ground truth → forecasting the future
A prediction ledger on geopolitics, markets, and conflict — forecasts with sources, scoring, and receipts.
A handful of contrarian calls — and a few we got wrong. Browse the full ledger for sources, audit threads, and verdicts on every entry.
When Bitcoin was near $50k and analysts were calling $250k by year-end, this called a peak-to-trough drawdown of at least 30% within 3–9 months and a new ATH thereafter. BTC fell ~50% by July 2021, then rallied to a fresh ATH of ~$69k that November.
When Biden's team was actively negotiating in Vienna and the consensus expected a return to the Iran nuclear deal, this called no US–Iran agreement during his term. None was reached by 20 January 2025.
At peak war-escalation panic — Trump's “Power Plant Day” ultimatum on the table — this called Rung D (no new civilian-infrastructure strikes) as the modal outcome at 48%. The window resolved with the Pakistan-mediated US–Iran ceasefire on 8 April.
Anchored on his Russia stance becoming a focal point, this predicted Trump would either be denied the nomination or beaten in the general. He won both, decisively.
Picked Pennsylvania's governor as the most likely VP nominee on swing-state and demographic grounds. She picked Tim Walz instead.
Long-form analyses, freshest first.
What Danny Citrinowicz, John Mearsheimer, and Behnam Ben Taleblu — three analysts with very different instincts — imply about the odds of a real US–Iran settlement.
What Islamabad will, and will not, produce — and why the real test is Lebanon and Hormuz, not the centrifuges.
Thirty-eight days of operational data show Iran's short-range missile campaign sustaining well above 20 launches per day — a resurgence the “90% destroyed” narrative cannot explain.