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

State of Trend Following in December

January 3rd, 2013 · 9 Comments · the State of Trend Following, Trend Following

State of TF
The year 2012 ended more or less as it started: down. Forget the intra-year peaks at +15%, these were all erased in the closing months of the year. A big up-and-down.
Happy New Year to all readers! Hopefully 2013 will bring better times for trend following…

Please check below for more details.

Detailed Results

The figures for the month are:
December return: -01.12%
YTD return: -4.14%

Below is the chart displaying individual system results throughout December:
StateTF December
And in tabular format:

System December Return YTD Return
BBO-20 -3.3% -5.28%
Donchian-20 0.85% 12.42%
MA-10-20 1.79% -0.07%
TMA-10-20-50 2.51% 1.18%
BBO-50 1.38% -3.21%
Donchian-50 1.26% 2.29%
MA-20-50 2.54% 0.06%
TMA-20-50-200 -0.27% -11.86%
BBO-200 -1.46% -9.83%
Donchian-200 -1.14% -6.88%
MA-50-200 -3.66% -22.81%
TMA-50-200-800 -1.99% -5.71%
COMPOSITE -0.12% -4.14%


Composite Index for 2012

Below is the performance of the average of all system/timeframe combinations used in the report for the year 2012:

Unfortunately, the year ends at nearly the lowest point of the last 12 months.

Appendix: System Details

System Rules and Parameters

All the systems were tested with the same simple position sizing rules of 1% per new trade. No other Money/Risk Management rules were used. No trade friction (slippage or commission) was applied. No return on margin is added to the system performance

The system rules are detailed on the Trading Blox online documentation.
The MA Crossover system was used with moving average pairs of 10-20, 20-50 and 50-200 days. The stops/position sizes are set at 2x, 3x and 5x ATR respectively.
The Bollinger Band system is the classic use of the Bollinger Bands with entries taking place at Breakouts. The parameters used were 20, 50 and 200 days with 2 standard deviations.
The Triple moving Average system was used with moving average triplets of 10-20-50, 20-50-200 and 50-200-800 days. The stops/position sizes are set at 2x, 3x and 5x ATR respectively.
The Donchian System is a simple version (with no Trade Direction filter) with channel lengths of 20, 50 and 200 days for entries (and 10, 25, 100 for exit). The stops/position sizes are set at 2x, 3x and 5x ATR respectively.

Portfolio Instruments

Covering over 50 instruments across Equities, Interest Rates, Currencies, Agriculturals, Metals and Energies, from around the world, the portfolio contains the following futures (CSI Symbols): AD, BP, C, CC, CD, CFC, CL2, CT, CU, EBL, EBM, EBS, ED, EOX, ESM, FC, FEI, FFI, GC, HG, ICL, IND, JK2, JP2, JP6, JR2, JRB, JTI, JY, KC, KPO, KTB, LC, LGO, LH, MFX, MP, NG2, RA, RS, S, SB, SF, SI, STW, SXE, TRY, US, W, YM, YTC .
Click here for a tabular view with description and exchange information.

Result Normalization

The system performances are adjusted for volatility to normalize the results. The normalization applied “baselines” the Max drawdown of the systems to a common value, and derive the resulting performance for each system.

A table showing each system performance numbers from 1990 to 2009 can be found on this page. Two extra columns have been added to show the “normalized return” and the multiplier coefficient to obtain this return (the multiplier coefficient is itself calculated by dividing an arbitrary Max Drawdown figure of 25% by the actual system Max DD).

Related Posts with Thumbnails


9 Comments so far ↓

  • Fred

    Your composite index wasn’t much different than the Newedge CTA Index which was down 2.96% last year. I was up 7.6% in 2012 which is 10.56% better than the CTA Index but not what I consider terrific.

  • Michael

    Terrific site and great work, Jez.

    Although trends in markets will never disappear completely, I wonder if the edge has eroded.

    95% of the speculative money entering the managed futures space is trend-following in nature. Entries and exits have to be more costly as a result, right? See commitment of traders data.

    Trending is caused, in part, by the discontinuous acceptance of market pricing, if you will. Bill Eckhardt’s statement that the random walk purists might have the last laugh is not a small one. The rate of information distribution is of course faster than ever and probably increasing exponentially. Does this diminish the value of older data in back testing? Were the Turtles truly just lucky, per your prior post?

