Thirty minutes is the window this site measures, because it is long enough to capture a reaction and short enough to belong to it. This asks what happened next: how far the Nasdaq 100 sat from the release price two hours on, against where it sat at the half hour.
| release | at 30 min | at 2 hours | ratio | is it over? |
|---|---|---|---|---|
| FOMC | 0.20% | 0.58% | 2.90x | No - more than doubles |
| Core PCE | 0.17% | 0.42% | 2.53x | No - more than doubles |
| Retail sales | 0.11% | 0.27% | 2.41x | No - more than doubles |
| JOLTS | 0.15% | 0.36% | 2.36x | No - more than doubles |
| GDP | 0.13% | 0.23% | 1.86x | No - roughly doubles |
| Payrolls | 0.38% | 0.64% | 1.68x | No - keeps drifting |
| CPI | 0.28% | 0.45% | 1.58x | No - keeps drifting |
| PPI | 0.24% | 0.35% | 1.47x | No - keeps drifting |
On the current record no release is finished at thirty minutes - every one of them sat further from the release price two hours later, by between 1.47 and 2.90 times. Both columns are the median distance from the release price, ignoring direction - a ratio above one means the price was further away later, not that it kept going the same way. Drift over two hours also collects everything else that happened in the session, which a thirty minute window largely excludes.
How far the Nasdaq 100 sat from the release price at the thirty-minute mark, and how far it sat at the two-hour mark, ignoring direction in both cases. Both are medians across the record, in percent. The ratio is the second divided by the first: above one and the index was further from the release price two hours out than it was at the half hour, whichever way it went. Thirty minutes is the window this site is built around, so this page is the honest check on that choice: it asks what the choice leaves out.
On the current record no release is finished at thirty minutes. Every row sits further away at two hours, by between 1.47 and 2.90 times. FOMC drifts furthest, 2.90 times, and the reason is the press conference: the statement prints at 14:00, the chair starts speaking at 14:30, and the second hour of the window is the market repricing the first hour. For FOMC the thirty-minute mark is closer to the middle of the story than the end.
The 08:30 releases have a different second event inside their window, and it is the same one every time: the cash open at 09:30. Two hours after an 08:30 print is 10:30, so the two-hour reading for CPI, payrolls, PPI, GDP, Core PCE and retail sales all include the first hour of the regular session, with its own flow and usually its own 10:00 data. Part of the drift on those rows is the release still being digested, and part of it is simply the market opening. PPI drifts least at 1.47 times; Core PCE and JOLTS more than double.
The thirty-minute figure is the number I size to, because it is the part of the move that belongs to the release and to nothing else. The two-hour figure is the number that stops me from expecting the story to be over. On FOMC that means I do not judge a position at 14:30, because at 14:30 the larger event has not started. On an 08:30 release it means I treat the open as a second release with its own risk, rather than as the release's echo. Those are two different reasons for the same shape of number, and the table does not distinguish them; the clock does.
Whether the move continued or reversed. The distance ignores sign, so a print that ran a percent up and then a percent and a half down shows a ratio above one on this page and a reversal on the page about the first five minutes. Two hours also collects everything else that happened in the session, which the thirty-minute window mostly excludes, and there is no way in this data to say how much of the drift was the release. It is not a recommendation to hold for two hours. It is a warning that thirty minutes is not the end.
Releases: Core PCE Price Index Consumer Price Index FOMC Rate Decision GDP JOLTS Job Openings Nonfarm Payrolls Producer Price Index Retail Sales