Systematic Trading research and development, with a flavour of Trend Following
Au.Tra.Sy blog – Automated trading System header image 2

Dear Mr Eugene Fama, are you kidding?…

October 14th, 2010 · 18 Comments · Trend Following

Or maybe in serious state of denial…

In this “open question” to Fama (and his sidekick French), I am referring to this recent entry on their forum:


“Market Timing with Moving Averages: An ancient tale with no empirical support”
There is only a short bridge to cross to extend this quote; and interpret what Fama is saying as “Trend Following does not work”.

Now, if I were dubbed as the father of the Efficient Market Hypothesis, as Eugene Fama is often described, and if my career were based on this theory, saying anything other than the above would probably equate to an academic hara-kiri. This slightly puts things in perspective about how objective this “insight” really is.

This reminds me of a quote by “very smart” quant trader Vic Niederhoffer (wait, remind me: how many times did he blow up?):

Trend Following is one of the stock market’s biggest con, I would even go so far to compare the concept to a cult like scientology… I get the same sort of value from Trend Following as I do from supernatural operators such as Uri Geller and horoscope readers.

On the other hand, here is a very much à propos quote from Jerry Parker, one of the most successful Turtle Traders and Trend Following Wizards:

The reason we make money? It’s the simple moving average systems. Two-thirds of that is what drives our profits. Our little filters to get in early, to get out quicker, volatility filters, if that is how we’re going to essentially generate returns, we’re going to be in bad shape. The core simple moving average or breakout systems [are key].
from The Complete Turtle Trader

For recap, here is a long-term track record from Chesapeake Capital, run by Jerry Parker (taken from this historical wizards performance data post):

Jerry Parker's performance
One single year in the red since 1988…

History is littered with “men of science” hanging on to their wrong theories, and there is no reason for this to have changed: last I checked nobody had discovered THE universal truth and knowledge…

I don’t know about you, but between an academic defending their theories with rhetoric and a successful practitioner, I know which one I am choosing to pay attention to.

Of course, Jerry Parker is not an isolated case and there are plenty of successful Trend Following Wizards using concepts similar to moving average market timing.

There is still the issue of survivorship bias when looking today at managers successful in the past (i.e. how many failed along the way, I do not know).
However there are many studies comparing buy and hold with moving average strategies, such as the Golden Cross. I’m not going to publish yet another one of these studies showing the risk-reward superiority of moving average market timing vs. buy and hold; but they are aplenty on the web (Mebane Faber’s timing paper/system, Blackststar fund’s Stock Trend Following study, or Mike Stokes’ post to name a few). Most of the better studies test for different lengths of moving averages to ensure that the concept works robustly across various timeframes, rather than with isolated “data-mined” parameter values.

How is that for empirical support?

One area where Trend Followers and Fama surprisingly seem to agree on is fat-tailed distributions (cf. Fama’s work on stable paretian distributions in market price returns – notably with B. Mandelbrot). I say “surprising” because I believe that fat-tailed distributions are one of the main source of Trend Following profitability (as per this post), and therefore a main reason for moving averages to work…

To me, this is plenty of evidence supporting the use of Moving Averages (or equivalent). I’d love to hear what you think about all these, Messrs Fama and French?

Related Posts with Thumbnails

Tags: ··

18 Comments so far ↓

  • Pretorian

    Regarding Eugene Fama, I think that if somebody has any useful idea of how the market works, he doesn’t become an academic, he uses the knowledge to make money. I don’t understand the EMH guys, I think not even one of his assumptions and conclusions is correct, the theory is so absurdly wrong that to me is like dicussing if Alchemy works. In any case, as Warren Buffet said, I would love for this theory to keep being mainstream so I can make money out of its believers. If Fama aknowledges that Mandelbroit has merit, plus recent discoveries in behavioral economics he should reconognize that his theory is absolute crap.