    Perhaps we dismiss counter trend too easily?

    Would love to hear your thoughts and good luck in the new year.

  • Jez Liberty

    I believe one of the main reasons for trend following to “work” lies in the behavioral biases of human nature – which does not seem to evolve too much (or too fast) when you look at the last centuries.
    Of course, as you mention, based on market environment investment go through different periods/cycles of performance (and not a great one for trend following at the moment).
    It sure seems like the “Turtle era” was much more rewarding and maybe these sort of numbers will never come back but the fact that the current under-performance and questions being raised about trend following “being dead” have happened before tend to make me think it is just another cycle..
    Just my thoughts. We’ll see what the future brings us!

  • Fred

    In my opinion, trend following and mean reversion/counter trend strategies both work. Of course, we do not know in advance which strategy will work the best in the coming year. Therefore, my approach is to use both types of strategies and perform statistical analysis of closed trades to confirm that each strategy still has a positive expectancy.

  • Michael

    Yes, Kahneman and Tversky’s work is informative and there are natural human biases that could explain trends. However, machines, when programmed to do so, do not have such biases. Look at some of the COT charts. It’s obvious what the net positions are of the large speculators. If almost all of the money is piling in and out based on similar signals, there should be some way of capitalizing on this.

    Fred, I also am stumbling upon your solution. However, what about the problem of negative correlation? Just as a mean reverting strategy might have worked well in the past two years, if sustained trends suddenly return, this strategy will likely underperform if not lose spectacularly, making for mediocre net/net results.

    I have not found regime switching to be of much added value, yet. Any luck with these?

  • Fred

    Michael: The last sentence in my previous post explains what I do in terms of monitoring the two broad types of strategies. You cannot know which strategy will be the best going forward for a given period of time. There may be times when you have to stop trading a particular strategy and there will be a time to start trading that strategy again. My preferred approach is to track all trades for each strategy and perform a statistical analysis of the past X trades to confirm that there is a positive expectancy. If the expectancy is negative, you have to ask yourself why should you continue to trade.

  • Michael


    Thanks for your much-appreciated comments!

    Following expectancy for X number of trades amounts to following the equity curve for that system. How can one know if it is a system that is “broken” or “not working” and deserves to be shelved or non-allocated vs. just a phase of the market enviroment. E.g. trend following systems for the past 2 years have been doing pretty lousy. Would this mean potentially these systems should not be traded until showing a return to profitability? Maybe allocation of funds between systems changes?

    Also, sometimes adding a system with negative expectancy to another system might actually improve the volatility-adjusted returns. Would this “diversification benefit” not be lost?

  • Jez Liberty

    Hi Michael,
    This is the $64,000 question… ;-)
    It depends whether you are referring to specific systems or strategy style in general. If the former, I suppose you could compare the specific system to a benchmark (like trend following in general, mybe using this report – that’s one of the reasons I created it..).
    If the latter, this is a more complex issue I think: is Trend Following dead/broken or simply in a down-performance phase?. I think it is hard to answer but one thing that seems to occur is the negative serial correlation of trend following returns (I cannot pinpoint various studies I have read on this – Schwager on his CTA book I believe is one who mentions it). The corollary is that it is a bad idea to stop trading a trend following system after bad performance as this is when it is more likely to return to best performance. Not an easy thing to do though..

  • Fred

    Deciding whether or not to temporarily stop trading a given system is a personal judgment but I believe it should be based on a statistical analysis of the past X trades. Sometimes a given strategy (the original Turtle’s rules for example) works until it doesn’t. At that point one should determine through walk forward testing if changes in the parameters are warranted or if the system should be canned. You can still monitor the trade signals generated by a strategy after it has been taken offline so you can see if and when it should be used again.
    I kindly suggest you read all you can by Howard Bandy if you have an interest in this.

    You and I have both read about how one would have benefited from investing in a given TF fund when it experienced a serious drawdown. You are wise enough to recognize the hindsight bias evident in this “observation”. The funds that had a serious drawdown and then closed aren’t included in these “observations”. Sometimes a drawdown presents an opportunity. Sometimes a drawdown is the beginning of the end. As far as funds are concerned, it may be very difficult to determine whether a drawdown is an opportunity or a threat. However, with regards to a single strategy used by an investor there are means of determining whether the strategy should continue to be used or, at the very least, temporarily taken offline.

Leave a Comment