If You Can’t Beat ‘Em, Join ‘Em
by Marc Lichtenfeld, Investment U Senior Analyst
Wednesday, September 7, 2011: Issue #1595
Computers conducting high frequency trades make roughly 70 percent of stock trades on all U.S. equity markets.
Every day, supercomputers – owned by hedge funds – buy and sell millions of shares of stock, and sometimes the holding periods are as short as milliseconds. The goal of this sort of trading is to take advantage of market inefficiencies and scalp profits all day, every day.
High-frequency traders generate approximately $21 billion in profits annually for the funds and investors that own the computers.
And high frequency trading is only going to increase. While regulators pay lip service to leveling the playing field for the little guy, in reality, the supercomputers are going to have even more of an impact going forward.
The New Center of the Stock Market Universe
In the future, the center of the universe when it comes to the markets will no longer be on the corner of Wall and Broad Streets in lower Manhattan. Instead, it’ll be a red brick building in Jersey City. That’s where a new facility is being built for the New York Stock Exchange. It will house several football fields worth of computer servers.
Funds will be able to lease space from the Exchange in order to place their supercomputers in the same building as the Exchange’s computers, taking precious milliseconds off of the speed of their high-frequency trades.
The average investor would be a fool to try to compete against the institutions in this style of trading.
But there is good news. Super-computing technology is becoming more readily available to the everyday investor.
Last week, I wrote a column about inventory turnover and how using one piece of data could help you beat the market by more than 21 percent. It literally took me longer to type in my instructions than it did for the computer program I was using to crunch the data and arrive at a conclusion.
And that was analyzing every publicly traded company in the United States.
Successful Stock Analysis
To be a successful stock analyst today requires more and more computing power.
On my laptop, aside from a lightning-fast internet connection, is a trading platform that gives me real-time order flows, charting, news and every other type of information that I could possibly need about the market or a particular stock.
I also have a program, linked to a supercomputer, which can crunch millions of pieces of data on thousands of companies in just seconds.
As a 15-year stock market veteran and a former NASD licensed analyst, I have a pretty strong idea about what makes a stock a winner. For example, I’m much more partial to cash flow than I am to earnings. Net income can be manipulated much more easily than cash flow.
I look at balance sheet items like inventory and accounts receivables to search for signs that fundamentals may be deteriorating or improving, before they show up in the revenue or profits figures.
But through a super-computing program that I call S.T.A.R.S., I was able to test my theories. The computer could tell me, in just seconds, which variables were positively impacting performance and which were not.
As I mentioned last week, inventory turnover was an important component for success. Surprisingly, certain technical indicators that I had relied on in the past were not.
Through the process of adding and subtracting a very wide range of fundamental and technical variables, I was able to create a reliable system that, in a 10-year backtest, outperformed the S&P 500 by 1,568 percent.
It also beat the market nine out of 10 years. What I liked most about the system was that during markets that were bottoming, it suggested lots of stocks, ensuring plenty of exposure to the market. However, when markets were topping or about to sink, like in early 2008, the system only recommended a handful of stocks, limiting exposure to the market. Even then it was able to pick winning stocks like Alaska Air (NYSE: ALK).
If you’re interested, you can learn more about S.T.A.R.S. and my new service, the Oxford Systems Trader.
But the purpose of this column isn’t to be a commercial for the Oxford Systems Trader. It’s to emphasize that if you’re going to do your own research, you need powerful tools…
Three Ways to Speed Up Your Trading in the Digital Age
Here are three ways you can ensure you’re up to speed in this digital age.
- If you’re a fairly active trader, or even just watch the markets during the day, get your broker’s latest trading platform. It’s usually free and will give you a great deal of depth into what’s happening in the markets and your stocks that day.
- If looking for your own stocks, use a stock screener with lots of variables. My favorite is from Morningstar.
- If you’re really committed to testing out your theories, get your hands on a program that allows you to backtest. Many brokers also provide this type of software for free. If not, there are commercial packages you can purchase from software vendors.
Whatever your style of investing, if you’re not harnessing the power of technology, you’re being left way behind. And chances are your investing returns confirm that statement.
The high frequency trading that’s conducted by hedge funds and their supercomputers are out of reach for most investors.
But thanks to improving technology, average investors can generate large profits that widely beat the market.
Good investing,
Marc Lichtenfeld
If You Can't Beat 'Em, Join 'Em,Any investment contains risk. Please see our disclaimer.
7 Responses to “If You Can’t Beat ‘Em, Join ‘Em”
Comments
By submitting your comment you agree to adhere to our Comment Policy and Privacy Policy.



Marc is a senior analyst at Investment U. His investment career started out at the trading desk of Carlin Equities in San Francisco, CA, where he executed dozens of trades each day for his clients.
IF YOU ARE INVESTING FOR DIVIDEND INCOME, YOU WOULD BE HOLDING FOR LONG TERM. WOULD SUPER COMPUTING BE USEFUL? WOULD IT BE NECESSARY TO SPEND MORE TIME AT THE COMPUTER? I REALLY APPRECIATE AND RECCOMEND THE OXFORD CLUB PORTFOLIOS. THANKYOU VERY MUCH. AL M.
Reply
Marc
Good article Why don’t you give us some software recommendations
Charlie
Reply
thanks Marc, insightful.
Reply
Hey marc, you old hippy!! Anyway, I appreciate your insight, and brilliance, and all the hard work you and your team put into the s.t.a.r.s. system. I’ve picked up your last pick and am waiting for the next. I believe as you do, that this system is going to kick some tail. Thank you again and just keep em coming. Yours in privacy, prosperity, and tradition. Steve F. E. Moline IL.
Reply
Right on, Steve (or should we say, Rainbow?).
Thanks for the kind words everyone….
As for Charlie’s question about software recommendations, many of the ones that Marc has seen aren’t very user friendly, so he was hesitant to mention them by name.
The ones that brokers provide are usually pretty good. For example, Marc really likes the offerings from TD Ameritrade. He’s also heard good things about Fidelity’s as well, although he said that he didn’t have any experience with it.
And in the article Marc mentioned the Morningstar stock screener, although that doesn’t allow you to backtest.
Hope that helps,
Investment U
Reply
I also would like to know any recommendations that Marc has for backtesting software since he brought up the subject.
Reply
I only see two major options for the retail investor as to backtesting: Zacks Research Wizard and Portfolio123.
Zacks, to the best of my memory, was somewhere between 200 and 300 bucks a month – and I’m not sure if that was the advanced version or not. As of last year it had survivorship issues and I wasn’t impressed with the robustness checks.
Portfolio123, on the other hand is $50 a month (they have a fee limited version), no survivorship bias, sophisticated portfolio tools, back-testing, robustness checks… basically an institutional grade piece of software for 1/20th the price. It takes a bit to learn how to program – unless you use the point and click menus – but thats kinda limiting.
You get a 30 day free trial but if you punch in the code HKURTIS you’ll get 45 days to see what I mean.
Reply