Every prediction we make is scored in public.
PLUS logs falsifiable forecasts — a claim, a metric, a stated confidence, and a horizon date — before the outcome is known. When the horizon arrives, the outcome is fetched from the metric’s official source and the verdict is written here. The ledger is append-only: corrections are published as errata, never deleted.
The prediction ledger
Ledger empty
No predictions have been published yet.
When PLUS publishes its first falsifiable forecast, it appears here before its outcome is known — and stays here after, whichever way it resolves. An empty ledger is the honest starting point, not a placeholder for an invented one.
Why we publish this
Intelligence you can’t audit is just opinion.
Anyone can publish analysis; almost no one publishes their misses. NirmanKit PLUS briefs include forecasts, so the only honest way to sell them is to keep score where you can check it. A prediction can only be scored against a metric from the approved catalog, by an automated resolver, from the metric’s official source — the analyst doesn’t grade its own homework.
Where the forecasts come from
From published briefs, not thin air.
Each prediction is attached to a human-approved intelligence brief analysing a real government decision. The briefs — and the operations tools built on them — are what a PLUS subscription buys. This ledger stays free either way.