# Gauge 2 · Barbell

**Published research summary · November 5, 2026.** Research status: all 12 registered Round-1 exports collected and validated; primary analysis unavailable under the frozen measurement rules. No headline effect or completed results JSON exists.

Barbell asks whether interest in luxury and budget fragrance is pulling apart alongside differences in household wealth and consumer sentiment. Its planned search baskets represent market tiers; they do not measure transactions, unit sales, or the incomes of the people searching. The registered comparison could not proceed because the collected search measures did not support the required mass basket.

The October 19 checkpoint is **12 of 12 verified Google Trends CSVs**: all four batches on October 14, 18 and 19, in America/Chicago. Exact bytes, immutable capture sidecars, hashes, registered settings and all 273 monthly observations have been validated. The collection and unlock requests are resolved. The observation range remains **January 2004–September 2026**, with United States, All categories, Web Search and literal Search terms. All scientific parameters and earlier captures remain unchanged.

The actual offline Trends run stopped at the registered quality gate. All 15 candidate terms trigger resolution review, but none has a stitched active-period median in the required 5–30 range for a mid-anchor, so the registered rule skips Round 2. The five mass terms have 8, 0, 26, 31 and 0 active months in their registered order; each falls below the required 36. The required mass basket therefore has no retained term. Required macro inputs, including Michigan income-tercile sentiment, are collected, but they cannot resolve this search-measurement limitation.

H2c and the other three primary tests were not run. Effects, confidence intervals, raw and BH-adjusted p-values, holdout performance and robustness results are unavailable. This is a measurement limitation, not a null finding or evidence that the proposed association is absent. No additional export is requested under the frozen plan; any redesigned study would need a separate registration.

**Public status copy:** “Barbell’s registered collection is complete. The search measures did not support a required basket, so the planned tests are unavailable.”

## Presentation of the unavailable result

Use a clearly labeled unavailable card without a numeric reading or a market-polarization claim. The quality-control diagnostics describe measurement coverage; they are not a search-spread finding or a substitute for the planned tests.

## Source and checkpoint note

Status above comes from `reports/gauge02-round1-diagnostic-2026-10-19.md` and its companion JSON, the actual-run record, and `reports/gauge02-collection-log.md`. The separate diagnostic reproduced byte for byte from the immutable captures; the actual pipeline also reproduced its failure checkpoint. Neither check establishes a completed end-to-end analysis or deterministic results JSON, because those results do not exist.
