself from trading is that too often our brains are fooled by randomness and emotional decision making takes over.

Life can be represented and explained thru numbers, and that includes the markets.

What is the difference between "wishful thinking" pattern/trend following and projecting statistically "significant" numbers analysis from the past into the future? Both are just predictions that what has happened will continue to happen. But anything that appears to "work" reliably in the past can start to fail miserably tomorrow; anything that appears to be random and statistically insignificant today can become the holy grail for the next year. Same achilles heel.

Pattern/trend followers and statisticians seem to share more similarities than differences to me.

 

[stephencrowley]

I completely agree that any pattern found in a simple univariate time series is bound to completely disappear sometime, the trick is to find 'hidden' time series which are constructed via other means that display short term patterns which persist over a long time that are not visible to the majority of traders.

Quote from illiquid:

What is the difference between "wishful thinking" pattern/trend following and projecting statistically "significant" numbers analysis from the past into the future? Both are just predictions that what has happened will continue to happen. But anything that appears to "work" reliably in the past can start to fail miserably tomorrow; anything that appears to be random and statistically insignificant today can become the holy grail for the next year. Same achilles heel.

Pattern/trend followers and statisticians seem to share more similarities than differences to me.

 

[cnms2]

Quote from illiquid:

What is the difference between "wishful thinking" pattern/trend following and projecting statistically "significant" numbers analysis from the past into the future? Both are just predictions that what has happened will continue to happen. But anything that appears to "work" reliably in the past can start to fail miserably tomorrow; anything that appears to be random and statistically insignificant today can become the holy grail for the next year. Same achilles heel.

Pattern/trend followers and statisticians seem to share more similarities than differences to me.

One difference: TA works on equity indices as well as on equities, trend following doesn't work on indices. TA indicators measure market internals too, price shapes don't. Both rely on statistics but on different sets of data. Now: I agree that a good trend following system could beat a poor TA based system. But I can't imagine an indicator as MACD stopping to work.

 

[illiquid]

Quote from stephencrowley:

I completely agree that any pattern found in a simple univariate time series is bound to completely disappear sometime, the trick is to find 'hidden' time series which are constructed via other means that display short term patterns which persist over a long time that are not visible to the majority of traders.

And how can you tell when that pattern you trade ceases to be significant and becomes no better than "random"? How can you tell that a period of drawdown in only a temporary setback, or a sign that your signal has truly lost predictive significance? How many times does it need to fail before you stop using it?

Isn't this the same issue facing those that jump onboard a trend and hold until it appears broken, albeit on another level? That is, staying with it while it works, and getting off when it stops? The same issues of rear-view projection apply in both cases.

 

[stephencrowley]

The patterns I've found are deep and persistant and change very slowly over time. There are temporary diversions from the long term during the day which I take advantage of and hedge appropriately. Basically, for the patterns I use, all money managers and traders in the world would have to have to agree at once to change their methods at once in some sort of grand conspiracy. These patterns do change, but I optimize daily and thus my models self-adapt to changing patterns and are always scanning the markets for new ones.

Quote from illiquid:

And how can you tell when that pattern you trade ceases to be significant and becomes no better than "random"? How can you tell that a period of drawdown in only a temporary setback, or a sign that your signal has truly lost predictive significance? How many times does it need to fail before you stop using it?

Isn't this the same issue facing those that jump onboard a trend and hold until it appears broken, albeit on another level? That is, staying with it while it works, and getting off when it stops? The same issues of rear-view projection apply in both cases.

 

[Chicken Little]

Quote from makosgu:

I have found it to be an acquired SKILL anyway one dices it up. ART implies that one can be creative about how to tap into the market's delivery stream when in fact, the market's delivery has no correlation with one's creativity in tapping tinto the stream. A painting is creative and artists are rewarded by the number of people who find and value the artists work to be creative. What underlies ALL effective tools wrt the market, is their ability to read and be in synch with the markets dynamics. Is it science? Too many textbooks have yet to answer this Q... Is it mathematical? Is it logical? Any system whether discretionary or mechanical need only do one thing, read, partner, and keep in synch with the market... The better the skill, the more effective the systems is, the greater the delivery... Like most things, people seam to not like to read, this is especially evident in ET wherever one finds long posts... Shortly thereafter, follow up posts state, "i didn't can't past your first sentence", "too long", "sum up in one word", "jibberish", "long winded babble"??? What is strikingly profound is that the market is also read. Market illiteracy is strongly coupled with failure. We all aware of the 95% failure rate. It might as well be the market illiteracy rate...

Hey Harvard boy use the spell check !!! LOL

 

[makosgu]

Quote from Chicken Little:

Hey Harvard boy use the spell check !!! LOL

LOL!!! Like a troll, you just hang on to everything I post. Spell check doesn't catch that because it IS spelled correctly. You misread the error as a spelling instead of grammatical. Punch in the sentence and run spell check, you will see that it does not find SEAM as misspelled. ET does not come with GRAMMAR check. "Jibberish" is the misspelled word for the record. What a piece of work you are. Your handle is getting handled all the time. PATHETIC!

 

[Chicken Little]

Quote from makosgu:

LOL!!! Like a troll, you just hang on to everything I post. Spell check doesn't catch that because it IS spelled correctly. ET does not come with GRAMMAR check. "Jibberish" is the misspelled word for the record. What a piece of work you are. PATHETIC!

Your ideas reflect the same attention to detail you give to grammar and spelling. Mullet...

 

[OddTrader]

Discretionary is winning:

http://www.barclaygrp.com/index.html

Q

Barclay CTA Indices => October ROR† => Percentage of funds reporting => YTD through October†
Barclay CTA Index => -0.17% => 84.31% => -0.72%
Agricultural Traders Index => 1.41% => 58.33% => 2.82%
Currency Traders Index => 0.53% => 82.05% => -1.36%
Discretionary Traders Index => 0.29% => 77.63% => 5.96%
Diversified Traders Index => -0.93% => 84.34% => -2.73%
Fin./Met. Traders Index => 0.66% => 86.11% => 0.55%
Systematic Traders Index => -0.20% => 86.36% => -1.79%

UQ