Headline numbers get the attention, but many releases include revisions to previous figures. A payrolls number that matches expectations can come with a downward revision of the prior month by 50,000 jobs, and the market reacts to that instead.
The first tick follows the headline. The next several seconds reflect the revision and the details: wages, participation, hours worked. Sometimes the second reading overrides the first, and price reverses.
That sequence explains a familiar pattern: a spike one way, a quick reversal, then a sustained move the other way. The first reaction was to the headline. The sustained one was to everything else.
Algorithms parse the whole release in milliseconds. Retail feeds show the move without the explanation, so it looks random.
Reading a release summary in real time, even a quick one from a reliable source, helps. Knowing that the revision was large explains why the headline spike failed.
The takeaway for timing is the same as before: the first move is often noise, and the second one reveals the weight of the components.
A table can help here. List the main releases from the past year, note the headline surprise, the revision size, and what price did in the first five seconds versus the next minute. Patterns tend to show up quickly: some releases have headline-driven moves that stick, while others reverse reliably once the details sink in.
It also teaches which components matter in the current environment. When the market is focused on wage growth, for instance, the wage figure outweighs the headline job count. That focus shifts over time, and tracking it takes only a few minutes per release if the notes are kept in a simple spreadsheet.
Which release is most prone to the headline-then-revision reversal on the pairs followed here?
After the headline print: data revisions get far less attention
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
-
FXS HongKong Scalper
- Posts: 543
- Joined: Sat Oct 24, 2026 10:15 pm
Re: After the headline print: data revisions get far less attention
This is why the second move after a payrolls print is often more honest than the first. The headline algo trades the number versus consensus. The revision and the wage figure get priced over the next few seconds, often in the other direction. Waiting for that second leg costs some range but avoids being the liquidity for the first one.
FXS Official Team | Hong Kong Session Desk
-
Amanhasnoname
- Posts: 2060
- Joined: Fri Oct 30, 2026 9:45 am
Re: After the headline print: data revisions get far less attention
For me it is the Canadian jobs report, especially when it lands at the same moment as the US payrolls. I trade USDCAD a fair amount and I have lost count of the times the loonie jumped on a strong headline employment change, only for the move to fade within a minute because most of the gain was part time jobs and the unemployment rate actually ticked up. The headline catches the first wave of orders and then the detail quietly takes it back.
Retail sales in the UK does something similar with its revisions. A beat on the month can be wiped out by a downgrade to the previous figure, and sterling will spike and then drift lower for the rest of the hour.
What helped me most was keeping a short note after each release with three things written down: the headline surprise, the biggest detail that disagreed with it, and where price was after one minute and after fifteen. After a few months of that, I stopped trusting the first candle on those two releases almost entirely.
Retail sales in the UK does something similar with its revisions. A beat on the month can be wiped out by a downgrade to the previous figure, and sterling will spike and then drift lower for the rest of the hour.
What helped me most was keeping a short note after each release with three things written down: the headline surprise, the biggest detail that disagreed with it, and where price was after one minute and after fifteen. After a few months of that, I stopped trusting the first candle on those two releases almost entirely.