Surprise size matters more than the headline in economic data

[Amanhasnoname]

The market doesn't react to the number. It reacts to the difference between the number and what was expected.

If payrolls come in at 200,000 against a forecast of 190,000, that's a small beat. If the forecast was 120,000, it's a huge one. Price movement scales roughly with the size of the surprise, relative to the typical error of forecasts for that release.

That's why it helps to know the usual range. Some releases miss consensus by a big margin regularly, so a modest miss means little. Others rarely deviate, and a small gap can be meaningful.

Revisions to prior data matter as well. A headline beat with a large downward revision can produce a confused reaction.

For scalpers, preparing means writing down the forecast, the previous figure and what size of surprise would count as large, before the release.

It makes the first seconds after the data less chaotic and helps avoid reacting to noise.

How do you prepare for releases: with scenarios, or by reacting?