Prediction Audit Trail
Every M3I prediction is timestamped and publicly verifiable. This page documents each prediction with the date it was made, the original call, the outcome, and links to the source posts.
5 of 5 confirmed · 1 in progress
Date Made
April 14, 2026 (Day 45)
Original Call
M3I composite model predicted framework deal resolution by ~May 14–15 with 60% deal probability. Pattern Engine sub-model initially targeted ~May 5 (Day 66); live 7-model composite refined to ~May 14–15.
Current Status
Day 77 (May 15) — DEFERRED TO ~JUL 31. No framework deal reached by the original target date. Resolution scoring extended to July 31. 45-day Lebanon-Israel extension confirmed May 15 resets the diplomatic clock. Dual-track process established: Pentagon security track (May 29), State Dept political talks (Jun 2–3). Day 91: Bloomberg confirms tentative 60-day MOU reached at negotiator level — Trump + Khamenei sign-off pending.
What the Model Got Right
- Trajectory: Called ceasefire → failed talks → escalation → de-escalation → structured diplomatic process. That is exactly what played out.
- Ceasefire timing: Called the ceasefire 3 weeks early (predicted Day 11, landed Day 39). 5/5 prior predictions confirmed.
- Escalate-to-de-escalate pattern: Called it as the dominant dynamic — confirmed by 4 rounds of talks, kinetic exchanges, and return to diplomacy each time.
- Senior envoy = endgame signal: Vance Islamabad, Witkoff Oman rounds, Trump-Xi summit, now dual-track Washington process — the pattern is playing out exactly as modeled.
- Resource war template: Correctly identified this as a resource war (not ideological), which determines the resolution pathway. Confirmed by every negotiation centering on oil, Hormuz access, and sanctions.
- Blockade as coercive leverage: Called that the blockade is escalate-to-de-escalate, not terminal escalation. Confirmed by ceasefire holding through 4 rounds + kinetic exchanges.
What the Model Got Wrong
- The date. Original target ~May 5, revised to ~May 14–15. Day 77 with no framework deal. The model was too aggressive on timing.
- Iran’s pain tolerance: Underestimated Iran’s ability to absorb economic pain and stall indefinitely behind 5 preconditions.
- Lebanon as a blocking variable: Did not model the Lebanon front as a separate complicating factor. The May 11 escalation (182 killed in a single day) nearly blew up the entire process.
- Nuclear enrichment deadlock: Did not model enrichment as a standalone blocking variable. Araghchi now admits talks are “in deadlock” with the enrichment issue postponed entirely.
- Number of variables: The model nailed the trajectory but could not nail the timing because the number of simultaneous blocking variables (Hormuz, Lebanon, enrichment, Khamenei veto, Trump-Xi, IRGC) exceeded what a date-specific prediction could absorb.
Honest Assessment
The model predicted the shape of how this conflict resolves — and that shape is confirmed. Every structural element the Pattern Engine identified (resource war dynamics, senior envoy engagement as endgame signal, escalate-to-de-escalate as the dominant pattern, blockade as coercive leverage) has played out. But the when was wrong. The Gulf War template predicted ~103 days for resource wars; we are on Day 77 with no deal and the diplomatic clock has been formally reset to a dual-track process running through July. The exit function is still running — it is just iterating through more loops than the Gulf War template predicted. Scoring formally deferred to July 31.
- Trump Project Freedom pause (May 5)
- Araghchi-Wang Yi Beijing meeting (May 6)
- China publicly demands Hormuz reopening (May 6)
- 14-point MoU nearing (Axios/CNBC May 6)
- Iran MoU rejected by Trump as “totally unacceptable” (May 10)
- Iran sets 5 preconditions; talks stall (May 11–14)
- Lebanon front explodes: 182 killed May 11 — highest single-day death toll
- Bahrain UN resolution gains 112 co-sponsors (May 13)
- Trump-Xi Summit (May 14–15): Both agree Hormuz must remain open; Xi pledges not to arm Iran; China committed to helping reopen Hormuz
- 45-day Lebanon-Israel extension confirmed (May 15) — ceasefire through ~early July
- Dual-track process established: Pentagon security track May 29, State Dept political talks Jun 2–3
- Araghchi admits enrichment talks “in deadlock” (May 15)
- MSV Haj Ali sunk off Oman — first vessel sinking in conflict (May 15)
- UK deploys HMS Dragon + multinational coalition to Hormuz (May 15)
- UKMTO 48 incidents since Feb 28
- Resolution scoring deferred to Jul 31. Resolves YES if signed MOU/framework/agreement-in-principle, joint statement on any major element (hostilities, nuclear, Hormuz, sanctions), mediator-announced framework accepted by both sides, or formal ceasefire extension/renewal. Partial deals on a single element qualify.
🔍 Independent Verification
M3I predictions are designed to be independently verifiable. Every prediction is timestamped on publication across multiple third-party platforms. No prediction can be edited, backdated, or deleted after the fact.
Substack Archive
All predictions are published as timestamped Substack posts. Substack records cannot be silently edited — all changes are logged.
m3iresearch.substack.com/archive →
X / Twitter Posts
Key predictions are posted to X with immutable timestamps. Search the @m3iresearch feed to verify any call.
x.com/m3iresearch →
Internet Archive (Wayback Machine)
This site and its pages are archived by the Wayback Machine. Snapshots provide a permanent third-party record of what was published and when.
web.archive.org →