My assumption so far is that trend following can be geared to look at many different trend lengths which determine how you get affected by noise/reversals (and the resulting losses). Possibly CTA are trying to capture shorter-term trends and got caught up in “chops” at that level – whereas there did not seem to be any chops at a higher timeframe (e.g. MA Golden Cross 200/50).

This post-mortem is probably a good idea for a future post… This is also one of the main reasons why I want to create a State of Trend Following report (I know I keep talking about it!.. It’ll be there, one day, soon). That report would have a few standard trend following strats, run with different trend/parameter lengths. This would hopefully give us more explanation as to why/how/where trend following CTAs produce the returns they do.

I have to admit that I was not following these performances so closely before I started the blog in September last year.
However, I remember that back in November and December I could anticipate correctly (after the facts, ie after the month was over but before I checked performances) how the performance would roughly turn out, for example November had a big upside breakout for Gold and many related markets (and November was widely positive for these funds) while that trend reversed in December (which was pretty negative). Whereas if you look at a wider scale/larger timeframe, Gold has been going up most of the year… (giving away that I am bit of a Gold bug on the discretionary side of things ;-)

@milk – yes, I compiled that list myself based on all the prominent trend followers I have come across and whether I could find their reporting. For actual performance reporting I get the figures from the funds website or AutumnGold CTA database when available (example for Abraham Trading)

I just can’t wrap my head around this…any thoughts?