Tempora Research · Prediction Tracker
Tempora keeps a public log of every dated, testable prediction we publish. Each row carries a date, a specific claim, the result window, and — once the window closes — the check-back. If a system claims to read timing, it has to prove it.
A claim qualifies for the tracker when it is dated, testable, and posted before the result is known. Retrospective analyses and pattern pieces stay in the research archive — they are not tracked here.
Every claim carries a public receipt: the method, the charts used, the result window, the test that would prove it wrong, and the result once the window closes. Corrections appear as follow-up notes, not silent edits.
If a claim fails its test, the failure stays public. The method is in question, not the analyst.
A note on rollout (April 2026): Tempora has published 43 research pieces so far. The tracker carries the ten that are forward, dated, and testable. The rest are method and explainer pieces — they stay live in the archive but are not tracked here.
| Ref | Domain | Claim | Window | Status |
|---|
Every Monday
Every Monday at 09:00 IST, the tracker re-checks. Closed windows that haven't been written up yet get flagged. Live windows stay marked as live. Where a claim has a measurable test — a market level, a seat count, a date threshold — we pull the data feed and check it.
The page updates. The receipts update. Every change has a public history.