End of one (strategy) era, the beginning of another

I've been running a self-rolled "alt-risk" strategy that can roughly be labeled something like "multi-asset credit allocation trend following using a systematic rules based approach" for more than seven years.  It actually dates to 2010, if not before, but the quality of my data before 2012 is suspect. I'd say I reliably have about 78 months of coherent data.  For a variety of reasons I am now closing the strategy. In practice it is, in fact, just being rolled into a broader strategy that is less data and time intensive (hey, I'm retired and there is a strong case for simplifying). But before I throw the last shovelful of dirt on the grave of my strategy I wanted to memorialize it for a second. Ignore that none of the strategy's timeframe overlaps any major financial crises.

Here is the strategy (green) up against two private, well known managed futures (MF) CTAs and one large AUM MF mutual fund doing vaguely similar trend following and where I also had material amounts of capital at stake over the span of Q1 2012 - Q2 2018. This is not a perfectly fair comparison but then again it kinda is....these guys get paid a ton to perform a particular task that I expect them to perform well. All of them play more or less the same role in my meta-portfolio. Displacement theory.

These are some thumbmail topline metrics:

   - CAGR over 78 months = 6.2% (6.5 from early 2011)
   - Std dev annualized from a monthly series = 3.8%
   - Sharpe: do your own based on your guess on Rf over that period, but I'd say: not bad
   - Fees = 0 but I estimate trading costs at 20-30 bps max but more like < 10-15 lately
   - Overhead: subjective but my time is ~free and I pay for some software but build the rest
   - Leverage has ranged from .5 to 1.5. 90% of the time it is less than 1.

This is the strategy plotted over backward-looking mean-var space over a slightly shorter time (higher vol and return vs above; I need to double check that I am apples to apples here which I'm not really since sd is calculated using a Morningstar method that will bias higher...oh, and returns are linear projections not multiplicative) but basically it's same basic idea in broad strokes. The 5-asset reference categories are spy efa agg iyr gld. The 2-asset reference is spy and agg.  The strategy I've been running is the green dot. The 2D 5-asset allocation space is based on a sampling technique. Co-var matrix is based on the history. There is a lot of imperfect measurement and assumptions in there and probably a few errors but so be it...

Conclusion

This was an interesting and worthwhile endeavor while it lasted.  On the one hand I think it was distinctly accretive to my meta portfolio by being additive to efficiency especially for the particular goals of that portfolio given then-current age(s), planning horizon(s), consumption rate(s), and purpose. On the other hand it often felt like riding a bike in first gear: a ton of heat, motion and effort in exchange for small amounts of forward progress.  It was efficient but not wildly remunerative...and it required a uniquely anal amount of detailed work-effort in terms of trading, data management, analysis, taxes and continuous strategy improvement. As one of my anonymous partially-helpful readers said once: "maybe you're trying too hard."

But, I do think that there is a case for paying a little bit to someone to do something like this for you instead of self-rolling it. For a rationale on this there are a ton of sources. Just follow the fin-twit crowd. Or read this recent piece from Thinknewfound (The New Glidepath). Or read pretty much anything from Mark Rzepczynski at Disciplined Systematic Global Macro Views.  There are plenty of others.  I will certainly continue something like this myself but not so obsessively as before and I will also continue to keep an eye out for someone to whom I can outsource this kind of thing, someone that will, net of their fees, do at least as well or better than my pared down effort and someone who does not have outrageous minimums or dirty and unappealing lockups.  Trust matters as well. We'll see.