Fundamental Analysis vs Price Action: What Earnings Days Do to the Chart

Michael Whitman · Trading since 2017, full-time since 2020 · YouTube creator, 137K subscribers · Published 2026-09-09 · Updated 2026-10-27 · Reviewed 2026-10-27

Fundamental analysis reads a company’s filed accounts to judge what the business is worth; price action reads the bars themselves, with no reference to the business at all. On today’s 30 Dow stocks from 2016 to 2026, the 1.66% of sessions that followed an earnings release carried 12.91% of all squared daily moves.

These two disagree less often than the arguments suggest, because they are rarely answering the same question. One is about what a business earns and is worth; the other is about how its price is behaving this week. Measured on the companies in the Dow today, earnings and price moved together loosely over a single year and much more closely over five or ten, and the days the accounts arrive are a small, loud slice of the chart.

What each one is

Fundamental analysis reads financial statements: revenue, margins, debt, cash flow and earnings per share, turned into a view about what a business is worth. Fundamental analysis covers it.

Price action reads the bars themselves: swing structure, ranges, breaks and the shape of individual candles, with no reference to the company at all. Price action covers it, and technical analysis is the larger family it sits in.

The inputs do not overlap. Most comparisons on this site set two readings of the same price series against each other. These two read entirely different things: one a set of filed documents that arrive four times a year, the other a sequence of prices that updates every session.

Where they differ

What timescale each operates on. A fundamental view resolves as earnings arrive and are priced, which takes years. A price action reading resolves in days. Both can be correct about the same company at the same moment, one about the business and one about this week’s order flow.

What each can see. Accounts reveal a company borrowing heavily, losing margin or running short of cash. A chart shows none of that until it moves the price, and by then the information is no longer news to anyone holding the filings.

What each cannot tell you. Fundamentals carry no date: a company can stay cheap for years. Price action carries no verdict on the business: a strong company and a failing one can print the same pin bar on the same day.

How each one is checked. A fundamental view is checked against the next set of reported earnings, which either arrive or do not. A price action reading is checked against the next few bars, and the same bar can confirm one trader’s read and refute another’s.

How closely earnings and price move together, by horizon

The question this page can answer with data is whether the businesses’ earnings and their share prices actually move together, and over what span. If they move together over years but not over months, then the division of labor below is not a matter of taste.

Over single years the link is loose. Ranking every company-year by its change in diluted earnings per share and by its change in share price, the two rankings agree only weakly: a rank correlation of 0.31 across 436 company-years in which earnings were positive at both ends.

Over five and ten years it is much tighter. Taking each company’s latest window, the same rank correlation was 0.43 for one year, 0.63 for three, 0.72 for five and 0.65 for ten. Companies that compounded earnings fastest over a decade were, with exceptions, the ones whose shares rose most.

Exceptions are real. IBM’s diluted earnings per share fell 16.8% over its ten fiscal years to December 2025 while its share price rose 125.1%. Microsoft’s earnings per share grew 7.13 times over its ten years to June 2026 and its price 7.29 times. Both are in the same index, and only one of them looks like the textbook.

The days the accounts reach the chart

Fundamentals arrive on a timetable, and the chart shows exactly when. Each quarter a company files its results with the SEC on a Form 8-K under item 2.02. For the 30 Dow stocks, the session that first traded on each of those filings was 1.66% of all sessions from 25 Oct 2016 to 23 Oct 2026.

Those few sessions were loud. Measured as squared daily moves (a large move counts for much more than a small one, as it does in any volatility measure), they carried 12.91% of the ten years’ total. For the typical stock the median move on a results day was 3.62 times its median move on other days.

The rest of the movement, about seven-eighths, came on days with no results release. That is the ground price action covers: the moves that come from positioning, the index and the rest of the market, which no filing explains on the day.

Which one to use: Fundamental Analysis or Price Action?

Use fundamental analysis when the holding period is years. Over that span what the business earns is most of what the measurement above found to matter, and the daily path of the price is noise you will sit through.

Use price action when the holding period is days. Over that span nothing in the accounts changes, so they have no bearing on whether this week’s breakout holds, except on the one day a quarter when the results arrive.

Use fundamentals to choose and price action to time. That is a genuine division of labor: one input picks the company, the other picks the entry and the price at which the entry is wrong.

And when a short-term trade spans a results date, treat it as a fundamental event. At one stock in six (IBM, Nike, Cisco, Walmart and 3M) a fifth or more of the decade’s squared movement came on those days.

Why the argument is usually a category error

Because value and price are allowed to differ for a long time. That gap is the premise of value investing, so a fundamental analyst is not embarrassed by a falling price, and a chart reader is not obliged to hold an opinion about earnings.

And because only one of them can see a company weakening. A chart shows a decline. The filings show whether it is a discount or a warning, and no amount of bar reading substitutes for that.

The original data

The companies. State Street’s holdings file for its Dow Jones Industrial Average ETF (DIA), dated 1 October 2026, lists 30 stocks; Visa is left out of the earnings-per-share test because its SEC EDGAR company facts carry no standard diluted earnings-per-share figure. These are the index’s current members, so every one of them is a company that did well enough to be in the Dow today. That survivorship almost certainly flatters how well earnings and price lined up, because companies whose earnings and shares both collapsed were removed from the index and are not in the sample.

