Buy the Rumor, Sell the News Explained

A company announces record revenue. Its share price falls the next day. The headline looks positive, so why are investors selling?

“Buy the rumor, sell the news” describes buying ahead of an expected event, then selling when the announcement arrives. If a positive outcome was already reflected in the price, confirmation may give existing holders a reason to take profits without giving new buyers enough reason to pay more.

That is one explanation for a sell-off after good news. The saying becomes more useful when you compare what was expected, what actually changed and how the price responded. Three NVIDIA earnings announcements show how the same company can report strong growth and get very different market reactions.

What does “buy the rumor, sell the news” mean?

The “rumor” can be a widely discussed expectation: a strong earnings report, an anticipated interest-rate decision or a possible regulatory approval. It does not have to be an anonymous tip or an unverified social-media post.

The sequence has two parts. Traders first build positions ahead of the event because they expect a favorable outcome. After the announcement, some close those positions because their original reason for holding has played out.

Here, selling can simply mean closing an existing long position. Opening a short position is a separate decision: the fact that someone takes profits does not establish that the price will keep falling.

You may also see the British spelling, “buy the rumour, sell the news,” or the variation “sell the fact.” The underlying idea is the same: prices can move in anticipation of information before it becomes official.

Why do stocks fall after good earnings?

“Good” usually compares the business with its past. A market reaction also depends on the future buyers and sellers had already expected. Revenue can rise sharply from last year and still leave investors reconsidering how much they should pay for the stock.

The result may already be priced in

If investors expect rapid growth, some may buy before the report. By announcement day, the share price may already reflect much of that optimism. Confirmation alone does not necessarily raise their estimate of the company’s value again.

An analyst consensus gives you a measurable reference, but it is only one reference. Estimates differ by provider and update over time. Individual investors can have expectations above or below that average.

Avoid explaining every decline with an invisible “whisper number.” If you cannot document a higher expectation from before the release, you cannot use it as an established fact afterward.

The outlook may matter more than the completed quarter

A report contains several pieces of information. Revenue and earnings describe the period just completed; guidance describes management’s expectations for a future period. Margins, costs and demand commentary add more context.

A company can beat the revenue forecast for the completed quarter while giving weaker guidance for the next one. Imagine revenue of $10.3 billion against a $10.0 billion forecast, a 3% beat, alongside next-quarter guidance of $10.2 billion when analysts expected $11.0 billion, about 7% short. Calling that report simply “good news” misses the disagreement inside it.

Some holders may take profits

A trader who bought before an event may decide to exit once the uncertainty is resolved. Other investors may rebalance or reduce exposure for reasons unrelated to the announcement. Those orders can meet buyers who are willing to transact only at lower prices.

A price decline cannot tell you exactly who sold or why. Profit-taking is a possible mechanism, not something a red candle proves. Wider market conditions and other news can also affect the day’s result.

Three real examples: NVIDIA earnings, 2024–2025

Each example compares the regular-session close on the announcement date with the next day’s close. All three reports came out after the market closed, so the window includes the overnight reaction. It is not the return of a trade entered after reading the news.

February 2024: strong revenue, then a further rise

On February 21, 2024, NVIDIA reported quarterly revenue of $22.1 billion, up 265% from a year earlier, and projected next-quarter revenue of $24.0 billion, plus or minus 2%.

Its shares rose from $674.72 at that day’s close to $785.38 on February 22, a 16.40% gain. These are prices on the share basis used at the time, before NVIDIA’s June 2024 ten-for-one split.

A trader who shorted the release on the theory that “good news gets sold” would have been on the wrong side of that move. An announcement can add enough new information to push prices higher still.

August 2024: record revenue, then a decline

On August 28, 2024, NVIDIA reported quarterly revenue of $30.0 billion, up 122% from a year earlier. The next day, its shares closed at $117.59, down from $125.61 before the announcement: a 6.38% decline.

The release also contained information beyond headline growth. GAAP gross margin was 75.1%, compared with 78.4% in the preceding quarter, and management projected next-quarter revenue of $32.5 billion, plus or minus 2%.

Those figures show why you need the full report, not just the headline. They do not prove what caused the sell-off, and the price record alone cannot say how much of it was profit-taking. What it does show is strong reported growth followed by a falling share price.

