Tuesday, March 27, 2007

Toby Crabel And The Epistemology Of Trading Expertise

I recently posted about opening range breakout (ORB) trades and illustrated an intraday example. An early presentation of ORB trading was Toby Crabel's book Day Trading With Short Term Price Patterns. It has become a classic text and, indeed, now fetches a handsome price, given that it is out of print. (Rumor has it that the author, a successful hedge fund manager who worked at one time with Victor Niederhoffer, has not pursued additional printings of the text, believing that it gave away too many of his valuable trading ideas).

I've read the book several times myself and do find it valuable. One of its great strengths is that it is an attempt to statistically test the efficacy of price patterns. Instead of merely asserting that a chart formation is bullish or bearish, Crabel actively searches for evidence. In this empirical approach, Crabel's work shows the influence of Niederhoffer.

(As an aside, allow me to mention that the works of many fine authors owe a debt to Niederhoffer, including his recent work with Laurel Kenner, which includes years' worth of financial columns and insights via the Spec List. The quantitative tradition is now very well established in the trading world; it's difficult--especially for younger traders--to recall that there was a time when that was not the case and when the idea of testing one's trading ideas was quite novel. Sadly, that empirical influence remains something of a novelty even to this day in the popular trading literature).

Less well appreciated is that Crabel's book is explicitly founded on the base of Ayn Rand's epistemology. Ayn Rand was a novelist and philosopher who developed a philosophy (Objectivism) that emphasized reason, political and economic freedom, and a heroic view of human potential. Her book Introduction to Objectivist Epistemology is an attempt to explain how the human mind is able to grasp reality. (Epistemology is the philosophy of knowledge). Central to Rand's account is the role of concept formation. "The ability to regard entities as units is man's distinctive method of cognition," Rand wrote (p. 7). This ability opens the door to both mathematical and conceptual reasoning.

Rand defines a concept as "a mental integration of two or more units which are isolated according to a specific characteristic(s) and united by a specific definition" (p. 11). The formation of concepts requires abstraction--isolating certain attributes from others--but also integration: combining concretes into a larger category. When we form the concept of a "trend", we are isolating certain aspects of price and volume and integrating these on the basis of a definition. Through ever-widening efforts at abstraction and integration, we expand our conceptual universe and extend our grasp of the world.

Crabel understood this, beginning with very simple notions of range and then investigating increasingly complex patterns over individual and multiple days. His abstraction of a "Principle of Contraction/Expansion" enabled him to widen his investigations to a variety of narrow and wide range price formations. He systematically investigated patterns of 2 bars, 3 bars, etc., using the chart to aid in the formulation of market concepts.

Crabel's book is best known for its treatment of ORB and narrow range (NR4, NR7) patterns, but I would argue that his greatest insight was his understanding of the role of epistemology in trading success. The successful trader may indeed trade patterns that appear to be simple. Behind these seemingly simple ideas, however, is a high degree of conceptual integration. Chess grandmasters do not see an assortment of isolated pieces on the board; they see formations that have strategic value. Similarly, skilled physicians don't perceive an array of disconnected complaints; they see interconnected symptoms that lead them to diagnoses of diseases.

Similarly, the successful trader is not mired in perceptual concretes. A good stock pick, such as the one mentioned in my recent post, integrates a wide range of information, from price action to the behavior of institutions, to actual earnings performance and industry trends. A good trade, such as the breakout example from my recent post, also integrates a large amount of data regarding the prior price range, recent and current volume patterns, and behavior during pullbacks. This integration is achieved perceptually, and it is the role of training to effect such perceptual transformation. The expert performer learns to see his or her domain in terms of patterns, whether they're the patterns of a chess opening, a defensive alignment in football, a military strategy, or a breakout from a trading range.

This is the great weakness of most efforts at "trader education". Such education consists of isolated Website posts, magazine articles, and conference presentations. Even books in the field fail to build a conceptual foundation for traders to help them understand *what* to trade and *why*. And, of course, few educational efforts help traders train their "eye" to see the patterns from their conceptual integrations. That, in trading as in chess and medicine, is a process that takes years of devoted effort.

Ayn Rand understood that philosophy is the most practical of disciplines. Without a solid epistemological foundation, what assurance do we have that we're trading anything other than randomness? Proper training for a trader is, at root, an epistemological undertaking. It transforms the knower by expanding the realm of the known. When you become expert, you forever see the world differently. You also think differently, guided by principles, not just percepts. Crabel understood that as few do. That, in itself, is justification for his book's reputation.


John Friedrich said...


You are in the zone! Man, your recent articles have been outstanding. Kudos.


Brett Steenbarger, Ph.D. said...

Many thanks, John. The feedback from readers has been helpful in focusing on topics that are of greatest interest--


AnaTrader said...

Brett, I add to what John said.

This post rang a bell - Crabel -when I first did a search on the web about his unique price patterns (mentioned by my mentor)and idiosyncracies!

Perhaps, I should read Niederhoffer, who is affordable. I am reading Ayn Rand's Atlas Shrugged (voluminous) which will take me a month to finish, as I can only devote an hour a day, more on weekends!

Thank you for your reading list embedded in your posts.

Brett Steenbarger, Ph.D. said...

Hi Anatrader,

I heartily recommend Victor Niederhoffer's writings, including the posts to the Daily Speculations site. There is much creative thought in those resources.


Troy said...

Mr Steenbarger,

Have you ever done any studies in reversal times. I use the them in trading stocks 9, 9:30 are a few.


vcn said...

Wow. Please keep up the insightful posts.

