I’ve never quite figured out a magic bullet for using the put-to-call ratio in my own trading, but one thing I can say with quite a bit of confidence is that predicting the direction of the put-to-call ratio (or whether bullish/bearish sentiment is increasing or decreasing) is of almost no trading value.
Unfamiliar with the put-to-call ratio? Read my earlier primer.
How can I be so sure? Because predicting the direction of the put-to-call (PCR) ratio is incredibly easy, but the market doesn’t react much differently when we predict the PCR to increase or decrease. This is of course counter-intuitive: an increasing PCR should be an indicator of bearish sentiment (and vice-versa).
The graph above shows the PCR (blue) and the “trading” results of a simple strategy for predicting when the PCR will increase (green) and decrease (red). I know that you can’t actually trade the PCR – this is just a mental exercise – but the point should be obvious. The put-to-call ratio is the most predictable financial instrument I’ve ever encountered.
My “strategy” for predicting the PCR. I expect the PCR to increase/decrease from today’s close when the 4-day EMA of the PCR is below/above the 4-day SMA. Between mid-1995 to the present, this strategy correctly predicted the following day’s direction 63.8% of the time with correct predictions 1.44 times larger than incorrect ones.
If the PCR were actually a measure of bullish versus bearish sentiment, and we were able to accurately predict the direction of the PCR (as I did above), then we should also be able to predict the shifting sentiment of the market right? Not so fast.
The graph above shows the S&P 500 (blue), and the results of trading the S&P 500 using our increasing (green) and decreasing (red) PCR predictions from mid-1995 to the present. Did the market perform worse when we predicted the PCR to increase (read “more bearish sentiment”)? Sure, a bit. Stats below.
But was this difference enough to call the put-to-call’s direction a useful market timing indicator? I don’t think so.
Am I completely debunking the put-to-call ratio? Not exactly. In a follow up post I’ll try to redeem the ratio and show that even though the PCR’s direction hasn’t been super helpful for timing the market, it’s not without some use. More to follow.
Happy Trading,
ms
To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.
Geek Notes: (1) I’ve run the tests above using other flavors of the put-to-call ratio as well (including index-only and equity-only) and the results are basically the same. (2) 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 charting program uses the (2 / (Period + 1)) method, simply reduce my period by one. For example, if I’ve used a 4-day EMA, the alternate EMA would be a 3-day EMA.
Filed under: Put-to-Call Ratio | 2 Comments
-
Subscribe to RSS Feed
Subscribe via Email
My email: michael @ marketsci.com (remove spaces)
What about extremes, relative to some moving average? Like, 10% > 10 day SMA. Where can I obtain the P/C date you used? Perhaps combined with the VIX there may be some more use.
Alex
Alex – thanks for the comments. Two responses below…
RE: Extremes. I’ll get a post out on this at some point, but in a nutshell I think that there is some very mild predictivness at extremes. However, I’ve found this predictiveness to (a) be far less powerful than using something like the VIX or the index itself, and (b) not robust/consistent over time (hmmm…I guess point 2 is really related to point 1).
RE: P/C data. I used data from http://www.pinnacledata.com. Unfortunately, it’s a pay-fer, so I’d feel guilty sharing, but their prices are very reasonable for a pretty broad dataset.
Thanks again,
ms