September 2026: the pre-registered New-Moon low test
A prospective, pre-registered test of the New-Moon / local-low correspondence, registered before the window opened and judged by rules fixed in advance.
The registered record
- Observation time
- 11 September 2026, 03:26 UTC
- Hypothesis
- A strict local low forms in the window T0:T+3 following the New Moon (11-14 September 2026 inclusive).
- Prediction
- If a qualifying low forms, measure the maximum upside at exactly 7, 14 and 21 days from the pivot. The New-Moon geometry strength recorded in advance is context only and is not evidence that a low must occur.
- Confidence / uncertainty
- Moderate on the direction of the lead, low on its timing. In 2026, 7 of the first 8 New Moons had a strict local low in T0:T+3; full-history behaviour is regime-dependent.
- Objective test criteria
- A strict 7-day pivot on the daily LOW: a day t in [T0, T+3] qualifies when LOW[t] < LOW[t-1], LOW[t-2], LOW[t-3] AND LOW[t] < LOW[t+1], LOW[t+2], LOW[t+3]. Strictly less-than on both sides; ties do not qualify. Defined on intraday lows, not closing prices. A qualifying pivot is not invalidated by a later lower low.
- Invalidation criteria
- No day in T0:T+3 satisfies the strict 7-day low pivot. A lower low forming later, inside or outside the window, does NOT invalidate a pivot that has already qualified - it is recorded separately as context.
- Technical context
- New Moon close $77,173.80. Lowest close inside T0:T+3 was $76,838.16 on 2026-09-13.
Predictions in this experiment
Outcomes are recorded as Consistent, Partially Consistent, Inconsistent,
Invalidated or Inconclusive. We deliberately do not reduce this project to a
win percentage — the numeric data is preserved for later statistical testing.