Earnings. Diluted earnings per share from each company’s annual 10-K filings, via the SEC’s XBRL company facts. Each yearly change pairs the two figures one 10-K prints side by side, this year and last, so a stock split between filings cannot distort it. Longer windows chain those yearly changes and are used only where earnings stayed positive throughout: six companies had a loss year inside their latest ten, and three had gaps or too short a record in the SEC data, leaving 20 for the ten-year figure, 24 for five, 26 for three and 28 for one.

Prices. Yahoo Finance daily closes, adjusted for splits but not for dividends, taken on the final trading day of each fiscal year or the nearest one before it. Dividends are left out on purpose: the comparison is earnings against the share price, not against total return.

How sure. With a shuffle test (200,000 random pairings), a five-year correlation that strong arose by chance in about 1 in 10,000 tries and a ten-year one in about 1 in 400. That says the link is not noise in these 20 to 28 companies; it does not say it holds for companies that left the index. Even over the latest single year, earnings and price moved the same direction for 22 of 28 companies (78.57%). The full company-year table is the Dow earnings and price file, and every window in the window file.

Results days (new test, 27 Oct 2026). The same 30 stocks, now with Visa included, as listed in DIA’s holdings file of 23 October 2026. Every Form 8-K carrying item 2.02 from the SEC’s submissions records, mapped to the first session that could trade on it: the same day if the SEC accepted it before 4:00 p.m. New York time, otherwise the next session. Two filings that land on the same session count once (3 such days). Item 2.02 also covers updates to an earlier release, so Honeywell, 3M and Nike have 48 or 49 results days, most stocks about 40. That gives 1,251 results days in 75,390 stock-sessions, 2,513 per stock. A day’s move is the log change from the previous close to that close, on Yahoo closes adjusted for splits and spin-offs.

One day can carry a stock’s figure. IBM’s results on 14 July 2026 (filed 7:01 a.m.) took the stock from $290.23 to $217.07, down 25.21%; that single session is 10.15% of IBM’s ten years of squared moves, and without it IBM’s share would be 19.85%.

Every stock, highest first: IBM 27.99%, NKE 26.77%, CSCO 22.88%, WMT 22.71%, MMM 20.28%, CRM 19.33%, UNH 18.33%, AMZN 17.26%, DIS 14.27%, AMGN 13.83%, GOOGL 13.83%, NVDA 13.58%, SHW 12.75%, MSFT 11.65%, TRV 11.51%, PG 10.95%, HON 10.51%, CAT 10.06%, AAPL 9.79%, MRK 8.70%, JNJ 8.51%, KO 7.87%, V 7.65%, MCD 6.87%, AXP 6.59%, HD 6.32%, GS 5.02%, CVX 4.75%, BA 4.41%, JPM 4.25%.

Checks. The volume line above is the timing check: 86.97% of results days traded at least 1.5 times the median volume of the 20 sessions before, which a misdated filing would not. Nasdaq’s historical quotes give the second source. Its files start on 26 Oct 2016, so its first change is on 27 October, and its closes are adjusted differently from Yahoo’s around the spin-offs at Honeywell (2018, 2025 and 2026), 3M (2024), IBM (2021) and Merck (2021), or around the August 2023 Kenvue exchange at Johnson & Johnson; those 9 stock-days are left out of the check. On the other 75,321 stock-sessions, the share is 12.83% on Nasdaq closes and 12.83% on Yahoo closes. Every results day is in the results-day file, and every stock in the results-day summary file.

When it fails

The characteristic failure is using fundamentals to justify holding a losing trade. A position taken on a chart reading goes against you, and rather than accepting the stop the trader reaches for the company’s accounts, which were not the reason for the entry and are being recruited afterward to avoid a decision. The valuation may even be sound, and it is irrelevant, because the position was sized for a days-long horizon and is now being held on a years-long argument with no change to the risk. The single-year correlation above is the reason this goes wrong so often: over a few months, good earnings and a rising price are only loosely connected.

A second failure is the reverse: abandoning a long-term holding on a chart signal that operates over days.

A third is holding a chart trade through a results date without noticing it. The median results-day move of the typical stock was 3.62 times its ordinary median, so a stop sized on ordinary days is sized for the wrong session.

A fourth is expecting fundamentals to provide timing, which they do not claim to do.

And a fifth is reading the long-run fit as a promise. It was measured on survivors, after the fact.

Questions people ask

What is the difference between fundamental analysis and price action?

Fundamental analysis reads filed accounts to judge a business; price action reads the bars with no reference to the business. They meet on results days: 1.66% of the Dow stocks’ sessions from 2016 to 2026, carrying 12.91% of their squared daily moves.

Which is better for day trading, fundamentals or price action?

Over days the accounts do not change, and about seven-eighths of the movement came on days without a results release. The exception is the results day itself, when the typical Dow stock’s median move was 3.62 times its ordinary one.

Do earnings move stock prices?

Over one session, yes: results days traded heavy volume 86.97% of the time and carried far more than their share of the movement. Over years, earnings growth and price change ranked alike at 0.72 across five-year windows.

Can you combine fundamental analysis and price action?

Yes, by horizon: fundamentals to choose what to hold over years, price action to time entries over days, and a note of every results date that falls inside a short-term trade.

Related

Fundamental analysis covers reading the accounts and what a filing contains. Price action covers reading the bars and the swing structure. And technical analysis is the broader school of chart reading that both camps argue over.

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