Ed,

I benefit greatly from your books, FAQ, Risk Management and Chartbook.  My whole family loves the Whipsaw Song!  I write to you to ask for feedback on my professional development.  First some background.

I recall over 2 years ago being abruptly fired from my middle management corporate job (of 17yrs) in a Dow company.  I experience depression and complete lack of energy in my entire body for 3 days.  Then I start to allow myself to feel the pain of losing the only work family I have known.  I cry in 1:1 meetings and share my feelings about parting with my co-workers.  This confuses my co-workers who have never seen me so open and vulnerable and they are very supportive.  Most important, it helps me move forward.  I close out on that part of my life as well as possible, doing more training than asked because I want my employees to experience great success.  I still sigh thinking about this loss but I realize now that the ending of this part of my life was a gift. 

During that old corporate life and especially after feeling the confusion and fear in the tech bust of early 00s, my primary hobby becomes trading.  At first I want to figure out the “why” to medicate my fear, but then I stumble across Covel’s Trend Following book, your work, etc and I begin to understand that following the trend is more productive.  In the last 15 years I read >200 trading books and use my excel finance background to help model and trade many different systems I find or develop through my growing understanding.  I read about everything I can find which seems to hold good bliss potential and then run small excel tests on the most promising.  I avoid the hucksters that show high returns and hide drawdowns and the hucksters that show back tested data without actual trading (unfortunately most info out there is junk and seems to fall into one of these 2 categories).  Through this slow study I educate myself to the point where my returns/risk is at parity or slightly better than recent stock market performance when fired.

My working life now is a “barbell” in Nassim Taleb’s Antifragile definition of having a mix of low and high risk business interests.  I have a small on line business I run (~20hrs/wk) for which I usually enjoy the process but have little interest in growing it.  It pays the bills and keeps me calm about the larger risks I take in trading.  (Or maybe this is my security blanket for slowly stepping outside of my old wage-slave life).  In any event I feel my purpose is trading.  Updating my system and trading takes <1hr/day but I spend 40-50hrs/wk researching and testing trading systems.  I regularly experience a sense of flow during my research.  I feel drawn to the work, and I get a little antsy and irritable if I haven’t been able to work on trading research lately.  My approach is to find new trend following systems to “fit into” a larger system so that bliss improves in that larger system.  (I define bliss as CAGR over both % drawdown and length of time of drawdown.)

Currently I have 17 systems and 2 “setups”.  Without going into details the names of the systems are: Global ETF, MOMO ATH stocks, VX, Bitcoin, Alt Coin, LT Bond Futures, Tropical Futures, Metal Futures, Nat Gas Futures, Nat Gas Curve, Crude Curve, Wheat Curve, Lumber Curve, Coal Stocks, POT (stock), RYN (stock), Seasonal Futures.  All systems are trend following with the possible exception of the last, Seasonal Futures, which could be considered trend following if looked at based on prior seasons or trend reversal if looked at over the course of the current year.  The 2 setups are trend reversals only to be traded in very specific post-crash conditions by purchasing options with small % risk: Stock Reverse, Bond Reverse.  However even these post crash systems could be thought of as “short-term trend following” instead of reversals.  One asset that is absent is FX.  Though I have tried many times I have been unable to fit in any FX systems because of the high correlation to existing futures return and the lower bliss function I generate in my FX systems.  This doesn’t feel right and I keep coming back to it.

I design my systems to stay out of the market as much as possible.  I measure the market or sub market for trends as a filter for many of the systems.  As a result there is no time when they are all “on”.  Right now 10 systems have active trades (some with multiple positions like MOMO ATH and Seasonal Futures) which is fairly typical.  Most systems have risk stops as % of portfolio values, some systems use time stops, volatility stops and other stops as well.  I have position sizing in some of the systems which factor positions by up to 10x from initial purchase.  Other systems maintain constant position size.  I choose all these system characteristics by trial and error in Excel optimizing for total system bliss.  Using Excel has many advantages for me: I have >20yrs experience in it so I can “think” spreadsheet, and it allows a great deal of flexibility in what can be measured.  But in designing each system individually using Excel (using individual spreadsheets for each system because of size/memory constraints) I feel limited because I am optimizing that individual system without much knowledge about how it will impact the larger system until I fit it in at the end (very time consuming).  A friend tells me I should learn a program like Python and do my research and back testing in Python.  I think he is right but I am not sure how important this is and how I should prioritize this.

My goal is to develop myself and my trading to achieve the highest return possible for 25% or lower drawdown over 6 months or less time.  While I have had offers of money management arrangements from family and friends I have no plans to manage other’s money or commercialize my trading for 3 reasons: I enjoy the flexibility of my time, I am sensitive to the fears of others, and do not enjoy the sales cycle.  This year my trading accounts are up 58% with 17% max drawdown.  I’m pleased with this result but I see a lot of room for improvement also as this year has had great trends in crypto currencies that my system has taken advantage of for much of the return.  While not all of the subsystems I currently have were in place; in 2016 I experience 4% return and in 2015 25% return.  However both of those years had larger drawdowns.  Also I back test out of sample 1990s data (using as many systems as possible/practical – bitcoin for example could not be replicated and MOMO ATH stocks in Excel is too time consuming).  This back test on ~70% of systems generates >50% returns and 20% max drawdown.  This is in line with what I expect because while my in sample system back test yields >100% CAGR and 15% drawdown, the drawdown is low because optimizing systems by its nature creates curve-fitting and the CAGR is high due to the incredible rise of crypto currencies.  The net result is that I’m getting good actual returns that are coming close to my out of sample testing.  Maybe this is the time for me to learn a programming language.  I feel a tightness in my neck as I think about learning a language.  And I know that by continuing to pursue system optimizations in Excel and possibly add new negatively correlated systems I will definitely be able to improve, but the programming language value is an unknown.  

Can you please advise me on where I might see the most benefit?  Should I focus more on learning a programming language like Python or continue to work with what I know in Excel and make incremental improvements?  Will the language knowledge possibly allow for breakthrough thinking/understanding of how my systems operate that I won’t be able to get staying with Excel?