Systematic Trading Diversification Done Right

Jason Russell and Nicholas Markos of Acorn Global Investments Inc. offer good insights into systematic trend following trading (PDF).

19 thoughts on “Systematic Trading Diversification Done Right

  1. Mike,

    This was an excellent post!

    Do you have the monthly performance data for the Barclay Systemic Trader index?

    Thank you,
    Robert

  2. I agree that the portfolio diversification argument is probably a good way to market an Alternative Investement fund to HNWI or fund of funds

    However systematic trading as a whole is bound to exhibit smoother equity curves (compared to Trend Following only) because of the composition of the index will contain Trend Followers as well as other (possibly opposite) strategies such as Mean Reversion, option writers, etc. (especially when measured via an index/benchmark with survivorship bias…)

  3. Very good post. Points out the “buy and pray” versus any systematic trading method including trendfollowing.

  4. Here is a thought: I have noticed over the years that in using the term “Trend Following” something very important is left uncommunicated. Trend Following is only the essence of the activity but what actually “exists” and what brings home the bacon is ones ability to stick to the system. This reminds of the existential dictum “existence precedes essence”. So, “systematic” is pre-trend following.

    An overarching attachment to the concept of Trend Following may interfere with execution of a system.

  5. Nice article and charts. I believe the Barclay’s chart and many other charts trying to highlight the long term success of trendfollowing fails to take into consideration survivorship. I am not disputing the benefits of systematic trading, but what about the funds that ‘blow up’. The street is littered with CTA’s and other funds that closed shop due to poor performance. Some of these funds even have performance that falls of a cliff before closing shop or having investors demand their money back (if in fact any is left). I do not mean to disparage Jason’s research, but to keep the playing field even all facts must be disclosed. Including survivorship would have a material impact on the cummulative returns.

  6. I know many traders who get thrown in the mix who were incompetent, not trading the style they were advertising (ex) or were not even systematic trend followers. Most of the aggregate stats don’t take into account individual detail. The phrase “CTA’s and other funds” can mean anything. The conversation is systematic trend following — not everything you might find in the kitchen sink. Bottom line, to me, if the aggregate stats don’t reveal exactly why some trader may have hit the skid, and considering I am aware of much behind the scenes stuff where bad traders work overdrive to appear good, just yelling ‘survivors’ — falls short.

  7. You always defend trendfollowing to the hilt, but never seem to acknowledge statistical evidence that may skew the biblical numbers you preach. Ask Jason Russell if he ever ‘jumps’ his system. If so and I can assure he has does that disqualify him from your definition of systematic trend follower? I can assure your there is a graveyard full of systematic trend followers that have been taken out of the market in a proverbial pine box. Rather than focus on the semantics of survivorship how about admitting to your TF brethren that data they are looking at may be skewed.

  8. You are right — I have done a little bit of trend following research and writing. So when someone walks into the room yelling fire, as if that somehow invalidates the strategy — I will point out errors. Look, the idea is to learn from people who have done it right, not from pretenders who made all the wrong steps. Can we learn from the bad ones? Sure, but those are different lessons. Bottom line, pretending the bad traders, who may or may not have even been trading the same, were on par with the good traders, and that the bad ones were just unlucky, makes little sense to me.

  9. I honestly don’t understand the problem some people have with TF (?). If ALL you take from it is to limit your losses and let your profits ride, you WILL come out ahead…unless you absolutely, positiviely suck as a trader.

    I don’t recommend it, but if ALL you did was say, “As soon as any losses hit 8% I will sell”, how can you NOT make money?

    I and a lot of others like me have buys and shorts that return as much as 900% over the life of the trade. If you have even one of these, that covers a lot of small losses.

    Bottom line: If you TRULY follow TF, you’ll do well.

  10. Re: Survivorship Bias.

    The statistics are just a starting point. TF is best understood by “seeing” the process, by seeing what doesn’t change, not by analyzing the results statistically. One has to remember there is a process, hidden from view, that generates those numbers.

    Those that put too much weight on survivorship bias do not understand process thinking. What matters is not “survivorship bias” but “how” they got there!

    What Michael is doing is pointing at such a process.

  11. Following on Trender’s thought…

    Let’s take one of the those so-called ‘biblical’ trend followers…Bill Dunn.

    I want to learn from him. Why has he excelled? What did he do for so long? Did he buy 5 stocks and sit on them for 40 years? Not exactly. His track record is many, many trades over all kinds of markets.

    Now, for example, do we include two failed Turtles in with Dunn’s track record even though anyone with a pulse can research those two Turtles and see that they don’t deserve to be in the same conversation as Dunn?

    I can go on…

    And of course Dunn ain’t the only one…so what can we learn from the winners and what can we learn from the losers? Assuming they are created equal is just that — an assumption.

  12. I guess the reason so many harp on “survivorship bias” is because it is so much easier to find great examples of it than a clear process.

  13. Zircon, you can argue suvivorship bias for any strategy, so really, what is your point? Ultimately survivorship bias has more to do with the practioner than the strategy.

  14. Regarding CTA indexes & Survivorship – I believe the Stark CTA indexes account for survivorship by taking them out of the index & recalculating.

Comments are closed.