    Regarding Parker and moving averages, the way I see it, the MA system (complete with filters, stops and so on) gives you the edge, whereas positions sizing makes you survive so I don’t think MA (or breakouts) in themselves are such a big part of profitability.

    However, having a look at the MA systems backtests that I have made, even the most naive and/or risky outperform buy and hold by a mile.

  • Motomoto

    While I am 100% with you Jez, this is always the problems with backtests as we all know there are issues with data, assumptions, etc; etc; My guess is that they probably make enough assumptions that are out of touch with the real world.
    I remember being told at university —–
    that arbitrage opps should not exist….. that was 20 years ago, and they still do.
    that the once in a 100 year event is just that, problem these seem to occur every two-three years. LTCM, Asian meltdown, Internet bubble, US housing bubble, commodities rally.
    There is theory and then there is reality and if the latest thoughts from the cosmologists are right, there are multi universes in reality with just as many theories….. so lets just deal with the real world as we know it. :)
    As they say – those that say it cant be done should not interrupt those who are doing it.

  • Friday links: bullet bonds Abnormal Returns

    […] Trend following and Gene Fama’s blind spot.  (Au.Tra.Sy blog) […]

  • Leonard Golub

    Hmmm. I would say Dimensional Fund Advisors, with over $200 billion under management, Fama a major shareholder and principal, goes beyond academic theory and into practice. But you may disagree.

  • Jez Liberty

    Leonard – I stand corrected, I looked over that fact and I agree this takes him beyond academia..

    I still believe he’s an academic first though (and this is how he is described on the Dimensional website as well, on this page).

    The cynic in me would think that he is probably “talking his book”. An even more cynic one would liken his position in Dimensional Fund to that of Robert Merton and Myron Scholes in LTCM (added “genius” on the fund roster) – oh wait, I see that Merton and Scholes are also involved with Dimensional… ;-)

  • Nick Iversen

    This report concludes “Over 5,000 popular technical trading rules are not consistently profitable in the 49 country indices that comprise the Morgan Stanley Capital Index once data snooping bias is accounted for … trading rules do not add value beyond what may be expected by chance when used in isolation”

    Looks like the apparent but not real success of moving averages is due to data snooping and chance.

    This paper and the ones it refers to are definitely worth a read if you think moving averages work.

  • Severus

    Great post !

  • Jez Liberty

    cheers Michael.
    btw, you might want to check your site: it’s quite often that I get a “connection timeout” (“The server at is taking too long to respond”) at weekends (like now), when all other sites seem fine. Not sure if it’s because I’m in London…

  • Michael Covel

    The only connection complaints I ever hear about are London-based. We can’t figure it out. You are not the first.

  • Pretorian

    Nick, moving averages (or any other technical based indicator for that matter) don´t work in isolation, they work in the context of a complete systems that includes crucially A DIVERSIFIED PORTFOLIO OF ASSETS. Trend following systems need different assets, so it is easier for them to find trends as often as possible, it will be impossible for them to work with indices (I assume you as talking about stock market indices) because they are all correlated. I suggest you read Michael Covel´s book to get an undertanding of the concept.

  • Nick Iversen

    Pretorian I agree with you. I regard technical analysis as an anchor (in the Kahnneman & Tversky sense) starting from which you can evaluate what’s going on in the markets. That’s why I’d never evaluate a technical rule in isolation as academic studies tend to.

    HOWEVER, most responses to The Fama statement (not on this site) seem to be saying that even in isolation technical rules do work and those arguments don’t really stand up in light of those academic studies.

  • prazor

    Jez – Good post!

    Don’t know how many research papers I’ve read
    that is littered with the emh nonsense.

    Even if the papers are analysing some proven anomaly in the market, they still devote a quarter to half of the paper to ramble about the
    definitions of the emh and making hypothesis
    (H0 and H1 etc).

    Emh is the definition of an inane question.

    Why not add, emh as THE best example of an inane question, to all dictionaries – starting with wikipedia… ;)

    Legitimate and serious researchers should ask other questions.

