WhitmanTrading

How to Trade the News

To trade unscheduled news, accept that the initial move happens before a human can read the headline, and trade the structure that forms afterwards instead. The first report is also frequently incomplete or wrong, which the second wave of price action reflects.

Unscheduled news is a different problem from a scheduled release. There is no forecast, no priced expectation, and no warning — and the initial move is made by participants reading the same words far faster than any person can.

Before you start

An honest assessment of how fast your information actually reaches you. Consumer news feeds arrive after the market has already moved, which is a fact about the technology rather than a criticism.

A rule for unverified reports, because the first version is frequently wrong. Early accounts get corrected, and the correction moves price again.

An acceptance that by the time you have read it, it is priced. That is not defeatism; it is the starting point for a workable approach.

The steps

1. Accept the first move is gone

A range-bound stretch of price with a sudden displacement.
The initial move happens faster than reading. Illustrative chart - not real market data.

Automated systems act on headline text in milliseconds. The gap between an event and a human reaction is not a skill difference and no amount of preparation closes it.

2. Establish what is actually confirmed

A slice of price data with an uncertain interpretation.
Reported and confirmed are different things. Illustrative chart - not real market data.

An early report is often partial. Whether it is confirmed, by whom, and whether the detail has changed since the first version is the substance of the event.

3. Wait for the second wave

A long-horizon price series with a secondary reaction.
The correction or confirmation is its own move. Illustrative chart - not real market data.

The initial reaction, then a reassessment as details arrive. The second move is slower, involves human judgement, and is the first point at which you are not structurally behind.

4. Let a range form before taking a view

A slow-moving stretch of price settling into a band.
Structure is what makes a trade possible. Illustrative chart - not real market data.

A high and a low made on real volume after the event gives you levels. Before that there is a fast market with nothing to trade against.

5. Expect the spread to stay wide

The first half of a price series with elevated costs.
Wide for hours, not for the headline. Illustrative chart - not real market data.

On this site’s shared series a round trip measures about 2% of the median bar range of 0.493, and it stays elevated long after the initial move while participants reassess.

6. Size for a second surprise

A section of a price series with a further displacement.
Developing stories develop again. Illustrative chart - not real market data.

A developing story can produce another move at any moment. Reduced size is the response, because there is no schedule telling you when the next update arrives.

7. Do not hold through an unresolved story overnight

The first half of a price series across a session boundary.
An unresolved story plus a closed market is a gap. Illustrative chart - not real market data.

An unresolved situation and a closed market is the specific combination that produces a gap through your stop, and only position size limits what that costs.

How to tell it worked

0 trades were taken in the first 10 minutes after a headline appeared.

The report’s confirmation status was checked before any position.

A range had formed before a directional view was taken.

And position size was below normal for the whole of a developing story.

Why you cannot be first

A candlestick chart annotated with the round-trip cost of a switch.
Chasing a headline pays the widest spread available. Illustrative chart - not real market data.

Machine-readable feeds exist and are used. Systems parse headlines and place orders without a human involved, and they are reading the same words you are about to read.

A section of a price series drawn without volume context.
And the first seconds are the thinnest book of the day. Illustrative chart - not real market data.

Which makes chasing the initial move the worst available trade. Widest spread, thinnest book, and an information position that is behind by construction rather than by effort.

Scheduled against unscheduled

A scheduled release has a forecast, and the expected move is priced in advance. You know the time, you know what is expected, and you can decide your position days ahead.

Unscheduled news has none of that. No forecast to react against, no premium priced beforehand, and no opportunity to have decided anything.

Which is why the approaches differ. For a scheduled event the useful decision is made in advance; for an unscheduled one there is no “in advance”, so the only workable version is to trade what happens after rather than what happens at the moment.

What the second wave actually is

Participants who read the whole story rather than the headline. The initial move responds to a sentence; the reassessment responds to detail, scope and whether the source is credible.

It is slower, larger in volume, and it is where the durable direction is set. The first move is frequently in the right direction and frequently overshoots, and the second corrects that.

Which is why waiting is not passivity. You are waiting for the phase where the price is being set by people making judgements rather than by systems matching keywords, and that is the phase your own judgement can compete in.

The cost is that you enter after part of the move. That is the trade being made — the part given up happened in a market with the widest spread of the day and no structure, so what is surrendered is mostly the least tradeable portion of it.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, 1 mentions news trading in the title, at 42,108 views — and it is not instruction-shaped. Central bank decisions appear in 7 at 1,723 and economic calendars in 0. The counts come from site/corpus_count.py.

A candlestick series with several gaps, the largest of them marked.
Most unscheduled news lands outside trading hours. Illustrative chart - not real market data.

1 video in 24,971, at 42,108 views. A single piece of coverage with a large audience, and the entire event-trading cluster — news, calendars, central bank decisions — totals eight videos across the whole corpus.

A stretch of price bars cut short at a decision point.
Headline just broke and price is running. Join it? Illustrative chart - not real market data.

The answer to the question on that chart is that the move you can see is the one already made. Entering there pays the widest spread of the day to buy after the informed participants have — and the early report may still be corrected, which moves price again.

When it fails

The failure is trading the correction as though it were the news, and it happens on developing stories. An initial report moves price sharply. A position goes on in that direction. Twenty minutes later the report is amended — the figure was different, the source was wrong, the scope was narrower — and price returns most of the way. The trade was taken on information that was accurate at the time and provisional by nature, which is what an early report always is.

The second failure is chasing the initial move. It is already made.

A third is normal position sizing. A developing story develops again.

A fourth is holding an unresolved story overnight. That is the gap case.

A fifth is trusting a consumer feed’s timing. It arrives after the market moved.

And a sixth is expecting an expected-move premium. Unscheduled news has none.

Gap trading covers news landing outside trading hours. Premarket and after hours is where most of it lands. And implied volatility is what a scheduled event has and this does not.

What I actually do

The honest position is that I am never first and cannot be. By the time a headline is on a screen I can read, automated participants have acted on the same words. What is left is the second phase — the range that forms once the initial reaction settles — and that one is an ordinary trade.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.