Condor Options

Some traders are initially attracted to options because of the leverage they provide.  But leverage, as anyone who’s followed the fate of the investment banks knows, is just a means for magnifying outcomes.  A leveraged risk-taker will experience more glorious wins and more disastrous losses, like a deranged person who shouts both poetry and obscenities (instead of whispering them quietly to himself, like the rest of us).

We use options not for the leverage, but to articulate views that are otherwise ineffable, or are expressible only crudely.  For instance, a stock trader who expects a period of inactivity or minimal movement has no obvious way to quantify that position, while an options trader can express the same view in several ways.  Options can thus be used to enhance the returns and reduce the risk of conventional long/short strategies.

Below is the equity curve of a profitable but not at all complex strategy that takes long and short positions in the S&P 500.  It exploits mean reversion tendencies on a multi-day timeframe, uses time-based exits, and is otherwise unremarkable (requiring only 9 lines of code).  Results are for a $100,000 portfolio from January 2007 to January 2009 and exclude transaction costs.

We selected the last two years of trade history for clarity’s sake, but the strategy exhibits two consistent problems over the last ten years: 1) very large occasional drawdowns, and 2) multi-month stretches of sideways churn.  The strategy gave back nearly all of its gains in the crash of fall 2008, and made little progress through the spring of 2008.  Simply put, relying on the time series of the underlying to generate profit has some real limitations for a strategy like this one.

However, if we can express the same core view – that there are market regimes during which mean reversion is strong – but use a different “language” to articulate that view (i.e. options, rather that the underlying asset), we may get significantly different results.  The equity curve below uses the same entry and exit conditions as before, but takes long and short positions by selling out of the money vertical spreads.  For example, the strategy generated a long signal for June 6, 2008.  With SPX at 1360.44, we could have sold the June SPX 1305 puts and bought the 1295 puts at the close for a net credit of about $1.40.  Like most of the positions in this strategy, that spread would have expired worthless, allowing us to keep the full credit received.  By contrast, a long position in ES futures or SPY shares would have incurred about a 2% loss over the same period.

The first advantage of the options-based approach is that it is more forgiving of incorrect directional signals: losing trades with the underlying become, in many cases, profitable options trades.  The options approach also reduces the two flaws we noted in the standalone strategy: 1) because we’re using risk-defined spreads, the large drawdowns occur with the same frequency but are less severe, and 2) because our profit potential isn’t restricted to the tick-by-tick changes in the underlying (that is, because we also rely on time decay), periods of sideways price movement can now also generate returns.

The differences between the two approaches are borne out in the numbers: the strategy expressed with options has superior risk-adjusted returns (Sharpe ratio), and a smaller maximum peak-to-valley drawdown.  Again, there’s nothing particularly special about this strategy, or about options as such.  What matters is that, in this instance, option spreads allow us to express our core view with more precision.  Although the investing public seem to have little tolerance these days for anything that smacks of financial complexity, the simple fact is that some financial propositions cannot be uttered in the language of buy, sell, or hold.  For more elegant elocution, derivative language is needed.

Image courtesy of Flickr user dailypic.