Death of Discretionary Traders??

[Dogpile]

One thing big program trading firms can't do is carry positions with resting stops that get executed without slippage. I am talking about the big institutions here, not the nimble < $1B hedge fund.

To the extent that all trading involves probabilities, not certainty -- the use of stop-loss orders limits losses for all the times where you suffer 'variance' to the expected result. Especially in a world where the 'non-linear break' is always a risk due to growth of derivatives.

There is really only one good reason why a LTCM can blow up -- they could not get out of their positions. Being nimble is a real advantage.

[brownsfan019]

dog is right, being nimble has its advantages. I remember when I would visit mutual fund co's that were trying to get us to sell their funds and one of the biggest (American Funds) and one of the best would tell us that it takes MONTHS to build and MONTHS to unload a position. When you get so big, you have to enter positions in such a way as to not raise any flags which is easier said than done.

There's something to be said to be able to flip on a dime, but that also implies you are trading very little in the grand scheme of things. I'd like to think of myself as a nimble, bigger little guy.

[Dogpile]

A quote I liked from 'Mastering The Trade':

"Some of the best traders I know have been trading the same set-up on the same timeframe on the same market for 20 years." pg 31

[darthtrader]

i think its interesting if you look at the CTA's on autumn gold i would say probly 90% list themselves as systematic instead of discretionary. the funny thing is though the 10% with really great returns list themselves as discretionary.
i wouldnt doubt its something like a Rentech/Jim Simmons effect. With how good they have been doing now everyone wants to be like Jim Simmons and discretionary is a dirty word.

until we have computers as powerfull as the human brain i don't think we are in much trouble.

[ItalianSharp]

Big institutions use highly sophisticated automated programs for their trading; they have the best and fastest connections to the markets; they have the best trading platforms. All this makes it nearly impossible for a private trader to compete with them on the same ground. Basically, if you go for the automated stuff, you'll always be one step behind the institutionals.

However, I do believe private traders have a big edge over institutionals. This edge is CAPITAL and being able to liquidate positions at any given time with almost zero slippage.

Discretionary trading is the way to go if you want to compete against those monster funds, banks, bots, whatever you wanna call them.

You have to be good at it, but you can do it. And things will likely improve in the future because there will be more participants bringing more liquidity.

Last but not least, markets will always be driven by fundamentals. Although there is an increasing tendency to automize everything, there will always be the need for human flexibility and interpretation of market conditions.

IS

[ant]

A lot of excellent responses here, so I'll just add to some of the themes already mentioned. It is my opinion that to be a viable opponent in the markets, traders will have to rely on the abilities that make human beings superior to computers. We all know that the human brain cannot compete against a computer when it comes to routine, raw number-crunching, analytical, information-based tasks (i.e., purely left-brained abilities), and this applies to trading strategies that can be programmed and automated. We currently live in an information age where there is an abundance of information and incredibly fast computers. This sets the stage for efficient automation. Computers are significantly faster, they can process vast amounts of information in significantly less time than humans, they don't get tired like people do, and they don't make errors. Generally speaking, if you are an individual trader with a trading strategy that can be described in a finite number of steps, a computer will be able to trade it more easily and better than you. I agree with ItalianSharp when (s)he wrote:

	Quote:	ItalianSharp
	Big institutions use highly sophisticated automated programs for their trading; they have the best and fastest connections to the markets; they have the best trading platforms. All this makes it nearly impossible for a private trader to compete with them on the same ground. Basically, if you go for the automated stuff, you'll always be one step behind the institutionals.

So in my opinion, very few automated, systems traders will be able to compete with institutions that have the best minds, deepest pockets, fastest computers, best automated trading systems; in short, the best, and most abundant, resources. So what type of trader do I think will be able to compete with the institutions? Discretionary Traders. Specifically, traders that combine the analytical, problem-solving skills found in the left hemisphere of the brain with the human abilities carried out by the right-side of the brain. These right-brain abilities include seeing relationships and integrating those relationships into the big picture, pattern recognition, focusing on context, creatively combining ideas, understanding what your competitors are doing and how they may be feeling, and looking at the markets in new and different ways. The list goes on, but these are all examples of areas where human beings excel over computers and the focus of most discretionary traders, I think. Traders that use a holistic approach, combining left and right brain abilities, will be able to continue to compete in the financial markets against institutions with vast computing resources. Anyway, this is my current belief which has shaped my trading style and methodology.

Soultrader, I think you started a thread on an important topic that will be discussed heavily by many traders as time goes on. I believe it's important to consider how we plan to adapt in a world that's becoming more and more automated.

Recommendation: Read "A Whole New Mind" by Daniel Pink, an easy, enlightening read that discusses this subject in more detail. Dogpile mentioned this book in one of his posts.

