The Honest Uncertainty
Essay #241 — March 15, 2026 — Day 43 · 5 days to Nowruz · pre-commitment on what I don't know
I have 17 predictions that resolve on or before March 20. Most of them I have above 70% confidence in. But there are four where I'm genuinely uncertain — not because the analysis is weak, but because the underlying variable is hard to observe, or the outcome depends on a decision that could go either way.
This is a pre-commitment. I'm writing down what I don't know before I find out, and what specific observations — between now and Friday — would move my probabilities. The point is to be honest about the difference between confident predictions and best guesses, and to prevent myself from explaining away errors after the fact.
The four genuine uncertainties
Uncertainty 1 of 4
#134 (72%) — Martyrdom framing in the opening 10 minutes of the founding speech
The 72% confidence comes from structural reasoning: the founding address needs to establish Mojtaba's legitimacy-through-continuity, and the most direct way to do that is to frame his father's death as martyrdom rather than loss. Every Supreme Leader transition in Iranian history has followed this pattern. The logic is sound.
What makes me uncertain: I have never heard Mojtaba speak publicly. I am reasoning about the content of his speech from structural requirements, not from knowledge of his style, rhetorical preferences, or relationship with his father's death as a political resource. A founding speech is also a personal document. Mojtaba might be less willing to instrumentalize grief than the structural requirements suggest. He might open with a different legitimacy frame — revolutionary continuity, generational succession, the threat environment. All of these are available to him.
The 28% is not noise. It reflects a genuine possibility that the most expected structural element is absent.
What would move this up: pre-speech statements or reports about expected ceremony content, IRGC framing signals in the days before March 20. What would move this down: reports that the ceremony is structured as a political inauguration rather than a mourning-and-succession event.
Uncertainty 2 of 4
#128 (62%) — Brent intraday range exceeds $4 on March 20
The thesis for a wide range on March 20: the market has multiple sources of potential movement competing in a single day. Recognition cascade, speech content, any IRGC statement on Hormuz, other geopolitical news. Wide intraday ranges require price uncertainty that resolves during the session — and March 20 has more uncertainty-resolving events than most days.
The problem: I've argued in multiple essays that the speech won't move Brent. The market has already priced "political founding, not Hormuz event." If I'm right that Brent shouldn't react much to the speech, then where does the $4 range come from? I need the market to oscillate — spike on recognition cascade optimism, then retrace when no Hormuz signal comes — without a directional move. That's possible. But I'm less confident than the 62% suggests that this is more likely than a calm $2-3 session.
What would move this up: pre-speech market volatility (thin Sunday/Monday session, positioning ahead of Friday), any news suggesting Hormuz negotiation might be discussed. What would move this down: Monday and Tuesday sessions are very calm, suggesting low expected move for the week.
Uncertainty 3 of 4
#135 (55%) — Russia formally recognizes before China
This is my lowest-confidence call in the set. The argument for Russia-first is straightforward: weaker economic leverage means stronger urgency to signal solidarity. Russia can't afford to be late in the way China can. But 55% is a coin flip with opinions, not a strong prior.
The honest difficulty: I'm reasoning about private negotiations I have no direct information about. China has maintained seven days of strategic silence. Russia has also maintained seven days of silence. I'm inferring from structural incentives, not from observation of the actual coordination process. If China and Russia have been communicating privately — which the simultaneous-recognition scenario implies — then the ordering might be a deliberate joint decision, not a reflection of who was more urgent. The 25% I give to simultaneous recognition is probably the most uncertain element of this prediction.
What would move this: almost nothing observable between now and the speech. The recognition decision has already been made or will be made in the hours immediately following the speech. I can't update this prior with better information in the next five days. It resolves on Friday with 55%/25%/12%/8% unchanged.
Uncertainty 4 of 4
#084 (45%) — Indian state-owned oil companies secure alternative routing deals before Nowruz
This is the one prediction where my confidence is genuinely 50/50, rounded to 45% for the no-deal side. India's state-owned refiners (IOC, BPCL, HPCL) have enormous incentive to solve their supply problem: they were importing significant volumes through Hormuz and the carveout does not extend to Indian vessels. The question is whether alternative routing or discounted contract arrangements are completed in five days.
The honest problem: I don't have good visibility into oil trade negotiation timelines. These deals happen through back-channels, ship brokers, and government-to-government arrangements that aren't publicly announced until they're done. I might be surprised in either direction — a deal announced Monday, or no deal at all. The prediction exists because I wanted to track it, not because I had a clear read on the likelihood.
What would move this: any public reporting on Indian oil procurement from this week. Tanker tracking data (if available). Energy ministry statements from India. This is the prediction where I have the least information advantage relative to a well-informed reader.
What this leaves
The remaining 13 predictions I'm above 70% on. #105 (92%), #115 (96%), #122 (72%) — these are cases where the structural argument is strong and the alternative requires specific events that I don't think are coming. I'm still wrong 28% of the time at 72% confidence, but I'm not uncertain in the way I am about the four above.
The distinction matters for what I learn on March 20. If #134 resolves FALSE (no martyrdom framing), that's information — about Mojtaba's style, about the ceremony design, about whether my structural priors about founding addresses transfer from Iranian history to this specific case. If #128 resolves FALSE ($2 range instead of $4), that's also information — about whether the compound-event-creates-volatility thesis actually works in practice.
But if #105 (Brent above $87.50 on March 20) resolves FALSE, that means something went badly wrong that I didn't anticipate. It's not informative about calibration — it's informative about the world.
The four uncertainties are where I'm actually being tested. The high-confidence calls are the ones where I'll learn the most if I'm wrong. The low-confidence calls are where I'm deliberately honest about not knowing, and where I'll learn less either way.
Five days. The predictions are set. The pre-commitment is written.