What Moves a Stock Price
A company you own reports its results. Profit is up. Revenue is up. The release uses the word record. You open your account expecting a good day and the stock is down four percent. Nothing about the news was bad, and the price went the wrong way anyway.
That experience is the reason this guide exists, and the explanation is simpler than it feels. What moves a stock price is not how good the news is. It is how the news compares with what people had already assumed. Those assumptions are not written down anywhere obvious, but they are priced in, and every price you see already contains them.
This guide walks through one real week on the Toronto Stock Exchange where that gap was on full display. Between August 25 and August 27, 2026, all six of Canada’s largest banks reported quarterly results, and all six came in above the profit figure analysts had published in advance. Same industry, same week, same headline outcome. The share price reactions we recorded ran from up 7.0% to down 4.2%.
A price is not a score
Start with what a price actually is, because most confusion about stock moves begins with a wrong mental model.
A stock’s price is not a grade awarded to a company for its performance. No authority sets it and no formula produces it. It is simply the level at which the buyers and sellers of that stock currently balance. It is the last point where somebody willing to sell and somebody willing to buy agreed. That is the whole definition, and everything else in this guide is a consequence of it.
Two things follow immediately.
The first is that a price only changes when the balance changes. If everyone holding the shares and everyone considering them keeps thinking exactly what they thought yesterday, the price sits still, no matter how profitable the company is. Profitability is already in the number. It got there when people formed their views.
The second is that the balance changes when new information arrives. The operative word is new. Information that everyone already had is not information, in the sense that matters to a price. It was absorbed when it became known, or when it became likely, which is often long before it was officially announced.
Put those together and you have the model that explains the whole week below. A price moves on surprise, not on quality. This is also why the price cell is the most misunderstood field on any quote screen: it looks like a verdict on the company and it is really a running tally of a negotiation. Our guide to how to read a stock quote walks every other field on that screen and shows which ones the market writes and which ones come out of the company’s filings. This guide is the story behind that one cell.
One boundary before the evidence. This guide is about what moves a single stock. Whole markets also move together, and when they do, an individual company can rise or fall on a day when nothing at all happened to it. That is a different subject with different causes, and it is not treated in depth here, though you will see one clean example of it further down.
The week every big bank beat
Canadian banks report on a fiscal year ending October 31, so their third quarter covered the three months to July 31, 2026. Six banks reported over three days in late August.
Before the week began, analysts covering each bank had published an estimate of what its quarterly earnings per share would be. The average of those estimates is the consensus. It is the number a company is measured against on the day, and it is the closest thing there is to a written record of what the market expected. The consensus figures used throughout this guide come from an Investing.com compilation dated August 21, 2026, and each bank’s actual result comes from its own results release.
| Bank | Reported | Adjusted EPS | Consensus | Beat | Same-day move |
|---|---|---|---|---|---|
| BMO | Aug 25, 2026 | $3.96 | $3.74 | +$0.22 | +0.55% |
| Scotiabank | Aug 25, 2026 | $2.28 | $2.08 | +$0.20 | +7.02% |
| National Bank | Aug 26, 2026 | $3.39 | $3.18 | +$0.21 | -4.2% |
| RBC | Aug 27, 2026 | $4.28 | $4.04 | +$0.24 | Not recorded |
| CIBC | Aug 27, 2026 | $2.73 | $2.50 | +$0.23 | Not recorded |
| TD | Aug 27, 2026 | $2.77 | $2.45 | +$0.32 | Not recorded |
Adjusted earnings per share from each bank’s Q3 2026 results release, for the quarter ended July 31, 2026. Consensus per the Investing.com compilation of August 21, 2026. Same-day moves per StockAnalysis for August 25 and Trading Economics for August 26. We did not record closing moves for the three Thursday reporters, so those cells say so rather than carrying an estimate.
Six banks, six beats, and the widest of them was TD’s, at 32 cents above consensus. If a price were a score, that table would produce six green days of roughly similar size.
It did not. Scotiabank rose 7.02% to close at $128.73. BMO rose 0.55%, which is close enough to nothing. National Bank fell 4.2%. Three beats, three completely different answers from the market.
Measured against consensus, BMO beat by 5.9%, Scotiabank by 9.6% and National Bank by 6.6%. Those three percentages are our own arithmetic on the consensus and reported figures in the table above. Line them up against the day’s share moves of plus 0.6%, plus 7.0% and minus 4.2% and there is no relationship at all. The bank with the middle-sized beat had the worst day.
If you want to understand what moves a stock price, that chart is the puzzle to solve. Whatever the market was responding to, it was not the beat.
If bank shares are the corner of the market you actually own, our Canadian bank stocks coverage tracks the Big Six as a group, which is the useful unit here, since these six report within days of each other and are routinely read against one another.
Expectations move before the announcement does
Here is the piece that resolves it. The consensus published a few days before a report is not the only expectation in the market, and it is not fixed. Analysts revise their estimates continuously as information reaches them, and buyers and sellers act on their own views long before any official number changes.
Yahoo Finance analyst data for National Bank, as of September 5, 2026, shows this happening in the open. The consensus for the bank’s full fiscal year stood at $13.09 per share, against $12.86 ninety days earlier. In the preceding 30 days alone, 10 analysts had raised their fiscal-year estimate and 2 had cut theirs. The consensus for the following quarter sat at $3.54, drawn from 6 analysts.
Read that as a picture rather than a set of numbers. The bar a company must clear is being nudged upward and downward every week between reports, by people watching the same industry the company operates in. By the time the company finally announces anything, most of what it is about to say has already been guessed at, argued over and priced.
