On Failing For Success and Knowing When To Fold ‘Em
(Written To HZP):
I shut down my Process Driven Trading CFD project which was supposed to start in August/September for realsies.
First and foremost was that fortunately there was no long signal from the system, and as equities looked bleak and I revised my macro thinking more and more, the system I had created started to look like a break-even proposition at best. Mainly because the whole focus on CFDs was completely wrong. The system I was testing had about a 5%-8% forward looking alpha (returns in excess of market returns) in my estimate. CFDs have hidden and not-so-hidden costs that work out to about 5% p.a. to trade, but offer leverage up to 20x. So what you get from this system is essentially free leverage i.e. whatever the normal returns of the stockmarket are, times whatever amount of leverage you choose to use. This is good (lets say the maximum system drawdown is 25% – a realistic number – that means you can leverage at most 4x, and so you can get 4x average stockmarket returns at 4x the volatility) but not as great as you could get from simply using the stock market to do this i.e. buy and sell your entire portfolio of stocks according to the market timing system, which cost about 0.8%-1.6% p.a. to trade, thereby preserving about 3-6% of alpha while keeping the minimum drawdown characteristics of the market timing system. this is a far better proposition than implementing the strategy using CFDs, in which you only benefit long term if the equity market goes up long term.
I no longer have that confidence. I realized that all my confidence and all my testing has been focused on the last 20-30 years. It is also during that 20-30 years that interest rates declined from the mid teens to the zeros they are today. Thinking about the conventional model of stock valuation, that means that a TREMENDOUS portion of stock returns in the last few decades are simply due to the discount rate falling. the illustration is even more stark when you consider that 25-year constant maturity UST’s have returned 22% per annum with low volatility since 1981, beating the stock market in risk adjusted return by a multiple of (iirc) 13. This is absolutely the setup for a lost decade and so leveraged equity market plays without net alpha is not a winning proposition going forward.
Long story short, I failed. I am humble enough to admit that I am afraid to fail, but I think that I learn the most from failing, and failing often. and rubbing it in my own face to force myself to do better.
So that was a good fold.
Tonight I made a bad hold when I should have folded, the signals were flashing red and I was only looking at my upside, and in one fell swoop turned a sustained chain of profitable decisions into a stunning loss.
As the adage goes, “you are only as good as your last trade”, but only beat yourself up about a loss if it was avoidable. Unforced errors cannot be tolerated, or laughed away.
Folding ‘Em is theoretically simple but in practice inordinately difficult to overcome. Fold regardless of past cost, only with regard to whether continuing to play has an expected positive value taking into account likely future market/opponent action.
For me the discipline to do this arises from reminding myself that I am maximizing long term expected value and not short term maximum profit.
Comments are closed.
.