November 2025: a rally at the open, a loss by the close

On November 19, 2025, NVIDIA reported quarterly revenue of $57.0 billion, up 62% from a year earlier, and guided to $65.0 billion, plus or minus 2%, for the next quarter. GAAP gross margin was 73.4%.

The next morning, the shares opened at $195.95, about 5.06% above the previous close of $186.52. By the close, they had fallen to $180.64, a 3.15% loss. The broader market reversed the same day: the S&P 500 ETF (SPY) opened higher and closed down about 1.5%.

This is the saying in its most visible form: the good news was confirmed, buyers paid up at the open, and sellers dominated the rest of the session. It still does not prove that profit-taking was the cause: wider worries about AI-stock valuations were also in the news that week.

Three cases cannot tell you how often each outcome happens. They do show that the observation window matters: the November reaction was +5% at the open and −3% at the close, so “how did the market react?” depends on when you look.

How can you tell whether news is already priced in?

You cannot read the market’s complete expectations directly from a chart. You can collect evidence that makes your interpretation more specific.

Start with a dated forecast. Save the consensus estimate, its provider and the time you recorded it. For earnings, check whether the figure is revenue, GAAP earnings or adjusted earnings, and which quarter it covers. Comparing different definitions can create a false “beat.”

Then examine the move before the event. Has the stock risen, fallen or stayed in a range? Compare the same dates with a relevant sector or broad index. A rise shared by the entire sector is weaker evidence of anticipation specific to one company’s report.

Finally, separate published expectations from your interpretation. “The stock rose before earnings” is an observation. “Everyone has already bought” is a claim you cannot establish from price alone. Tools for reading market sentiment can provide context, but no single reading settles what is priced in.

The aim is to define what would count as new information before you know the outcome. That makes the analysis harder to rewrite with hindsight.

A practical checklist for trading around news

Before the announcement: define the event and your exposure

Check the release time and time zone with the issuer or official publishing body. A calendar entry for an earnings call may refer to the discussion after the results, rather than the moment the figures become public.

If you already hold a position, decide whether your plan involves exiting before the release or accepting the risk of holding through it. An after-hours announcement can change the price before your instrument next becomes tradable.

If you have no position, write down the information and price behavior you would need to consider an entry. “The headline looks good” is too vague to evaluate later.

At the release: read the surprise and the outlook

Use the official release to compare actual results with the forecast you saved. Keep current-quarter results separate from next-quarter guidance. Look for material changes in margins or demand that the first headline may omit.

For economic releases, also check revisions to earlier figures. A strong new reading can arrive alongside a downward revision to the previous period. For a central-bank decision, the statement about future policy may change the interpretation of the rate decision itself.

There is no obligation to trade the first move. If you cannot reconcile the headline with the full release, keep observing until you can state what changed.

After the release: test the idea against the price

Choose an observation window in advance: for example, the first completed 30-minute candle of the next regular session, followed by that session’s close. Use the same window when reviewing comparable events.

A rise that falls back below a recorded pre-event level gives you different evidence from a rise that holds above it. Neither observation is a complete entry rule. A trade still needs a defined entry, invalidation point, exit and position size appropriate to the instrument.

Check the actual spread and trading hours. Fast moves and price gaps can produce worse fills than planned; a standard stop does not guarantee the exit price. Keep these execution costs in any strategy review.

Is “buy the rumor, sell the news” a reliable strategy?

The phrase leaves the crucial trading rules unspecified. It does not tell you which events qualify, how early to buy, when to sell, or how much to risk. Without those choices, there is no single strategy to test.

It also does not distinguish between taking a profit before the announcement, selling immediately afterward and opening a new short trade. Those approaches expose a trader to different price moves and execution conditions.

To evaluate a defined approach, record all qualifying events, including uneventful releases and cases where prices keep rising. Apply the same timing rules and include costs. Selecting only memorable reversals will make the saying look more reliable than the evidence supports.

For practice, choose one upcoming scheduled event and write down the forecast, your plan and your observation window before it happens. Review those notes after the window closes, even if you place no trade. The useful skill is comparing expectations with new information without rewriting your original view.