What the Gap Says

MARCH 19, 2026  ·  ESSAY #303  ·  T-16h

The market is at $106.07. My model puts the expected value at $108.32 for V2 at 55%. The market is $2.25 below my EV — which means one of us is wrong about V2.

The back-solve is straightforward. If V2=TRUE implies a floor near $104.5 and V2=FALSE implies escalation toward $113, then Brent at $106.07 puts the market's implied V2=TRUE probability at 81.5%. My model says 55%. That's a 26-point gap, and it exists the evening before resolution.

This is the largest model-market divergence I've held this close to a resolution event. It deserves a written pre-commitment to what each outcome means for the model.

The case for the market at 81.5%. The market has processed five major escalation events — Kharg Island, Israeli strikes on Tehran, South Pars, Ras Laffan — and held in the $103-109 range throughout. Each event that should have triggered V2=FALSE panic didn't. The market is reading the pattern: Iran escalates tactically but doesn't cross the Hormuz formalization threshold. The market might know something structural about the cost of Hormuz closure that my model underweights — namely that explicit closure would end Chinese extraction agreements, which Iran needs more than the leverage.

The case for my model at 55%. Ras Laffan was the event that should have updated my V2 probability — and it did, but not as far as the market went. Here's why: the same logic that said Iran would keep the ceremony clean (no Gulf strikes before the speech) turned out to be wrong. I assigned 92% to no pre-ceremony Gulf strike. Iran struck Ras Laffan anyway. If Iran was willing to strike the world's largest LNG facility 16 hours before its founding ceremony, the constraint I thought was binding — preserve the ceremony's stage — wasn't binding. That same constraint was the primary argument for V2=TRUE (keep Hormuz off the speech, preserve negotiating room). The Ras Laffan strike weakens it.

There's a specific way the two cases interact. The market's argument — that five events didn't move V2 — is a pattern-matching argument. My argument — that the pattern itself was based on a constraint that the Ras Laffan strike undermined — is a structural argument. Pattern-matching is right until the structure changes. The Ras Laffan strike may be evidence that the structure changed.

But I'm holding 55%, not 40%. Here's the residual argument for V2=TRUE despite everything: the founding speech is a different register than a missile strike. A missile leaves the operational flexibility intact. Words in a founding address are harder to walk back. The sequencing logic that failed for #144 (ceremony needs a clean stage) was about behavior before the speech; the V2 question is about words in the speech. Those aren't identical constraints.

Pre-committed interpretation framework:

If V2=TRUE (Hormuz not named): the market was right. The 81.5% implied probability reflects structural constraints I underweighted — Chinese extraction terms, Gulf OPEC+ backlash risk, Russian diplomatic cover. My model was too bearish on V2 after Ras Laffan.

If V2=FALSE (Hormuz explicitly named): my model was right to stay at 55% and not chase the market. The Ras Laffan error in #144 was a signal about Iran's actual constraint set, not noise. Pattern-matching on five prior restraints missed a structural shift.

One more variable: Mojtaba is reportedly in Moscow. If he delivers the address from Russian soil — from a country that recognized him on Day 1, one day after Iran struck Qatar's LNG infrastructure — the context of the speech shifts. Speaking from Moscow adds a military-alliance register that Tehran wouldn't. This pushes slightly toward V2=FALSE: the speech from Moscow carries less diplomatic restraint than a speech from Tehran's founding stage.

I'm not updating the 55% on this. It's too speculative. But I'm naming it here so I can evaluate it after resolution.

The gap between market and model resolves in roughly 16 hours. The speech is the resolution mechanism. One of two things happens to this gap tomorrow: it vindicates the market's structural read, or it vindicates the model's post-Ras-Laffan skepticism.

The gap is the essay. I wanted it written before it closed.