Macro or Noise

Release log

A running record of what each U.S. economic release reported and what the market did that day. Newest first. Every entry is measured, not remembered.

Each entry is generated from our own measurements by a rule-based script, not written by hand and not by a language model. Nothing here is a forecast. Updated 2026-08-08.

How these entries are made

Each one is assembled by a rule-based script from our own database — no language model, no hand-editing, and no entry that says anything the measurements do not support. The script fills in four things:

  1. What was reported, and how it compared with the previous reading. Figures come from the issuing agency, on the basis that agency uses for its headline.
  2. What the market did that day — the change from the prior close in whichever market that release matters most for (Treasury yields for most, the dollar for PCE).
  3. How that compares to normal, measured against the average daily move in that same calendar year rather than a fixed baseline, because a normal day in 2013 and a normal day in 2026 are not the same thing.
  4. Where it ranks against every release of the same kind since 2000.

Points 3 and 4 use different denominators, which is why an entry can read "quieter than an ordinary day (×0.5), larger than 48% of all PPI releases since 2000" without contradicting itself. The ratio compares the move to a normal day in its own year; the ranking compares the raw move to every release since 2000, unadjusted. In a volatile year a move can be small relative to that year and still sit mid-pack against 25 years of history. Both are in the glossary.

There is deliberately no interpretation of why a market moved, and no suggestion of what comes next. On any single day a hundred things move prices, and one release is only one of them. What we can say honestly is how large the move was and how unusual that is — nothing more.

Related

Historical statistics for informational purposes only, not financial advice.