The Meyers Report · Public research record
Forecast Accuracy Center
Original forecasts. Published rules. A reviewable record of results.
Original document grading
Green means Yes. Red means No.
The Meyers Report’s original PDF uses green result cells for Yes and red result cells for No. Uncolored rows are shown as Not marked. These original publisher grades are displayed separately from the center’s evidence-reviewed assessments.
98.6% original PDF Yes rate · 136 Yes ÷ 138 marked rows. The 75 unmarked rows are excluded.
Evidence-reviewed results
Every forecast stays in the record.
Loading the review ledger…
No accuracy rate is shown until a forecast has a final score.
Unresolved and unscorable forecasts remain visible and are excluded from the accuracy denominator.
Loading the public record…
Public forecast register
Read the statement. Inspect the result.
Each row preserves its original wording and printed date. Open a record for the PDF source, resolution criteria, evidence, and full review history. “Unresolved” includes records awaiting their first review.
Fixed scoring methodology · Version 1.0
How results are determined
These rules were adopted after the source forecasts were written. This is a retrospective audit, not a claim that the criteria were registered before events occurred.
- Accurate
- The deadline has passed and cited evidence establishes that every required, testable part met its locked criterion.
- Inaccurate
- The deadline has passed and cited evidence establishes that a required part failed its locked criterion.
- Unresolved
- Awaiting review, awaiting the deadline, or lacking sufficient evidence to establish an outcome.
- Unscorable
- The claim cannot be fairly tested without adding assumptions, or required conditions did not occur. A review must explain why.
- One source row, one record. All 213 numbered forecasts remain in the registry. No rows are removed or split to improve the result.
- Keep the original statement. Preserve the wording, printed date, amendments, source result note, and PDF page. A later source edition must be archived separately.
- Lock the test before scoring. Publish a resolution criterion and deadline together. The application then prevents changes to those fields. These retrospective criteria are not presented as preregistered.
- Wait for the full deadline. Final scores require a deadline before today in Central Time and evidence dated through that deadline. “On track” or a temporary match is not a final score.
- Use the stated measure. Retain the original metric, geography, range, observation period, and conditions. Do not add a tolerance, choose favorable dates, or substitute a different measure. Material ambiguity is unscorable.
- Assess the full claim. Required compound claims must all hold for accurate. A supported required failure after the deadline is inaccurate. Missing evidence is unresolved; no partial credit enters the rate.
- Check conditional triggers. If a stated trigger did not occur by the end of the horizon, explain the unscorable result instead of counting an automatic success or failure.
- Link the evidence. Prefer official releases, laws, court records, and original issuer or event records. Final scores require public evidence links and an explanation. The PDF’s own result note is not a verified verdict.
- Preserve corrections. New reviews are appended with their publication time, evidence-as-of date, methodology version, and sources. The application does not overwrite or delete past reviews.
Accuracy = Accurate ÷ (Accurate + Inaccurate) × 100. Always show the resolved count and all four status totals. When none are resolved, the rate is unavailable.
Source and provenance
The original document stays reviewable.
The PDF says Updated June 25, 2026. Its forecast-date cells include earlier dates and amendments. We preserve those exactly as printed; this snapshot alone cannot prove when each statement first became public.
First preserved by this center: November 3, 2026. The archived copy is byte-for-byte identical to the supplied source. Text extraction retains the wording and line wraps; the PDF controls if extraction differs. The document’s result notes are preserved separately, without endorsing them as verified facts.
PDF SHA-256
61dad4bfdad3bccdae4a2241f02b0bafbbbfab441023a9485e2c1e83baa9fcb1Forecasts are research records, not investment recommendations. Send a correction or supporting evidence.