When Markets Move

Know which days are going to be rough.

Every Sunday, one page on the economic announcements coming that week — and what markets actually did the last two dozen times each one landed.

Real numbers from the briefing. Pick an announcement:

Turbulence, not fortune telling

Those bars say how often each market made a move of half a percent or more on that announcement's day, across the last two dozen occasions. Bonds react to almost every Fed decision. The euro mostly ignores them. Knowing that before Wednesday afternoon is worth more than another opinion about where prices are heading.

In every issue

  • The week's announcements, with the consensus forecast and last month's number.
  • How six markets behaved on the last 24 occurrences of each one.
  • Last week in review — what actually happened, against what usually happens.
  • Live prediction-market odds on rates, inflation and recession.
  • World events with a plausible link to oil, gas and currencies.

Never in any issue

  • Buy or sell calls. Nothing here tells you what to do with your money.
  • Direction. Public data supports how much and how often, not which way.
  • Cherry-picked track records. Every number is recalculated weekly and can be checked.

Where the numbers come from

Nothing in the briefing is proprietary, and that's deliberate — every figure traces back to a source you can check yourself.

Federal Reserve Economic Data

The exact publication date of every inflation, jobs, growth and spending release, straight from the St. Louis Fed's own database.

Daily price history

Four years of closing prices for the dollar index, gold, the S&P 500, the Nasdaq 100, 10-year Treasury yields and EUR/USD.

The economic calendar

What's scheduled in the week ahead, with the consensus forecast and the previous reading for each announcement.

Public prediction markets

Live odds on rate decisions, inflation and recession — priced by people with real money at stake, not by commentators.

How it's put together

For each announcement, we take its last two dozen publication dates and measure what those six markets actually did on each of those days. That produces two numbers per market: the average size of the move, and how often it exceeded half a percent. Compared against a normal trading day, that's the whole method.

Two limits worth stating plainly. The measurements are close-to-close, so a sharp move that fades by the closing bell is understated. And two dozen occurrences is enough to see a pattern, not enough to bet a mortgage on — which is why the briefing reports frequencies and never predictions.