Skip to main contentChartCompass
Menu

All Insights · market-education

Explaining market moves without confusing correlation and cause

A practical evidence ladder for describing what moved, what changed nearby, and how much confidence a causal explanation deserves.

By ChartCompass · Fri, 04 Sep 2026 07:34:33 GMT · 6 min read

A sharp price move invites a tidy story. The problem is that markets absorb many pieces of information at once, while the price chart records the result rather than the reason. A headline beside a candle is evidence of timing. It is not, by itself, proof of cause.

A useful explanation therefore begins with an observation and earns its confidence one step at a time. It should tell the reader what is measured, which information was actually available, what else moved, and which competing explanations remain plausible.

Start with the event, not the story

Describe the move before interpreting it. Record the instrument, exchange session, comparison point, size of the move, volume if available, and the timestamp and timezone. A move from the previous close, a move from the opening print, and an intraday reversal answer different questions. Extended-hours trading may also separate the market reaction from the regular-session candle.

Then establish a release timeline. Scheduled statistics have an official publication time. The U.S. Bureau of Labor Statistics, for example, publishes a calendar in advance and explains that its principal releases are distributed at designated times. Company filings have acceptance timestamps; central-bank statements and press conferences have known windows. Use those records rather than the time a secondary article happened to appear.

This order matters because a chart can appear to lead or lag a headline when the displayed timestamps use different timezones, when a feed is delayed, or when the market had already priced an expectation. The relevant surprise is often the difference between the outcome and what participants expected, not simply whether the published number rose or fell.

Build an evidence ladder

A disciplined market-move note can use four levels:

  1. Observed: price, volume, volatility, rates, currencies, commodities, and sector peers changed over a specified window.
  2. Contemporaneous: an identifiable filing, data release, policy statement, or verified report arrived during that window.
  3. Mechanically plausible: the information can affect expected cash flows, discount rates, risk premiums, or positioning in a way consistent with the move.
  4. Causally supported: a suitably narrow design separates the event from other news and shows that the measured surprise, rather than the scheduled announcement alone, explains the response.

Most everyday explanations should stop at level two or three. That is not a weakness. “Shares fell after guidance was reduced, while peers were little changed” is often more accurate than “guidance caused the fall.” The second statement requires stronger identification.

Federal Reserve researchers illustrate why narrow windows help. High-frequency event studies examine asset prices close to a policy announcement to reduce contamination from unrelated developments. Even then, a central-bank communication can contain several kinds of information: a policy-rate surprise, news about the reaction function, and information about the central bank's view of the economy. One label such as “hawkish” may hide those distinct channels.

Test rival explanations

Before attaching a cause, ask what the chart would look like if the preferred explanation were wrong.

  • Did the whole market, the sector, or only the company move?
  • Did interest rates, foreign exchange, or a relevant commodity move at the same instant?
  • Was there another filing, court decision, regulatory notice, analyst action, or macro release in the window?
  • Did the price begin moving before the supposed catalyst became public?
  • Was the move a continuation of an existing trend or an abrupt break?
  • Did volume and options activity confirm unusual attention, or was liquidity thin enough to amplify a small order imbalance?

A peer-relative move supports a company-specific explanation more than a broad risk-off session does. A refinery stock rising alongside product margins has a different evidence profile from every cyclical stock rising after a rate surprise. Neither pattern proves causality, but each changes which hypotheses deserve attention.

Treat language as part of the method

Confidence should be visible in the verbs. “Coincided with” reports time. “Was consistent with” adds a plausible mechanism. “Investors focused on” needs direct evidence such as management commentary, a filing, or broadly documented market reporting. “Caused” should be rare.

Avoid reverse-engineering a narrative from price direction. Good news can accompany a decline if expectations were higher, if forward guidance deteriorated, or if discount rates moved. Bad news can accompany a rise if positioning was defensive or the outcome was less severe than feared. Price is not a vote on the moral or absolute quality of a release; it is the clearing result of orders under existing expectations and liquidity.

Social posts and sentiment summaries can identify questions to investigate, but they are weak foundations for factual claims. Investor.gov warns that social sentiment tools can reflect inaccurate, incomplete, or misleading material. Trace important claims back to an issuer filing, official release, regulator, exchange, or clearly attributable reporting.

A reusable explanation template

Write a short explanation in this sequence:

  • Move: what changed, over which exact window, and relative to what benchmark.
  • Dated evidence: the primary release or filing and its publication time.
  • Mechanism: which cash-flow, discount-rate, risk, or positioning channel could connect the evidence to the move.
  • Cross-check: relevant peers, market factors, volume, and other assets.
  • Alternatives: at least one reasonable competing explanation.
  • Confidence: observed, plausible, or strongly supported.
  • Next evidence: the filing, transcript, later data, or market normalization that would strengthen or weaken the account.

This format turns an explanation into an auditable research note rather than a memorable story. It also makes later review possible: when new filings or revised data arrive, the original timestamps and assumptions remain visible.

Limitations

Public timestamps can be delayed, corrected, or displayed in different timezones. Consensus expectations may be incomplete, peer baskets can contain mismatched exposures, and price reactions can reflect private positioning that public evidence cannot reveal. Even a narrow event window does not prove that every trade responded to the same information. AI-generated summaries can omit alternatives or state a plausible narrative too confidently.

This framework is educational and does not identify what any person should buy or sell. Open ChartCompass to inspect a selected move beside dated market context, source status, and explicit evidence limits.

Sources

Return to Insights