66.73 Million at 4:40 and 7.75 Million at 3:15 Are Not Subtractable

66.73 Million at 4:40 and 7.75 Million at 3:15 Are Not Subtractable

Two numbers on Cboe's page for Thursday, 20 August 2026 invite a quick calculation. The U.S. Options Current Market Statistics summary displayed 66,728,881 matched contracts at 4:40 p.m. Central Time. Lower on the page, a separately labelled Cboe Exchange Market Statistics table showed 7,751,037 contracts at 3:15 p.m. The difference is 58,977,844 contracts.

That subtraction is accurate arithmetic. The 58,977,844 figure has no documented relationship to activity after 3:15 because the records carry different table labels and different timestamps. A put/call ratio drawn from the earlier Cboe Exchange table has the same boundary: it describes that table, at that time, rather than the final six-venue market total.

Thursday, 20 August was the latest completed U.S. options session when this article was prepared. The useful task is to read the labels before turning a dashboard number into a market signal.

Observed facts: two records on one page

Cboe's Market Volume Summary lists six venue groups and a 16:40 Central Time data stamp. Its rows reconcile to the displayed total.

The page also contains a table headed "Cboe Exchange Market Statistics for Thursday, August 20, 2026." At its 3:15 p.m. Central Time line, that table displayed 4,547,516 calls, 3,203,521 puts, 7,751,037 total contracts and a 0.70 put/call ratio.

The headings supply the boundary. One record is a Market Volume Summary across six named venue groups at 16:40. The other is a Cboe Exchange table at 15:15. Neither label says that one field is a subset of the other at the same moment.

Model output: the difference fails the scope test

The arithmetic is straightforward.

66,728,881 - 7,751,037 = 58,977,844

The result is a difference between a six-venue final summary and an earlier, separately labelled exchange table. No common time-and-scope basis connects it to an interval measure, Cboe's post-3:15 share or a transfer of activity between venues.

Think of a football stadium with a full-stadium attendance count at 4:40 and a turnstile count at one gate at 3:15. Subtracting the gate count from the stadium count yields a number. It does not reveal how many people arrived after 3:15 because the counters cover different places as well as different times.

This check is especially useful when a dashboard places several panels close together. Shared branding and a shared date do not supply a shared denominator.

The put/call ratio belongs to its own table

Within the 3:15 Cboe Exchange row, the displayed ratio follows the familiar puts-divided-by-calls calculation.

3,203,521 / 4,547,516 = 0.70446, displayed as 0.70

That is a model output based on the two reported columns in that row. A final market-wide put/call ratio, a position view and an order purpose require fields beyond that row. The page's columns do not identify the underlying security, strike, expiration, premium, customer, buyer or seller, or whether a transaction opened or closed a position.

OMP's explanation of Cboe's index and equity put/call split applies the same discipline to option class. Its guide to fragmented U.S. options market share explains why venue labels matter, while volume and open interest separates a day's trading from outstanding contracts. The options-chain guide and implied-volatility guide identify the contract fields that an aggregate volume panel cannot provide.

Interpretation: make every subtraction earn its denominator

The 20 August page is a reminder that the first calculation is often a data-alignment check. Before comparing two options-market figures, record four fields beside each number:

1. The table heading and venue or market scope.
2. The exact timestamp and time zone.
3. The unit measured, such as contracts, calls, puts or a ratio.
4. The contract detail supplied, including any underlying, expiry or customer classification.

Perform a subtraction, ratio or trend comparison only when those fields align with the question being asked. For a mismatch, retain the values as separate observations. Cboe also states that the information is furnished without responsibility for accuracy and that its use is subject to its website terms, so a dated capture remains a research record rather than a permanent market fact.

Options involve loss, liquidity, transaction-cost, tax, exercise and assignment considerations. This material is general education, not personal financial, legal or tax advice or a recommendation to buy, sell or hold an option or security. The practical rule is concise: calculate from one labelled series at a time, then move to the option chain for the particular contract under review.

Sources

• Cboe U.S. Options Current Market Statistics, data as of 16:40 Central Time on Thursday, 20 August 2026