Opening 10-minute aggregate options pressure predicts 09:41-to-close underlying return

The simplest version of the oldest question about options flow: when the opening tape shows more premium going one way than the other, does the underlying follow? A yes would have meant that aggregate options pressure is a directional signal readable in the first ten minutes and usable for the remainder of the session, which is a large and easily exploited inefficiency. It ran first in this ledger for exactly that reason: it is the cheapest strong claim to falsify, and a fast clean null was a live and acceptable outcome.

Tests the AGGREGATE options-net-premium representation of opening options pressure (dollar net premium, all trade types pooled). A null here does not falsify trade-level, aggressor-partitioned (side-partitioned flow), or opening-position variants of the broader opening-pressure mechanism (amendment 9). Follow-ups register as new hypotheses with parent lineage.

Two liquid index proxies over seven months of sessions. The signal is opening net premium: net options premium summed over 09:30–09:40 and divided by gross premium over the same window, so the quantity is a unit-free imbalance rather than a dollar amount. The outcome is the log return from 09:41 to the close; the start sits a full minute past the end of the formation window so the signal is actually available when a position would open. Estimation pools both symbols with symbol intercepts and clusters standard errors on sessions. Inference is a permutation test with 999 draws against a preregistered positive direction: more buying pressure, higher subsequent return.

No. The pooled coefficient came in at 0.0018 with t = 1.57 and a permutation p of 0.13 over 140 sessions: correctly signed, small (a fully one-sided opening tape maps to under twenty basis points of subsequent return), and comfortably inside what reshuffling the signal produces by chance. Aggregate opening options pressure carries no rest-of-session directional information at this granularity. The result did two jobs. It retired the first and most obvious representation of the flow-to-direction mechanism, and it exercised the whole instrument (registration, frozen signal construction, permutation inference, a recorded disposition) on a claim the lab was willing to lose. Four further representations followed it, and the family closed null.

Formation windows close roughly a minute before outcome windows open, and measured feed publication latency consumes most of that slack. Live availability of each signal at the formation boundary is therefore tight rather than generous.

One of the two symbols looked stronger on its own (t = 2.70), but the registered test is the pooled one and that split is recorded as a diagnostic, not as evidence. A subgroup does not redefine a preregistered claim; it stands as a possible descendant hypothesis and nothing more.

The registered claim covers the aggregate representation only: net premium with all trade types pooled. A null here does not falsify trade-level, aggressor-partitioned, or opening-position variants of the same mechanism, and the lab tested several of those as separate registered experiments rather than reading them into this result. Like every result the lab has published so far, it is a discovery-substrate result: preregistered, but not yet confirmed on data the research process has never seen.