Market Sentiment: What the Relative Equity Put/Call Ratio Tells Us

click to enlarge

As the chart above indicates, rises in the relative equity put/call ratio (which I operationalize as the 10-day equity put/call volume ratio divided by the 200-day ratio) have been reliably associated with market bottoms since 2004. On Friday, we saw the 10-day ratio exceed the 200-day ratio by over 20%.

Since 2004 (N = 1033 trading days), we've had 129 occasions in which the relative put/call ratio has met or exceeded the 20% threshold. Twenty days later, the S&P 500 Index (SPY) has been up by an average of 2.22% (103 up, 26 down), much stronger than the average 20-day gain of .11% (537 up, 367 down) for the remainder of the sample.

More broadly, when the relative equity put/call ratio has been above 1.0 (meaning that we're seeing more puts traded relative to calls over the past 10 days compared to the past 200 days), the next 20 trading days in SPY have averaged a gain of .84%. When the relative ratio has been below 1.0, the next 20 days in SPY have averaged a loss of -.31%.

While it is certainly possible for the market to get weaker in the short run even when the ratio has exceeded 1.20 (50 of the 129 occasions were down after five trading days), it's generally paid to fade extremes in the relative equity put/call ratio.

Comments

[User 142222]

hi,
Nice article, is the put/call ratio available in real or close to real time publically for free?

[primorec]

www.cboe.com/data/Intr...

[bhakta]

I agree, the market is very over sold, but it is not every year, or every decade that the entire system stops working and worldwide lack of confidence prevails

[Dynamo2000]

I believe it has long been observed that put/call ratios are correlated with moves in the market. I have never however read of any reasonable explanation as to why the ratio of the VOLUME of puts to calls should be so correlated. One would think that the level of implied volatility might be correlated - but that the VOLUME should be correlated I find very puzzling. Can anyone suggest a plausible explanation as why such a correlation should exist?