You managed to succinctly state what I've been trying to understand and put together for the last few months . Essentially, I've been breaking down price patterns into basic concepts and attempting to figure out the their (inter)relationships to price.

I would like to mention "Fooled By Randomness" by Nassim Nicholas Taleb , which also lays out a solid philosophical foundation for the author's trading style and philosophy of life. The work provides another view which is in contrast with that of Niederhoffer's.

PS, I'm reading Ayn Rand's "Capitalism: The Unknown Ideal" right now. Adding Niederhoffer to my reading short list.

- v

MarketProfessor said...

Dear Doc,

I have accidentally discovered Victor's book EdSpec and through it your site. Or is it predestined? Like what is said in ALCHEMIST by Paul Coelho. If you desire a thing strongly, the universe will conspire to make you achieve it.
Conceptual foundation is at the root of Trader education. I feel Concepts come first then only the micro rules. Unfortunately popular trading literature and books (Except few like Vic & Doc ) emphasise rules which make the trader confused. High level ideas such as contained in DAILYSEC, TRADERFEED give meals for a life time.

Brett Steenbarger, Ph.D. said...

Hi Troy,

It's an interesting topic, but not one I've ever tried to study. Thanks for the suggestion--


Brett Steenbarger, Ph.D. said...


Thanks for the book recommendation. I think those items on your reading list will prove stimulating.


Brett Steenbarger, Ph.D. said...

Hi Market Professor,

Thanks for the comment. In my latest post, I try to sort out the difference between trading by perception and trading by rules. It's a fascinating topic, and it's fascinating to see how market concepts become embedded in perception and action--


Carl said...

Thank you Brett for devoting your time and energies. These communications are invaluable. You continue to so eloquently emphasize a balance between the quantitative and empirical; ultimately holistic approaches (everything matters all at once).

Rand is an often sourced influence on the recently deceased Robert Anton Wilson. So many of the ideas you present synchronize so readily with his writings. The books Prometheus Rising and Quantum Psychology I recommend as populist syntheses of Rand and other "observer principle" tomes.

Jim Buhrmaster said...

A few observations:

1. Niederhoffer has been using a quantitative approach to financial analysis since his college days. In the mid-60s he published several papers, including at least one on price reversal patterns on the NYSE.

2. Niederhoffer founded the NYC Junto, which in part focuses on the philosophy of Ayn Rand. Also, before Crabel, Monroe Trout worked for Niederhoffer and named Rand Financial after Ayn Rand.

3. I remember marking the Opening Range on hand drawn charts more than 20 years ago (and it certainly wasn't an original idea of mine -- in fact I have nearly always found that my "original" ideas have been predated by someone else's work). I have measured expansion/contraction for so long that I had forgotten that Crabel had written about it. Perhaps Crabel's contribution was to quantify some of the short-term methodologies that had been floating around. In that sense he was a pioneer and unfortunately still is as the quantitative approached is eschewed by most technical traders.

Brett Steenbarger, Ph.D. said...

Thanks, Jim, for the historical background!


Philip said...

It's always easy to talk about patterns and indicators (TICKI) after the fact. If the patterns and related indicators are so neatly laid out and logical each time, then trading would be much easier.

Chuck Lebau back tested every indicator against random entry and found that most if not all have no predictive value. The found that the best traders in the industry are right only 30% to 40% of the time (so much for indicators). They just cut their losses short and have these huge winners. See http://www.amazon.com/Technical-Traders-Computer-Analysis-Futures/dp/customer-reviews/1556234686

I know successful traders who has been trading for decades and they trade nothing but price because they told me that they finally gave up on trying to predict markets with indicators. When they stopped trying to predict with indicators, that's when they started to become successful.

Maybe you should talk about patterns and indictors and then back it up by showing us whether you traded those patterns and indicators. This will be much more credible.

Don't take it wrong. I respect your work and the tremendous amount effort you put in. While I like your trading psychology and learning to trade process work, I have my deep reservations when you are talking about indicators as if they are logical predictive tools. In my humble opinion, you are leading beginning traders astray.

Brett Steenbarger, Ph.D. said...

Hi Philip,

You raise valuable points; thanks for the comment.

LeBeau's research and conclusions re: trading are accurate within the context of his investigations. He was looking at system trading, largely of a trend following nature. In that context, it is true that one can be profitable on far less than 50% of trades and still be profitable. Indeed, the dyed in the wool trend followers eschew any attempts at market prediction and simply trade trends across the board in diversified markets, allowing the expected distribution of outcomes to work to their favor over time.

There are, however, many traders--including some I work with personally and intimately--that rely on pattern recognition (and sometimes empirical tests of those patterns) to make their money. Those traders are right far more often than they're wrong, and I've seen the account statements to prove it.

I'm certainly remiss if my post implied that perceiving and acting upon market patterns is easy. On the contrary, as I try to note in recent posts, it takes years of concentrated practice--and even then only a relative handful of traders can consistently make a solid living from their work. (Not unlike other performance fields, such as golf or music).

It's not the specific indicator that's important: it's the pattern of trading over time and the relationship between price change and sentiment. Other indicators (such as Market Delta) could have illustrated the pattern equally well. If people think they can make money by selling when an indicator is at one level and buying at another, of course they'll fail. The key is tracking the waxing and waning of supply and demand over time. For that, volume and sentiment add important dimensions to price, particularly when traders think in terms of such concepts as "value" (from Market Profile) and how volume behaves as price moves away from volume.


Brett Steenbarger, Ph.D. said...

My bad again; my ADD is acting up! That last sentence in my reply to Philip should end with, "as price moves away from *value*".

More Ritalin please... :-)