Day Trading Systems

Toby Crabel, who manages $1B, never had a down year, and his worst month in 8 years (last month) he was down 2% and some change, is 100% systematic.

A quote from "The Day Trader's Advantage" by Howard Abell

Crabel:
I know one thing for me is that, in order for me to make money, the more outside pressure I have, the harder it is for me to trade. I think the only way you can circumvent that is to make it systematic.
 
On the subject of trading systems and software (the original topic) I second ttcrep's suggestion of
www.wealth-lab.com

Its unbelievable to me that this is not the hugest thing in trading right now. I used Tradestation, MetaStock, Neuroshell, etc. and nothing comes close to this platform that was developed by a single individual, working day and night, who loves every aspect of what he's doing.

Its an open-source environment (many trading systems discussed on this forum and in all the books or posted on the site for people to use for live signals or just backtest). There's a pascal-like language for developing your own systems and most of the major indicators are built in. And, most importantly, you can backtest across thousands of stocks and optimize across lists of stocks - something that is difficult to do in all the big commercial software platforms.

Why am I posting such an enthusiastic review?:
a. I hope to see this guy get what's he deserves for putting in so much quality time and effort.
b. I think the quality will get even better if there is a larger community of talented traders posting systems, exchanging ideas, suggesting new features, etc.

The only forum that I go to other than this one is the wealthlab forum.

-jaltucher (on wealthlab).
 
I certainly think that a good purely mechanical system can be devised but whether it can be done for daytrading I have my doubts. Its like making a map of the coastline of Florida, it depends on what scale you use. If you use one inch equals fifty miles the task is not to difficult. If you use one inch equal one inch, the permutations are practically infinite.
 
Originally posted by andrasnm
Aside from effective bionics (robotics) whose costs itself would negate it's trading usefulness, you currently have no technology off the shelf to do a program like that. Current programs are basket trading versus indices, baskets against other baskets but simply put NO DT program !!! Not for a
while.....
We are in fact quite a while away from it. 'Data' (in startrek could do it). Build me a 'Data' and I will find you capital.
More...

If you are talking about a very complex analysis, such as "discretionary trading," then you are correct. However, if you are saying that automated trading is not possible that is not correct. If you wanted to trade a simple crossover system, a turtle 20-bar channel breakout, an aberration 80-bar BB breakout, or a variety of other systems that have clearly defined rules for entries & exits, you could hook up a computer today and go to the beach while the computer executes trades for you. (Of course, you'll have to account for reliability & availability of the computer, connection to the exchange, etc. etc. -- but I'm only talking about the capacity to do so, not whether you should let it trade unattended).

Now whether that system is profitable or not is a completely different topic, but give me a mechanical system and I can have it trade automatically for it. Do I do it myself? No, and for two reasons.

1) I don't have the appropriate setup to ensure 99.99% reliability of hardware/software/order entry system/network connection etc. I do have contingency plans in place for if something goes awry (such as picking up the phone to call order desk to exit positions).

2) There is definitely joy in trading and joy in pulling the trigger. I have strongly considered doing automated trading where I just sit and watch the computer put in orders, thereby being able to respond in case of unexpected situation (like both dsl lines going down or hardware failure). But it is definitely possible and there are a lot of fund managers and broker-assist programs that do nothing but that.
 
All:

I was reviewing another chat forum based on messages involving Tony Oz. In one of the topics, Tony shares some generic advice as to what he wished he had done initially, but didn't. The first thing on his list was to develop a profitable trading system so that he would at least be following trades that had good mathematical probabilities of success. Other market dynamics can assist in confirming an entry or exit, but the trading system itself seemed to be a requirement. If Tony reads this, perhaps he could confirm or deny what I read on another forum.

Stopped at the library tonight to pick up some books. Picked up the following:
- Long-Term Secret to Short-Term Trading - Larry Williams
- Martin Zweig's Winning on Wall Street
- The Innter Game of Trading - Koppel & Abell
- The Speculative Stragist - Slatyer
- The Day Traders - Millman
- Trader Vic - Methods of a Wall Street Master

Most of these books won't have much to do with Trading Systems, but rather the phychological aspects of managing a trading career and the emotions associated with it.

In my opinion, it is very possible to create a profitable trading system. But the system needs to be adaptive to the point where it adjusts rules based on current market conditions (volatility or average true range, that type of thing). Developing such a system for day trading is next on my list.

Now for one last comment on trading system development. Most people make trading decisions based on a trading system, market conditions, gut feel, or a "monkey see, monkey do" process. I believe there a traits among successful traders than can be captured in an adaptive system. Granted, the system will probably require multiple dimensions to capture all the market dynamics involved in an individuals decision making process but, none the less, should be able to be done. Is this a pipe dream, or is it a reasonable goal?

I'm surprised I haven't heard of big insititutions developing profitable mechanical systems. Perhaps very few exist that can be applied at the individual level.

Keep up the comments,

TAHOE
 
Tahoe,

I'm surprised I haven't heard of big insititutions developing profitable mechanical systems.

And you never will. :) Just finished re-reading Schwager's original Market Wizards book, and while some trade solely on gut (Mark Weinstein comes to mind), many of the big boys make it clear that they developed essentially mechanical systems with the help of engineers, computer wizards, etc. And it took them quite a few years to do it and at great expense. Needless to say, they won't be sharing their hard-earned secrets and methodologies with us peasants. Frankly I don't blame them, I wouldn't either. Would you?
 
Originally posted by Magna
Needless to say, they won't be sharing their hard-earned secrets and methodologies with us peasants. Frankly I don't blame them, I wouldn't either. Would you?
More...

The geeks shall inherit the earth I'm telling you! First it was Bill Gates with software, then it was dotcom kids who cashed out while the cashing was good. But soon it will be the master traders of the world, hopefully a geek like me. :D

Heck, according to my backtesting, with conservative pyramiding techniques I should be trading 1000 S&P contracts in about 32 months. I hope the system is valid from now til then. :p
 
Nobody at my firm trades mechanical set-ups, the market is always changing, and if you can tweak your system according to market conditions, you already know so much about trading that you probably don't need a system to do the job for you.

I wonder why newbies keep on looking for systems, every single trade I do is discretionary, so is 99.9% of the traders at my firm. Stop throwing out those hedge fund examples, they probably don't trade intraday NYSE (at least not extensively), I can guarantee you that tape reading is the most important element on NYSE trading and I don't believe anyone on this board can program that.

Last but not least, at least more than half of us don't even trade with technical indicators, and even if we do it is probably a moving average that we only use half the time, in most cases, the better the trader, the less number of indicators he uses, at least for NYSE, as tape beats everything else and indicators just get in the way.

As for stops, me and a lot of the traders at my firm only use hard stops when we can not watch the position on the tape window. There is no point, the specialist will take your stop in a hurry, and those quarter point losses will destroy you. The specialist can all of a sudden print size at most unexpected prints and a lot of times it is better to buy more than sell. In a fast paced game many times I would sell market way before my stop is triggered because I know the trade is probably dead and I will take some rediculous spread damage if I allow the stock to say, print below the figure. My hard stops are only for emergency purposes, and I don't expect them to be triggered.

I can see people making some sort of system that trades e-mini's or even QQQ's, but for individual stocks intraday? What are you smoking?
 
You must log in or register to reply here.
Back
Top