Jun

4

 I'm reading Trading as a Business by Charlie Wright. Pretty good book profiling the evolution from discretionary trader to systematic trader. One of those books where I found myself laughing at having been down the paths. More trend following oriented but I think it is a pretty good synopsis of the systematic world and he covers some bases that added value in terms of elements to consider in one's trading (or at least mine). Decent set of checklists.

Do systematically inclined speculators recommend similar books (besides Victor Niederhoffer's and Larry Williams books).

Also, Tradestation seems to do most anything a trader would want in terms of trend following testing. I have never used it though.

Thoughts?

George Parkanyi writes:

The only flaw I find with systems is that they immediately stop working as soon as you try to use them. I think people need to do more research on fading systems.

Christopher Tucker writes: 

Where's the "like" button on the Speclist?

Steve Ellison adds: 

Yes, even systems I developed myself stop working when I try to use them because of data mining bias. Even if there legitimately is an edge, some component of the good backtesting performance is better-than-average luck. 

Leo Jia writes: 

The word "enlightened discretionary" is very appealing. The reason for it, I guess, is because of the word "enlightened" more than the word "discretionary". Everyone hopes to be enlightened in someway. Being enlightened seems to be a spiritual consummation. But I guess that is not the first and real reason why people are after being enlightened. The real reason is that it is mystic and mostly unattainable. This coincides with a human nature of always craving for what they don't have, which is among the reasons why most people are persistently unhappy.

I feel preferring discretion to system is quite illogical. Aren't whatever rules one uses as a discretion by nature a system? It perhaps is not explicitly sketched out, but it by all means is a system of rules that resides in one's head. Couldn't that be phrased and then programmed? I agree some are not very easy. But are they really impossible?

Gary Phillips writes: 

I've been doing this long enough to instinctively know what works and what doesn't. I only need to look at my P&L for empirical confirmation. If in doubt I just try to see the market for what it is and not what it appears to be. One needs to understand market structure, liquidity, and price action and develop a framework for analyzing the market, somewhere between bottom-up & top-down lies the sweet spot. This allows you to see the market in the proper context and provides you with a compass, which will keep you from feeling lost and will show you the way.

Craig Mee writes: 

Aren't ?

Hi Leo, you probably could say "whatever rules one uses as a discretion by nature is a system", but a system may not have the ability to load up once the move kicks (obviously it can be programmed) but at times the opportunity may appear intuitive, and  a trader can do that on relatively short notice, whilst keeping initial risk limited.

Interesting, Gary, the issue with systems seems to be at times data mining against price action and structure which gives strength of understanding. The HFT may work on massive turnover, low commissions and effectively front running, and unless you have those edges then it appears difficult to succeed from a data mining basis (and relatively scary trading something that you don't effectively understand from a logical point of view). However classifying a markets structure, and working off 3-4 premises no more, (as I believe more would allow any edge to be diluted across a range of options), and the ability to leverage once on a move, appears to be something you can work with. This is purely from a hands on execution basis, no doubt the pure programmers can weigh in.

I remember speaking to a guy who professionally programs for others… (admittedly a lot of retail), and we were talking about what are the laws in place for him to not front run me after developing a system I gave him…and he was like "mate, to be honest (probably insinuating "dont flatter yourself") 97% don't make a dime." That was certainly probably expected I suppose, but to hear it in technicolour was confronting and I was surprised he said as much.

Gary Phillips writes: 

I really don't believe that discretionary trading today, is any harder than it used to be. The emotional aspects, and risk management, have essentially remained the same. Methodology is different, because algorithmic driven HFTrading has forced intra-day traders to change from momentum chasers to mean reversion traders. And as you stated, there are countless global/macro concerns as a result of the financial crisis and continued global easing. So, it does demand a broader universe of knowledge, and revamped techniques and benchmarks, but it still boils down to identifying what is truly driving price and how it is being driven. 

I guess this is what gives you the elusive *edge*. But, as we used to say the *edge* can sometimes be the *ledge.* That being said, trading doesn't have to be about being right or wrong the market, or predicting where the market is headed in the next moment, hour, day or week. Trading can be nothing more than a probabilistic exercise, and a trade nothing more than a statistical data point - the next event in a series of events governed by the statistical random distribution of results.

 

Kim Zussman writes: 

"Trading can be nothing more than a probabilistic exercise, and a trade nothing more than a statistical data point - the next event in a series of events governed by the statistical random distribution of results."

One would suggest that trading is a waste of time if your historical or expected mean are random.


Comments

Name

Email

Website

Speak your mind

3 Comments so far

  1. Gary Phillips on June 4, 2012 7:41 pm

    The point is Kim, you simply can’t treat trades and prediction the same way, and far too many focus on prediction. Prima facie, it seems like trading is a form of prediction but it really isn’t. If you you implement a methodology that solely attempts to predict the market, it won’t work. You have to concentrate on projecting losses, risk management and finding something that works, but if you are directly looking for prediction, it can be counter-productive or even self-defeating.

  2. Brendan on June 5, 2012 7:01 pm

    Sorry for the newbie question, but which books are "Vic and Larry's books" (from original posting). By Vic's book, I assume you mean the "Trader Vic" book? Which one is Larry?

    Many thanks

    Ed.: George had already read Victor Niederhoffer's book(s) and Larry Williams' book(s) so he wanted suggestions for other books. 

  3. Chris Figaro on June 6, 2012 12:47 pm

    I agree 100% with Gary: “I really don’t believe that discretionary trading today, is any harder than it used to be. The emotional aspects, and risk management, have essentially remained the same.”

    No matter what edge one might think they have, any successful trader must adhere to a strict risk management system and not let his or her emotions effect the decision making process. In a way, all good traders are systematic in that they will, or should, look at an opportunity mathematically in terms of risk/reward and size accordingly.

    In terms of Tradestation…it’s great for single stock backtesting, however, it does not allow for backtesting a portfolio. Tradingblox would be the program of choice but a little less user friendly.

Archives

Resources & Links

Search