    Prof. Thorp put it elegantly:

    The question isn’t “Is the market efficient?” but rather “How inefficient is the market?” and “How
    can we exploit this?” – Edward Thorp.

  • prazor

    Would you pick up the $100 bill?

    One emh-economist is strolling down the street with a companion. They come upon a $100 bill lying on the ground, and as the
    companion reaches down to pick it up, the economist says, “Don’t bother – if it were a
    genuine $100 bill, someone would have already picked it up”.

  • proud_


    “I don’t understand the EMH guys, I think not even one of his assumptions and conclusions is correct, the theory is so absurdly wrong that to me is like dicussing if Alchemy works”

    You are so far from understanding it (like the original poster himself). The main concept of EMH is absolutely right (though I would question some of its details, but it is mostly right).

    Let`s assume that you are right, and there exists a strategy that beats the market based on the historical data. This is then a trading opportunity leading for extra profit. If traders, who are constantly searching for trading opportunities/strategies, paying millions of dollars to research/learning, find this trading opportunity, than it has to disappear. As people trade with a winning strategy, than the stock price incorporates that strategy, too, until you can not trade with it for extra profit. It`s really simple, I can`t believe how people are not understanding it (plus it has proving data).

    So, if such strategies exist even for a limited time, the cost to find it is really high (fastness, computatinal power), and then the opportunity fades, leading to an efficient market.


    You are also wrong.

    If every dropped $100 bill has a number on it that represents how many people picked it up and checked its validity (like simple and very complex strategies have empirical tests), and the bills with 0 on it means an 50% winning strategy, than an EMH economist would say, that 1) don`t bother picking up a bill with a number 1000 on it, and 2) you have 0.0001% chance of finding a valid bill (with 0 on it) on the streets, as millions are searching for it every day.

  • proud_


    Let`s assume that you are 100% sure, that the price of a stock will rise tomorrow. If you are not the owner or a worker of the firm itself, the probability that other people also know that tomorrow will rise the price is very high.

    Now why would you wait for tomorrow to buy? You would buy it right know. And the others think the same, so if a group of people have information about a future price, than that information moves the price almost instantly.

    The only chance you have to beat the market is when you have insider information, that is not reflected by the stock price yet.

    P.s.: a trend following strategy indicator is also an information that says: the price will be here in the future. As the trend following strategy is not used only by you, the price reflects all of the trend following strategies, removing the extra profit from them.

  • proud_

    News overreaction is not an anomaly itself.

    In the first half of the overreaction, traders do not have information about the target price, just about the direction. Why? Because they do not have time to evaulate the effect of the news event, hence the strategy is faster, but shifts the price too much.

    The second half is the evaulation section. Slower strategies can calculate the correct price after the event, but are slower. So a correction follows the overreaction phase.

    You can not exploit this knowledge of overreaction. Why?

    Because you have to be really fast on deciding that an event is good/bad (which can be done fast by computers, too), but in this phase you don`t know the correct price where it will be standing after all this (which price is over the target price and which price is not), or you can calculate the correct target price (which can be done a little slower, but also really fast), but can use this information only after you actually have that calulation result. You probably can`t beat any group of people (machines): event direction + target price calculators.

    Do this little test: measure the speed you react to a two state event (e.g.: flashing red or green light), and a calculation task (actual price, new price – new price is higher or lower?). After a number of tests, you can`t beat yourself in any of the two phases, that represents the fact that millions of people try to beat each other on these phases in news event trading, and that proves EMH right.

  • “Market Timming with Moving Averages: an Anncient Tale with no Emperical Support”… -Keneth French/Eugene Famma, Oct. 11, 2010 « globalmomentum

    […] on Automated Trading System while researching back-testing software (tired of Excel I guess).   Dear Mr. Eugene Fama, are you kidding?…  As I continued to read the post I almost couldn’t believe it, denial is a powerful thing, […]

Leave a Comment