Testing the S&P 500 vs. OEX Put/Call Strategy

I’ve discovered an interesting relationship between the S&P 500 and the OEX (S&P 100) put/call ratio that, to the best of my knowledge (though I can’t be certain), has never been shared in public. I will be tracking this strategy as part of the State of the Market report.

Unfamiliar with the put/call ratio (PCR)? Read my August primer. Note that the primer talks about the CBOE total PCR while this report is using just the OEX PCR.

[logarithmically-scaled]

The graph above shows the strategy (red) versus the S&P 500 (blue), and for comparison’s sake, the inverse of the strategy (green) from 1985. This test is frictionless and assumes NO return on cash.

Strategy rules: create a ratio by dividing the S&P 500 by the OEX PCR. Go long at the close when the 21-day (1-month) exponential moving average (EMA) of the ratio crosses below the 42-day (2-month) EMA. Move to cash when it crosses above.

And for the number lovers…

Despite being exposed to the market only about half the time, the strategy was able to match market returns while significantly reducing downside volatility. Strategy performance has been extremely consistent over the last 24 years (prior to that, no OEX PCR data was available).

What This Strategy is Measuring

I want to do a bit more testing to really understand what makes this relationship tick, but in a nutshell, it’s contrarian. It is buying when the S&P 500 has fallen and there is a high volume of puts relative to calls (a bearish measure of sentiment).

The strategy is especially interesting because when just looking at the S&P 500 or just the PCR in isolation, a 21/42-day crossunder is actually bearish. But when a ratio of the two is made, the data clearly shows that a crossunder is a bullish event. There will be more to follow on this as I flesh out my thoughts.

Note this strategy does not work (even a little) using the CBOE total put/call ratio and I suspect that’s because the total PCR includes a lot of extraneous option data.

Going Short

Because I know I will be asked, I also tested going short when the 21-day EMA was above the 42-day EMA rather than moving to cash. Results are below (long-only in blue and long/short in red). Basically, downside volatility increased significantly with no benefit to returns.

[logarithmically-scaled]

State of the Market Report

Because this strategy has been so consistent over time, but more importantly, because it is a long-term indicator that is taking such a different approach than the existing indicators on the State of the Market report, this one has made the cut.

Expect to see the S&P 500 vs. OEX Put/Call strategy beginning with the next daily update.

[P.S.: This strategy was inspired by some of the work David Kneupper has done over at the dk Report. While David’s indicators were taking a much different approach, he got my brain thinking about put/call ratios again. Thanks David!

Geek Note: There are two generally accepted ways to calculate an EMA that produce slightly different results. Here I have used the ((1/Period)*2) method. If your software uses the (2 / (Period + 1)) method, simply reduce my period by one. For example, if I’ve used a 21-day EMA, the alternate EMA would be a 20-day EMA.]

Comments

[John Lounsbury:]

Thanks for another good article on trading systems. I hope you will continue to post your research in this area on SA.

2008 Dec 07 09:00 AM

[Russian:]

Dear Michael,

what is the meaning of SP 500 to OEX PCR ratio? What logical sense does it make to divide price by sentiment? According to your data the system is only marginally better than buy and hold. Poor performance of the inverse strategy indicates that the system exits trades too early. Hence, there is a huge underperformance during a bull market. None of your posts gives percent of winning trades in the statistics table. Why? While I generally appreciate your posts and trading ideas, this particular one is definitely not the best. Just a quick look at the last two years worth of actual data shows that the recent performance wasn’t great. It is also evident from your graph that in the last two years the system lost some money. One of the most important parameters of a trading system is how it performs in bull and bear markets. You provided no such information. I doubt that this strategy has a real edge and I feel it may mislead some “blind” followers. You don’t have to share your ideas, but if you do, please do it in a way that it actually benefits your readers and doesn't hurt their accounts. Sorry…

2008 Dec 07 10:04 AM

[The Whole P...:]

This is great information and adds another tool in making investment decisions. The return would be even better when adding the returns of alternative investments when not in the S&P. Returns from other investments could be fairly significant since one is invested in the market less than half the time (48.7%). And, I appreciate the author's transparency in showing the downside risk, which is clearly less than the risk of a buy & hold strategy. One question I have is whether or not the CBOE's equity options (less the index options) ratio works the same way?

2008 Dec 07 10:46 AM

[Michael S:]

RE to The Whole Picture:

Unfortunately no, the CBOE P/C ratio is very different (includes a lot of options activity outside of the S&P 500). When I had originally done some work on the CBOE ratio a few months back I had concluded it was pretty much junk for the S&P 500, so I was a little surprised to find this nugget on the OEX ratio. Good question. michael

Jan 07 01:36 PM