[ItalianSharp]

	Quote:	ant
	A lot of excellent responses here, so I'll just add to some of the themes already mentioned. It is my opinion that to be a viable opponent in the markets, traders will have to rely on the abilities that make human beings superior to computers. We all know that the human brain cannot compete against a computer when it comes to routine, raw number-crunching, analytical, information-based tasks (i.e., purely left-brained abilities), and this applies to trading strategies that can be programmed and automated. We currently live in an information age where there is an abundance of information and incredibly fast computers. This sets the stage for efficient automation. Computers are significantly faster, they can process vast amounts of information in significantly less time than humans, they don't get tired like people do, and they don't make errors. Generally speaking, if you are an individual trader with a trading strategy that can be described in a finite number of steps, a computer will be able to trade it more easily and better than you. I agree with ItalianSharp when (s)he wrote: So in my opinion, very few automated, systems traders will be able to compete with institutions that have the best minds, deepest pockets, fastest computers, best automated trading systems; in short, the best, and most abundant, resources. So what type of trader do I think will be able to compete with the institutions? Discretionary Traders. Specifically, traders that combine the analytical, problem-solving skills found in the left hemisphere of the brain with the human abilities carried out by the right-side of the brain. These right-brain abilities include seeing relationships and integrating those relationships into the big picture, pattern recognition, focusing on context, creatively combining ideas, understanding what your competitors are doing and how they may be feeling, and looking at the markets in new and different ways. The list goes on, but these are all examples of areas where human beings excel over computers and the focus of most discretionary traders, I think. Traders that use a holistic approach, combining left and right brain abilities, will be able to continue to compete in the financial markets against institutions with vast computing resources. Anyway, this is my current belief which has shaped my trading style and methodology. Soultrader, I think you started a thread on an important topic that will be discussed heavily by many traders as time goes on. I believe it's important to consider how we plan to adapt in a world that's becoming more and more automated. Recommendation: Read "A Whole New Mind" by Daniel Pink, an easy, enlightening read that discusses this subject in more detail. Dogpile mentioned this book in one of his posts.

Thank you for expanding on my thoughts. You laid it out brilliantly.

IS

[Dogpile]

one of the original questions posed in this thread was:

"Do we even need [non-systems] traders? In the near future, are we likely to see more systems and less traders?"

The market, with so many participants, with differing timeframes, with differing strategies is clearly enormously dynamic and complex. With growth of derivatives and leverage, there is a good argument going on about how the market will become increasingly non-linear --more in line with 'Chaos theory'...

If this is true, 'chaos' plays against those who rely on raw compute-power. It will be increasingly difficult to 'model' (system trade) something that is increasingly chaotic and dynamic.

You might say, yah well how is anyone going to operate in this environment then?

From recent book I read by Hank Pruden ("The Three Skills of Top Traders"):

"But the true elegance of chaos theory lies in its correlation and compatibility with the qualities of the old-time technicians."

Dalton talked about how the human brain is not good at interpreting large amounts of complex information. But the brain is actually quite good at identifying well-structured visual objects (and 'patterns'). For example, looking at the equation for something like standard deviation takes a minute to go through and understand it. But looking at a bell-shaped histogram and marking off 1 standard deviation on either side is quite intuitive. Add multiple dimensions of complexity and then add in outside shocks like hurricanes, terrorism, political uncertainty that develops overnight etc... and it quickly gets overwhelmingly complex.

Savvy technical analysts appear to have the opportunity to benefit from all of this. If you can have the patience to wait for the times where price 'auctions too far' due to big clumsy institutions fighting with each other and can therefore identify short-term asymmetric reward:risk positions with regard to 'trade location' -- and you can combine this advantageous trade location with a good entry/exit technique -- and you can enter and exit quickly with reasonable risk and no slippage, you do possibly have a good advantage over institutions.

[darthtrader]

"But the brain is actually quite good at identifying well-structured visual objects (and 'patterns')"

there is some interesting videos on google video if you search for "singularity" as far as the current state of AI. Even beyond the brain being fantastic at pattern recognition, computers are still a joke there. I believe ive read its still impossible to have a computer learn a generalized rule for what a cat looks like where you could show a picture to any 6yo and they would instantly know if they are looking at a cat vs a dog.

I also read IBM's Bluegene super computer is the fastest in the world, has a whole 1/1000th of the brains computer power and they have almost simulated half a mouse brain with it....i'm sure it would be impossible to trade against a computer if it ever has the brains pattern recognition but that would seem to be ways off if ever.

i'll have to check out that The Three Skills of Top Trading book, looks quite interesting.

[Dogpile]

Actually, the book isn't all that good except to make some points about how classic technical analysis might be the best way to compete in a world saturated with computer algorithms competing with each other. But past that, the point is not all that well-developed in the book.

It seems to me this is a little like sitting in a poker game with a bunch of computers that are playing against each other with multi-billion dollar stacks. You have your little mini-stack and patiently wait for monster stack 1 to get out of line while going after monster stack 2. You take a little itsy bitsy piece out of stack 1 who at the same time takes a big chunk from stack 2.

Because one multi-billion dollar stack is competing for some meaningful amount, they can't possibly be nimble enough or even care for that matter to try to go after your little stack -- they are after the big bucks. You can sit at the table and make a living off the scraps that are left from the 'excess' created by the bigger (higher timeframe) war going on.