The Put/Call Ratio Manual

Research note. Every figure computed at render from CBOE's official daily statistics as we collect them (January 2025–present) — the same file behind the live put/call tracker. Methods in How we checked it; editorial standards in How we use AI.

What is the put/call ratio?

Count every put option traded on CBOE today, divide by every call. That's it — a one-number summary of which direction the options crowd paid for. Puts profit from declines, so a rising ratio reads as fear; calls profit from rallies, so a falling ratio reads as greed. The twist that made the gauge famous is contrarian: at extremes, the crowd has usually already acted, which is why heavy put buying has so often marked lows rather than predicted declines. As of August 3, 2026, the total ratio printed 0.80, putting the 9-day average at 0.92 — the 80th percentile of our record, a state our tracker labels elevated fear.

The ratio today — all three series

The most recent session in our record is August 3, 2026 (CBOE publishes each day's figures the following morning, so the reading runs one session behind by construction). It printed: total 0.80, equity-only 0.55, index-only 0.88. The 9-day average of the total sits at 0.92 — the 80th percentile of every session in our record, a state the tracker labels elevated fear. The last-10-sessions table below shows how it got here.

Equity vs index — the two ratios live on different planets

The single most common misreading of this gauge is quoting "the put/call ratio" without saying which one. The equity-only ratio averages 0.58 in our record — individual-stock options skew heavily to call buyers. The index-only ratio averages 1.05 — index options are dominated by institutions buying portfolio protection, so puts outnumber calls on a normal day. An index reading of 1.05 is Tuesday; an equity reading of 1.05 would be a panic. The total ratio blends the two, which is convenient and slightly treacherous — moves in the blend can come from either crowd.

What our record shows — one clean episode, honestly labeled

Our data begins in January 2025, so this Manual has exactly one major fear spike to show rather than decades of folklore. It happens to be a textbook one. On April 8, 2025, the 9-day average hit 1.09 — the highest in our record — during that spring's selloff. From the spike day, SPY returned +13.8% over the next month and +26.1% over the next quarter: the crowd bought maximum protection within days of the low. The three lowest readings in our record (0.758, 0.761, 0.762) all came in calm, rising markets in mid-2025 and 2026, and preceded nothing dramatic. One spike and a handful of lulls is an anecdote, not a study — we label it that way on purpose, and the tracker keeps the count as the record grows.

Where this gauge will mislead you

• Our percentiles rank a young record. "80th percentile" here means 80th of ~400 sessions since January 2025, not of history. CBOE retired its free archives in 2019; we publish what can be verified and nothing more.
• Volume is not direction. A traded put counts the same whether it was bought in fear or sold for income. Heavy put volume usually means hedging demand — usually — but the ratio cannot distinguish a protective put from a cash-secured put sale.
• The blend hides its parts. A rising total ratio can be equity traders panicking or institutions rolling index hedges — different stories, one number. Read the split chart first.
• Zero-DTE changed the furniture. Same-day index options now dominate volume, and their hedging flows have shifted all the baselines. Treat any pre-2020s rule of thumb ("0.7 is greed, 1.2 is fear") as folklore from a different market.

The last 10 sessions

Full dataset: put_call_ratio.json.

How we checked it

Ratios come from CBOE's official daily market statistics page, collected each day (the publisher posts each session's figures the following morning, so the site's reading runs one day behind by construction). The 9-day average is an exponential moving average; percentiles rank the latest average against every session in our record; forward returns from the spike are SPY price returns computed at render. When CBOE restates a day, our next collection picks it up.

Frequently asked questions