What the S&P 500 Does on Macro Data Days: CPI, Jobs and the Fed
Updated October 9, 2026
A release day is not a normal day#
Macro data comes out at a fixed time and packs into minutes what other days spread over hours. The table measures two things in ES, over the releases from October 1, 2021 to September 30, 2026: how much bigger the first hour after the release is than the same hour on a normal day, and how often the direction of the first five minutes was still standing at the close.
CPI, the inflation report, moves it the most: the median first hour is 4.0 times the normal range, over 59 releases. Next come the Fed decision, at 3.5 times over 40 meetings, and the jobs report, at 3.1 times over 57. Producer prices (1.8 times) and retail sales (1.6 times) move it much less.
It is a median: half of the CPI releases made the hour bigger than that and the other half smaller. Some release days barely move at all.
Macro data days in ES: how much the first hour widens and whether the first reaction holds
ES
Swipe to see the sample →
| Release (ET time) | First-hour range (× normal) | First 5-minute reaction holds at the close (%) | Releases |
|---|---|---|---|
| CPI (inflation) | 4.0 × | 65.5% | 59 |
| NFP (payrolls) | 3.1 × | 55.4% | 57 |
| FOMC (Fed decision) | 3.5 × | 60.0% | 40 |
| PPI (producer prices) | 1.8 × | 55.2% | 58 |
| Retail sales | 1.6 × | 49.2% | 59 |
n = 273 · 2021-10-08 to 2026-09-16 · exchange data
Releases from 1-Oct-2021 to 30-Sep-2026. Prior price = close of the minute before the release (8:30 ET; FOMC 14:00 ET). First-hour range = high minus low from the release minute to 59 minutes after it. Normal range = median of that same hour over the 20 previous sessions without any of the 7 releases (CPI, NFP, PPI, retail sales, GDP, PCE, FOMC); the cell is the median of range / normal range. First reaction = direction from the prior price to the close of the 5th minute; it holds if the 16:00 ET close is on the same side of the prior price (flat reactions excluded). ES front contract, CME 1-minute bars, no costs.
The weekly report, every Sunday
One email a week: what moved the macro picture, next week's calendar, rates and COT positioning, and the new study with its sample size. No signals; unsubscribe in one click.
Does the first reaction hold?#
On CPI, the direction of the first five minutes was still standing at the 16:00 close in 65.5% of releases with a reaction: two out of three. One of the 59 had no reaction and is left out of the percentage. On the Fed, 60.0%.
On the jobs report, 55.4%, barely better than a coin. On producer prices 55.2%, and on retail sales 49.2%. On those three releases, the first reaction says very little about how the session will end.
A direction that holds does not say how far price travelled or whether it was tradable: there are hours between minute five and the close, and a reaction that ends in your favour may have gone through a big pullback first.
Chasing the release after minute five#
A common idea is to wait for the release to "settle" and then trade with the move. To measure it, each release is classified by how bonds (ZN) moved in the first five minutes: a strong drop, no surprise or a strong rise, compared with the median of previous releases of the same report, without looking ahead.
The big reaction happens before minute five. On CPI, from the release to the close ES lost 0.65% on average when bonds fell hard (21 releases) and gained 0.44% when they rose hard (28). From minute five to the close, by contrast, what was left was +0.05% and -0.04%.
None of the minute-five-to-close averages is statistically different from zero, on CPI, jobs or the Fed. Several rows have fewer than 15 releases and are marked as small samples. The honest reading is that after minute five there was no reliable edge left to chase: most of the reaction had already happened.
