Dark Pools - A Book Review

I read a lot of books on finance and investment theories, etc.  And, at the risk of sounding arrogant, mostly I don't learn a whole lot from them that's new.  After doing something for 30 years, you get to the point, I guess, where there isn't a whole lot to learn.

Well, last month I was in Anchorage, Alaska visiting my first grand- child, little Nora Emily; and my son and I visited the Anchorage library as the women went off with Nora to see the midwife for Nora's two-week checkup.

First, let me say that Anchorage has an excellent library.  They are putting that pipeline money to good use.

As always, I gravitated to the new book section and lasered in on the investment section.  There I came across Dark Pools.  I hadn't intended checking anything out; but, after sitting in one of their cushy chairs and perusing a couple of pages, I saw that this book had a lot to teach me on the important topic of high frequency trading.

The author, Scott Patterson, takes you from the beginnings of high frequency trading, day trading, and machine trading and describes the people who were instrumental in building the key constructs.  He describes how algorithms are able to hide trades until the last minute and then jump in front of the line when there are sufficient limit orders (i.e., depth in the market) to ramp up the probability of achieving a gain while limiting the possibility of a loss.  You thought that because you were the first one to put in a order to buy Intel at $20.00 that you get first dibs when a trade goes off at that price?  Think again.

The pioneers in this market did accomplish a lot of good.  They hastened the demise of the trading in eights system that blatantly ripped off small investors.  They led to a system where, today, trades are executed fast and efficiently.  They forced lower brokerage commissions.

Still, many of the principals admit they created a monster and that steps need to be taken to regain control.  In fact, at this late date, there appears to be no definitive answer to what caused the May 6, 2010 flash crash and knowledgeable people in the book argue that it easily could happen again, in even a bigger way - a pleasant thought for Joe and Daisy retiree trying to live out their days in calm comfort, not to mention for investment advisors in Glenelg, Maryland!

The last part of the book will be of special interest to some readers.  It describes Alex Fleiss's work in creating a system called Star  - an attempt to create a digital Warren Buffett.  It describes how Star, an artificial intelligence system, analyzes massive amounts of data, computes their correlations, and determines what to buy and whether to be in the market.  If a falling Euro, an increase in the price of corn, and a drop in mortgage rates correlates with rising utility sector prices, Star will find it in short order.  It is a system that learns as it goes and discards trading approaches that don't work and keeps those that do.

Star started buying up stocks including financials in a big way in early 2009 to the dismay of Fleiss who felt, along with much of the investment community, that the world was coming to an end.

As we know, of course, Star turned out to be a Star (sorry...couldn't resist that) as markets roared upwards after March 2009.

All in all, I found Dark Pools to be a fascinating read, and I feel like I learned a lot about a subject I need to understand.

Here is a recent interview on high frequency trading with Dave Cummings, owner of TradeBot Systems, one of the people profiled in the book:

http://finance.yahoo.com/video/cnbc-29093777/the-business-of-high-frequency-trading-31129268.html