Futures Signals

Module 1. Fading a Predictably Failing Trend Following Technique

This module is placed as the first since it is the only module that has survived the numerous iterations of my Option Project strategy AND is closest in methodology (albeit with numerous enhancements) to the original option strategy (OPv1 – or Option Project version 1) that went live 1/1/09.

This module is based upon a futures trend following system that has historically BEEN WRONG over 70% of the time and is currently (i.e. since 10/11) wrong about 80% of the time.

The accompanying Excel Spreadsheet is a static representation of the dynamic data that normally gets updated daily (and even intraday) via DDE links.

A Pause to point out a few things…

1. You’ll quickly notice that if one were to fade the signal generated by the trend following system one would wind up being “right” over 80% of the time for these markets since 10/11.

2. You’ll also quickly notice that the average win when the signal was right was about 3.5 times the size of the average loss.

If you had “bet” a hypothetical $100 each of the 472 times against the signal you’d have won $822.31 the 384 times the signal was wrong and you’d have lost $478.77 the 88 times the signal turned out to be right – exclusive of commissions and slippage – for a net profit of $343.44 or 72.8 cents per $100 bet or about $728 per $100k bet. Another way to look at this is a profit of about $728 per futures contact IF one assumed the average one was valued at $100,000. Of course this doesn’t seem like much but when one compares the return relative to the margin required instead of the notional contract value it quickly becomes apparent that the return is much more desirable – e.g. if the average margin requirement was $5,000 per futures contract the return would be on average $728/$5,000 or approx. 14.56%.

Of course the above example is overly simplistic and in fact impossible to achieve as the “bet” must necessarily be different for each market due to the contract multipliers and other factors which I’ll go into later (risk standardization based on vol and the percentage of one’s bankroll one wants to risk on any one trade are two prominent examples).

Additionally, one would almost surely not ride out EVERY loss to its final and maximum culmination. Would one? Of course not. This is where risk management techniques get built into the model to cut one’s losses before they get “too big” or ideally, before they get even as big as the average expected win – or depending on one’s own comfort level, some multiple of it. Much like the entry points generated, one should not enter without clearly predefined exit points.

Back To Methodology

So, although one of the modules DOES in fact deal with trading futures based on this trend following technique  – with NUMEROUS risk management enhancements to keep losses small –  THIS module deals specifically with creating a futures option strategy to take advantage of both the known failing of the trend following technique as well as the expected results of it – i.e. the size (and relative size) of the wins & losses of it.

So, armed with the trend following behavior the idea was to create an option strategy that sold calls when a long signal was generated & sold puts when a short signal was generated. All that had to be determined was what expirations, what strikes, and what quantity to sell.

Based on several variables a formula was created which accounted for the current price of the underlying future, the contract multiplier of the underlying future, the volatility of the underlying future, & the percent of equity to risk on each trade, the preferred weighting of each particular market, and a time adjustment in the volatility multiplier based on the remaining days until option expiration, among other factors such as option deltas & thetas, e.g. Additionally, clearly defined exit points or risk mitigation techniques were put in place immediately upon successful execution of any new position.

Initially, and to prove up the robustness of the strategy, ALL MARKETS had the same methodology (but over time I have been able to customize the input requirements for each market over the 2004-2008 5 year back testing period and 2009-2012 4 year live trading period) to optimize market specific characteristics that have robustly presented themselves over the 9 years to date WITHOUT CURVE FITTING.


Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: