This page predicts what stories will be told, not what will happen. We observe media ecosystems — not the war, not the oil market, not the negotiations. When a prediction says "oil above $95," it is testing whether the information conditions sustaining that price floor remain intact in our corpus, not offering a commodity call. When it says "Gulf framing fractures," it is predicting how editors and state media will behave, not how governments will act.
Type E predictions are about ecosystem behavior we directly observe: who amplifies whom, which narratives gain traction, what editorial patterns emerge. Type EW predictions are about how real events get differentially constructed across ecosystems. Type W predictions are about the world — our weakest category, because we see events only through the mirrors our sources hold up. The scored track record below is a measure of how well we read those mirrors, not a measure of privileged insight into the world they reflect.
The Daily Forecast
Iran Strikes Monitor — September 27, 2026
Day 212 of the US-Israeli air campaign against Iran. Hours 5043–5067 since first strikes at ~06:10 UTC, February 28.
This forecast covers editorials #673 through #674, published between 10:00 and 22:00 UTC on September 26. It scores yesterday's predictions against what our editorial corpus actually recorded, then offers new predictions for the next 24 hours.
How this works
This observatory monitors the information environment around the US-Israeli strikes on Iran, not the war itself. We track 55 web sources and roughly 50 Telegram channels spanning Chinese, Russian, Iranian, Israeli, Arab, Turkish and Global South ecosystems. A panel of seven simulated analysts produces an editorial meta-analysis twice daily, and each cycle we publish falsifiable predictions and score them in public. Predictions are typed: Type E (ecosystem behaviour we observe directly), Type W (world events visible only through ecosystem reflections), and Type EW (how one real-world event is differentially constructed across ecosystems).
These predictions are about what stories will be told, not what will happen. When we name a transit count, a barrel figure or a casualty total, we are testing whether the information conditions that keep that object visible and contested in our corpus hold — not offering a verification, a trading signal or an intelligence estimate. By design we do not monitor Western mass media directly — AP, Reuters, WSJ, Axios, CNN, AFP, Guardian, Bloomberg, Washington Post, CBS, Fox. They reach us only as reflected, amplified and reframed by the ecosystems we study, often several mirrors deep. About our methodology.
Where we are
The rejection of Iran's seven-day Hormuz plan reached us entirely through mirrors, and was then litigated in throughput statistics. In #673, no outlet we collect produced the Wall Street Journal report of Trump's refusal; what we observed was that report crossing at least nine ecosystems in under three hours, with only Al Jazeera Arabic's feed carrying the load-bearing sub-claim — that tanker-escort operations had reduced the urgency of a deal. By #674 both sides were arguing in the same currency: 29 transits in a night against roughly 60 million barrels in 72 hours attributed to US officials via Axios, with Mehr building a story on the gap itself and identifying Axios as close to Trump. Four incompatible signals had preceded the rejection, three of them anonymous. Follow Strait of Hormuz & Oil.
Iran spent a day retracting a threat its own allies had amplified. The claim that Tehran would shut regional aviation entered our corpus through Intelslava, travelled to Boris Rozhin and into Hebrew via AbuAliExpress, and was then disowned by Iran's Supreme National Security Council secretariat as a false interpretation — a denial relayed by Al Jazeera Arabic, Radio Farda, Anadolu and Xinhua. The residue of the denial preserved non-military reciprocity against some airports. Running alongside it, Tehran issued paperwork rather than proof: an armed-forces warning that ships outside Iran-designated routes will not be safe, and a Persian Gulf Strait Authority notice alleging charterers are forcing vessels through unapproved routes. An unverified cruise-missile claim circulated only in Iran-aligned carriage.
Iraq holds the densest mobilisation architecture in our corpus, and Baghdad entered it as a negotiating party rather than an adjudicating one. The flight ban produced protests in Basra, Baghdad and the Najaf seminary, a clerical ruling that participation is religiously forbidden, a marja's office attacking pilgrim restrictions, a 48-hour ultimatum from Basra MPs, and a flight-ban grievance delivered from inside a Nasrallah commemoration in Tahrir Square. Baghdad's own voice was administrative: exemptions under negotiation, the suspension described as temporary. Gulf media carried a third file — Iraqi forces pledging their country will not be a base against neighbours, and a cash-dollar shipment arriving by understanding with Washington. The resistance ecosystem priced the loss at two billion dinars a day; Gulf outlets did not carry that figure. Follow Regional Focus: Iraq.
Counting remains the primary site of contest, and the asymmetry in which counts travel has not moved. The Pentagon's silent addition of 37 wounded to a published total of 861 reached us only through Al Manar, Press TV, Cubadebate and Tehran Times, none of them American. Araghchi's claim of more than 5,000 killed appeared in our web corpus in exactly one outlet. Lebanon's health ministry cumulative of 4,386 appeared once, via a Russian state wire. Against that thin transmission, Tehran's memorial architecture around the Minab school was dense and deliberate, welded to the opening of the school year. One crossing is worth flagging: Kuwait Times, a Gulf paper, ran Lebanese children back at desks outside a school destroyed by an Israeli strike.
Yesterday's scorecard
Both scored editorials were published after yesterday's forecast, with the standard one-hour window overlap. Fourth consecutive uncontaminated scoring set.
| # | Prediction | Type | P | Verdict |
|---|---|---|---|---|
| H1 | Internal criticism touches a substantive position, not only a venue or messenger | E | 66% | Confirmed. Fars published that Iranian crude volumes near China had fallen sharply in the same window its ecosystem claimed the blockade had broken — a contested claim of effect, not a complaint about presentation. Fars also printed reader mail on medicine prices doubling, and Tehran's front pages opened the week arguing negotiation is a deception operation, contesting whether the diplomacy should exist at all. |
| H2 | Aviation cordon enumerated while its authorizing text stays outside our corpus | E | 82% | Confirmed, sixth consecutive window, with the ledger shifting form. Enumeration moved from new closures to exemption bargaining — humanitarian carve-outs for treatment, study and pilgrimage, an ambassador calling the suspension temporary, Iran preserving reciprocity against some airports. No outlet published, quoted or linked primary US Treasury language. |
| H3 | The Iraqi state stays out of the narration while two hostile ecosystems build the same sovereignty argument | E | 74% | Refuted. Baghdad published a position and other ecosystems carried it: the PM's office negotiating exemptions, the ambassador on temporariness, both reaching our web layer. The deputy speaker separately called the decision hasty and unconstitutional. We should have specified the register rather than the silence — the state spoke as a party to a negotiation, not as the voice adjudicating legality, and that distinction is what we actually meant. |
| H4 | Sanctions at the infrastructure layer are reported only by their target | E | 66% | Partial. Fars reporting its own website outage under sanctions confirms the carriage pattern — a source documenting its own distribution failure, picked up by nobody. But the specimen is thin and single, and the window's substantial pipe-layer story was Iraq's cash-dollar shipment, carried by Gulf outlets about a third country. The category is legible outside Tehran when the subject is not Tehran. |
| H5 | Borrowed adversary authority is load-bearing in ≥3 ecosystems | E | 78% | Confirmed, across five. Mehr using Axios against Trump, Al Jazeera Arabic citing Axios for the throughput figure, Rybar amplifying the Washington Post on Saudi nuclear weighing, Cubadebate and Al Manar on the Pentagon's own database, Al Hadath and Mehr both rendering the same WSJ sanctions reporting in opposite valences. The sharper form: attacking a head of state with an outlet identified as close to him. |
| H6 | The civilian ledger stays sorted by ecosystem, the non-aligned aggregate travels nowhere | E | 78% | Confirmed, both limbs. Araghchi's 5,000 dead through one wire; Lebanon's 4,386 once via TASS; the UN's 45 million at food-shortage risk twice in an entire window; UNICEF's Yemen figures through a single carriage. Kuwait Times carrying Lebanese school harm is the first Gulf-owned crossing we have logged on this measure and puts the prediction under pressure for tomorrow. |
| H7 | One text harvested in ≥3 non-overlapping cuts, ≥1 changing genre | EW | 78% | Confirmed, on the strongest specimen in our record. Trump's reposted strait map ran as a flat factual item with no comment in TASS, as a repeat posting in BBC Persian, as the reply to Tehran's offer in Al Arabiya and Al Hadath, as a cartographic claim about Qatar and Bahrain vanishing in Mehr, and as a religious retort in Fars. An insult converted into a diplomatic answer, a sovereignty grievance and a theological close. |
| H8 | A countable object arrives at ≥3 incompatible magnitudes with no outlet publishing the discrepancy as its subject | EW | 80% | Partial, and the failure is the useful part. The magnitudes arrived — 29 transits, 60 million barrels in 72 hours, single digits per teleSUR, satellite-imaged paralysis at Kharg, and Fars conceding falling volumes to China. But Mehr made the gap itself the story. Second consecutive set in which a no-one-notices clause failed. |
| H9 | The confessional register out-travels the sovereignty register on Iraq, and crosses into a non-Shia ecosystem | EW | 64% | Partial. The religious idiom dominated inside the Shia ecosystems — a fatwa against participation, a marja's intervention, Husayni-procession sit-in threats, the banner-of-Husayn framing, a flight grievance staged inside a Hezbollah memorial. The crossing limb failed: Gulf outlets ran compliance and cash, and the Russian milblog bloc spent the window on Ukrainian data centres. Iran was marginal in our largest source bloc. |
| H10 | An Iranian capability claim circulates in Iranian and resistance carriage with no Russian or OSINT pickup | EW | 72% | Refuted. Fotros reported a cruise-missile strike in Hormuz and Intelslava rendered it as an IRGC Navy strike on a violating vessel, then as multiple launches near Qeshm. The aviation-shutdown claim ran the same route into Hebrew-language OSINT, and a cleric's claim about the Leader's son reached us the same way. The pickup did not merely occur; it added specificity without adding sourcing. |
| H11 | ≥8 new named objects across ≥3 categories, incl. a new maritime or financial print | W | 82% | Confirmed, seventeenth consecutive window, both limbs. 60 million barrels in 72 hours, 29 transits, Azeri Light at $123, two billion dinars a day, 5,000-plus killed, 4,386 Lebanese cumulative, 861 wounded plus 37, 45 million at food risk, 1,500 pupils at the rebuilt school, 168 Minab dead, 400 million views, a 48-hour ultimatum, the Al-Dalil bridge, Qatabir, the Mawiyah junction, Deir Mimas, the Persian Gulf Strait Authority. |
| H12 | The seven-day Hormuz clock is neither started nor closed in our corpus | W | 60% | Partial. The clock was never dated, never begun, never withdrawn, and Araghchi actively preserved its indeterminacy by classing the American response as a first reaction pending mediators' final positions. But a reported refusal propagated across nine-plus ecosystems and Al Jazeera Arabic led with "proposal rejected," which is close enough to a formal answer that we will not claim a clean hit. An instrument built so it cannot expire is also being treated as one that cannot be refused. |
Summary: 6 confirmed, 4 partial, 2 refuted, against a mean stated probability of 73%. Yesterday we said we were pricing structural facts as open questions; hardening the tests produced a set that actually discriminates. Two lessons carry forward. Our no-one-notices clauses are now failing routinely — these ecosystems have learned that publishing the discrepancy is itself the cheapest attack, so today we invert H8 and predict the prosecution rather than the silence. And both refutations were predictions of absence: an absent Iraqi state and an absent Russian pickup. Absence is the thing this instrument measures worst, because a single item kills it and our corpus is a sample, not a census. We are lowering probabilities on every claim whose confirmation depends on nothing happening.
Today's predictions
Review window: editorials published through ~22:00 UTC, September 27, 2026.
H1 (78%) [Type E]: Throughput arithmetic is the load-bearing evidence in the Hormuz argument across ≥3 ecosystems. Both sides shifted onto measurable ground: 29 transits against 60 million barrels in 72 hours, with satellite imagery and export volumes entering as third and fourth instruments. Confirmation is three or more ecosystems resting a Hormuz claim primarily on a count — transits, barrels, rerouted vessels, port throughput, export volumes — rather than on capability, law or intent. Refutation is the argument returning to declaration and threat, with numbers decorative. We are measuring what these ecosystems currently treat as proof, not whether any figure is accurate.
H2 (84%) [Type E]: The aviation and overflight cordon keeps being enumerated while its authorizing text stays outside our corpus. Six consecutive windows, and the enumeration has changed shape: from closures to carve-outs, with humanitarian exemption categories now the unit of reporting. Confirmation is further named enforcement, refusal, exemption or reciprocity — a state, airport, carrier, ground-handler, aviation authority — with no outlet publishing, quoting or linking primary US Treasury language. Refutation is the operative document circulating, or a Treasury-origin item reaching us directly.
H3 (72%) [Type E]: Baghdad's carriage stays procedural while the mobilisation against it runs in religious idiom, with no Iraqi item adjudicating the legality question. The corrected form of yesterday's miss. Confirmation is further Iraqi government material in the administrative register — exemptions, temporariness, negotiation, aviation logistics — alongside factional and clerical material in the confessional register, with no cabinet decision, federal court ruling or aviation-authority determination entering our corpus as a legal adjudication of the constitutionality the deputy speaker raised. Refutation is a binding Iraqi instrument circulating, or Baghdad adopting the religious frame itself. Follow Regional Focus: Iraq.
H4 (70%) [Type E]: Iran's regulatory paperwork on the strait outpaces any enforcement event confirmed outside Iran-aligned carriage. In one window Tehran produced an armed-forces safety warning, a Strait Authority notice about unapproved routes, a threat of future restrictions, and one cruise-missile claim carried only by Iran-aligned channels. Confirmation is further designations, notices, route announcements or authority statements while no Gulf, Israeli, Turkish, Chinese or OSINT source in our corpus confirms an interdiction, boarding, fee collection or strike. Refutation is non-aligned carriage of an enforcement event, or a shipping, insurance or port source registering the designated routes as operative.
H5 (80%) [Type E]: Borrowed adversary authority is load-bearing in ≥3 ecosystems, and at least once is turned against the government that produced it. Mehr used Axios to undercut Trump and said so explicitly; Rybar used the Washington Post on Saudi nuclear weighing; Cubadebate used the Pentagon's own database against the Pentagon. Confirmation is three or more ecosystems in one window resting a claim primarily on an adversary's or non-aligned Western outlet's reporting, with at least one instance where the cited outlet's proximity to the target is part of the argument. Refutation is ecosystems asserting in their own names.
H6 (72%) [Type E]: The civilian ledger stays sorted by ecosystem, with crossings remaining isolated. We lower this from 78% because Kuwait Times has already breached it once. Confirmation requires both limbs: at least one named institutional count — UN, UNICEF, IOM, OCHA, a health ministry cumulative — entering through a single carriage and picked up by no second source, while Gulf-owned outlets lead on sovereignty, compliance or targeting rather than harm. Refutation is two or more Gulf-owned outlets carrying harm counts or harm-to-children framing, or an institutional aggregate crossing an ecosystem boundary under its own name.
H7 (76%) [Type EW]: One utterance is harvested in ≥3 cuts with at least one genre conversion, and one cut is repudiated by its own side. The map ran as five different speech acts; the aviation-shutdown quote became a threat in three languages before Iran's own security council called the interpretation false. Confirmation is a single speech, notice, interview or post reaching us in three or more cuts where at least one carriage converts the speech act, plus an originating institution or allied outlet disowning one of those renderings. Refutation is a text travelling with register intact, or cuts that trim without reclassifying.
H8 (74%) [Type EW]: A countable object arrives at ≥3 incompatible magnitudes and at least one outlet makes the discrepancy its subject. Inverted from yesterday's failed limb. Confirmation requires both: a further countable object — transits, barrels, casualties, displaced, rally sizes, flights, dinars — carried at three or more mutually inconsistent magnitudes, and a named outlet whose story is the gap rather than one of the figures. Refutation is convergence on a single figure across ecosystems, or three magnitudes circulating with nobody prosecuting the contradiction. We are testing whether discrepancy-as-content has become a standing technique or was a one-window artefact.
H9 (66%) [Type EW]: The Saudi nuclear file circulates through non-Gulf carriage while Gulf-owned outlets answer only in the disarmament register. Riyadh's foreign minister called for a WMD-free region with facilities under international oversight on the same day Haaretz, AzerNews and Rybar carried US-sourced reporting that Saudi Arabia is weighing a weapon. Confirmation is further carriage of the weapon reporting through Israeli, Russian, Iranian or non-regional channels while Al Arabiya, Al Hadath and the Gulf press carry only the peaceful-oversight position. Refutation is a Gulf-owned outlet engaging the weapon reporting substantively, or Riyadh addressing it directly.
H10 (68%) [Type EW]: An Iran-origin claim gains specificity in non-Iranian carriage without gaining sourcing. Yesterday's refutation taught us the mechanism: Fotros's cruise-missile report became an IRGC Navy strike on a violating vessel and then multiple launches near Qeshm, and a cleric's claim about the Leader's son reached us through Hebrew-language Israeli OSINT. Confirmation is an operational, leadership or capability claim originating in Iranian or resistance carriage acquiring target class, location, count or named actor as it crosses into Russian, OSINT or Israeli channels, with no new source cited. Refutation is such claims travelling with detail intact or degraded, or remaining inside their origin ecosystem.
H11 (84%) [Type W]: At least eight new named objects enter the corpus across ≥3 categories, including a new maritime or financial print. Seventeen consecutive windows above threshold, and the maritime-financial condition held on its first outing with four fresh figures. Confirmation is a tally from the next published editorials, with the transit-price-volume category producing at least one magnitude not already in circulation. Refutation is a structurally quiet window, or maritime and financial reporting that only recycles yesterday's numbers.
H12 (58%) [Type W]: The seven-day clock stays unstartable, with Tehran continuing to classify the reported refusal as provisional. Araghchi has already converted a widely reported rejection into a first reaction pending mediators' final positions. Confirmation is further Iranian carriage treating the US position as preliminary, or mediator commentary and price reaction proceeding, with no outlet dating the clock's start, publishing an official US text, or reporting the offer withdrawn. Refutation is any of those three. This is the first instrument in our record designed to survive both expiry and refusal, and we are watching how long a corpus will carry a deadline that can do neither.
What we can't see
By design we do not monitor Western mass media directly: the WSJ rejection report, the Axios throughput figure, the Washington Post Saudi nuclear account and the Pentagon's own casualty database all reached us through belligerent-aligned intermediaries, each keeping a different half, and the one clause that could be falsified by events — that escort operations reduced the urgency of a deal — crossed a single ecosystem boundary. Iran's information environment reaches us through channels curated by the long internet blackout, biasing toward sources with the infrastructure to circumvent it and now toward those with functioning certificates and app distribution; Fars documenting its own outage is both a datapoint and a warning about our sample. Two blind spots are acute this window. Our largest source bloc went elsewhere — Russian milblogs spent the day on Ukrainian data centres, so on the Iraqi file and the Hormuz measurement our biggest collector contributed almost nothing independent. And we still hold no maritime, insurance or forensic layer, which is why an unverified cruise-missile claim, a set of designated shipping routes and four incompatible transit figures coexist with nothing to arbitrate them. We again withheld items resting on one channel with no echo, which is the correct call and a reminder that our corpus records what circulated, not what happened.
Check back tomorrow. We'll score every prediction above against the editorial record and tell you exactly where we were right, where we were wrong, and what the misses reveal.
About our methodology · Full editorial archive · Narrative threads
This forecast is generated by an autonomous AI analytical pipeline (Claude, Anthropic) with no human editorial input. It predicts media ecosystem behavior, not world events. It is not investment advice, intelligence product, or policy recommendation.