Chasing the release: what ES did after minute 5, by how bonds reacted
ES
Swipe to see the sample →
| Release · bond reaction (ZN, first 5 min) | Release to close (mean) | Minute 5 to close (mean) | Up from minute 5 to close (%) | Releases |
|---|---|---|---|---|
| CPI · bonds down hard (yields ↑) | -0.7% | 0.1% | 52.4% | 21 |
| CPI · no bond surprise (small sample) | -0.0% | -0.1% | 50.0% | 10 |
| CPI · bonds up hard (yields ↓) | 0.4% | -0.0% | 57.1% | 28 |
| NFP · bonds down hard (yields ↑) | -0.3% | 0.1% | 63.6% | 22 |
| NFP · no bond surprise | -0.2% | -0.3% | 33.3% | 21 |
| NFP · bonds up hard (yields ↓) (small sample) | -0.2% | -0.4% | 30.8% | 13 |
| FOMC · bonds down hard (yields ↑) | -0.5% | -0.2% | 43.8% | 16 |
| FOMC · no bond surprise (small sample) | 0.2% | 0.1% | 50.0% | 14 |
| FOMC · bonds up hard (yields ↓) (small sample) | -0.0% | -0.4% | 50.0% | 10 |
n = 155 · 2021-10-08 to 2026-09-16 · exchange data
Releases from 1-Oct-2021 to 30-Sep-2026. Bond reaction = move of the 10-year note (ZN) from the minute before the release to the close of the 5th minute. 'Hard' = larger than the median absolute 5-minute ZN move over the PREVIOUS releases of the same kind since 2010 (no later release is used; the first 10 are unclassified). Bonds down = yields up. Minute 5 to close = from the close of the 5th minute to the 16:00 ET close, what was still ahead once the bond reaction was known. ES front contract, CME 1-minute bars, no costs. 'Small sample' = fewer than 15 releases.
How it is measured#
The reference price is the close of the minute before the release: 8:30 New York time for all of them except the Fed, which publishes at 14:00. The first-hour range runs from the release minute to 59 minutes later.
The normal range is the median of that same hour over the previous 20 sessions with none of seven releases: CPI, jobs, producer prices, retail sales, GDP, PCE and the Fed. The cell is the median ratio between the two.
The first reaction is the direction from the reference price to the close of the fifth minute. It holds if the 16:00 close stays on the same side of the reference price. ES front month, one-minute bars, no costs.
What this data does not say#
It does not say whether the release was better or worse than expected. The table measures the price reaction, not the surprise, and the same direction can come from good data one month and bad data the next.
The samples are short: 40 to 59 releases per report. Five years mostly cover one regime, the 2022 rate hikes and what followed, and the numbers may change in a different backdrop.
It does not subtract the spread or slippage either, which around a release are bigger than in a normal session.
How to use it#
As context when preparing the day: if there is CPI at 8:30, the first hour will probably be several times bigger than usual. That matters for position size and stop distance, not for direction.
And as a vaccine against a popular idea. The first CPI reaction holding two out of three times does not make the others a signal: on the jobs report and on retail sales the first reaction is close to a coin.
This content is informational material about historical data and is not personalized investment advice. Past results do not predict future ones, and trading futures carries a risk of loss.
Free PDF: 20 futures figures with their sample
The figures from our studies on one sheet, each with its n and date range (in Spanish for now). Subscribe to the weekly report and download it right away; the link also comes in the welcome email.
Frequently asked questions
- How much does the S&P 500 move on the inflation report (CPI)?
- In ES, the first hour after CPI had a median of 4.0 times the normal range for that hour, over 59 releases between October 2021 and September 2026. Half of the days moved more than that and the other half less.
- Does the first reaction to CPI hold until the close?
- In 65.5% of releases with a reaction, two out of three: the direction of the first five minutes was still standing at the 16:00 close. It does not say how far price travelled in between.
- Does it help to enter after the first minutes of a release?
- In ES there was no reliable edge left. Classifying each release by how bonds moved in the first five minutes, the average result from minute five to the close was not statistically different from zero on CPI, jobs or the Fed. The big reaction had already happened.
- What about the jobs report (NFP)?
- The first hour is 3.1 times the normal one, but the first reaction only held to the close 55.4% of the time, over 57 releases: barely better than a coin.
These numbers, instrument by instrument
See all 30 instruments →These base rates, on your chart
The indicator is open source and free: it marks each session's Initial Balance and gap, and prints beside them the historical rate at which that bucket fills, with its sample size. Same data as the tables above.
How to cite this data
The tables on this page are our own work, built from exchange data. You may reproduce them, translate them or build on them, including commercially, as long as you credit the source and link back to the original page. If you use them, tell us: we like knowing what they get used for.
Licence: CC BY 4.0
Suggested citation:
Perfiltrade (October 9, 2026). What the S&P 500 Does on Macro Data Days: CPI, Jobs and the Fed. https://www.perfiltrade.com/en/learn/sp-500-on-macro-data-days
Try these concepts with real data
7 days free, no card. An analysis tool, not investment advice.
See plans →