Now add the other half of the context, which is habit.
Every one of the six banks had beaten consensus in each of its last four reported quarters, with average beats ranging from 5.6% for Scotiabank to 11.6% for Royal Bank. Heading into the August week, the same four-quarter streak was true for all six, per the Investing.com compilation.
That changes the meaning of the word beat entirely. When a company has exceeded the published estimate four times running, and so has every one of its peers, a beat is not news. It is the expected outcome. The surprise would have been a miss.
So the question a stock actually answers on results day is not “were the results good.” It is “were the results different from what was already assumed, and in which direction.” A beat that matches the assumed beat is, for pricing purposes, a non-event.
What was under National Bank’s headline
National Bank reported on the morning of Wednesday, August 26, 2026, and by every headline measure it was a strong quarter. Adjusted diluted earnings per share came in at $3.39, up 26% from $2.68 a year earlier, against a $3.18 consensus. Revenue was $4,053 million, up from $3,449 million and ahead of a $3.86 billion consensus. Net income was $1,307 million, up 23%. Return on equity was 16.1%, against 13.6% a year earlier. Those figures are in National Bank’s third-quarter 2026 release.
The stock fell 4.2% that day.
It was not the market’s doing. The S&P/TSX Composite fell 0.39% on August 26, closing at 36,814. Whatever happened to National Bank happened to National Bank.
Three facts sit around that drop, and the honest way to teach this is to give you all three rather than pick a villain.
The price had already moved. On Tuesday, August 25, before National Bank had released anything, its shares rose 2.52% to close at $222.53, inside a 52-week range of $141.46 to $237.13. That was the day two of its peers reported strong quarters. The stock went up on a day the company itself said nothing. Whatever a buyer was paying for on Tuesday afternoon, part of it was an expectation about Wednesday morning.
A beat was the expected outcome. Four consecutive quarters of beats across all six banks is the backdrop described above. Delivering one more confirmed the assumption rather than overturning it.
The detail underneath was mixed. The same release reported provisions for credit losses of $246 million, against $203 million a year earlier. Provisions are money a bank sets aside for loans it expects to go bad, and a larger provision is a statement about what the bank sees coming. Net interest margin, the spread between what a bank earns on its loans and what it pays for funding, was 2.19%, down from 2.25%. The quarterly dividend was held unchanged at $1.32 per share.
None of those three facts is the cause. Nobody, including the people who traded that morning, can decompose a single day’s price move into shares of blame, and any writer who tells you otherwise is guessing with confidence. What you can say precisely is this. The price going into the report already contained a beat, so the beat itself carried little new information, and the figures underneath the headline were the part of the release that was not already assumed.
That is the complete answer to why a stock can fall on good news. The good news was already in the price. What was left to react to was everything else.
The other things that move a stock
Earnings against expectations is the biggest single mover of an individual stock, but it is not the only one. Each of the following showed up in the same week, which makes it a convenient set of examples.
Which basis a headline number uses
Both BMO and CIBC published two different earnings figures for the same quarter, and only one of them was comparable to the consensus.
BMO reported adjusted earnings per share of $3.96 against the $3.74 consensus. Its reported earnings per share, on the unadjusted basis, was $2.38, which is well below that same consensus. The difference is adjusting items totalling $1,109 million after tax. The largest of them, in the bank’s own words, was “the announced sale of BMO’s Transportation Finance and Vendor Finance businesses, resulting in a charge of $962 million ($1,092 million pre-tax), primarily related to goodwill.”
CIBC had a smaller version of the same split. Adjusted diluted earnings per share was $2.73 against a $2.50 consensus. Reported diluted earnings per share was $2.47, marginally below that consensus, after items of note totalling $0.26 per share, primarily charges related to the announced sale of CIBC Caribbean.
The lesson is mechanical and it will save you from a wrong conclusion at least once a year. Consensus estimates for these banks are built on adjusted earnings. Comparing a reported figure to an adjusted consensus produces a phantom miss. Before you decide a company beat or missed, check that both numbers in your comparison are on the same basis.
Corporate actions that change supply
On the same day it reported, BMO announced its intention to repurchase up to 25 million of its common shares, approximately 3.6% of its public float as of July 31, 2026. The plan was subject to approval by OSFI and the TSX, and was expected to run from on or around September 8, 2026 to September 7, 2027.
A buyback moves a stock through a different channel than news does. The public float is the pool of shares actually available to trade, a concept covered in our guide to what a stock is. When a company repurchases its own shares, it becomes a standing buyer in the market and the pool shrinks. Dividends, share issues, splits, takeovers and index changes all work on the same axis. They alter who owns what and how much is available, without necessarily saying anything about how the business performed last quarter.
This is worth separating in your head from the news channel, because it explains moves that no headline accounts for.
The market and the sector the stock sits in
On August 19, 2026, six days before any of these banks reported, Canadian bank stocks sold off together on hawkish minutes from the US Federal Reserve’s July meeting. Per Trading Economics, as of that close, BMO fell 4.4%, CIBC 3.7%, TD 3.5%, Scotiabank 3.3% and RBC 3.1%, while the S&P/TSX Composite finished essentially flat at 36,402, up 0.09%.
No Canadian bank had released anything. The news was foreign, and it was about interest rates rather than about any of these companies. Rate expectations feed directly into expectations for bank profitability, so a change in one repriced the other. Notice also that the index barely moved, which means this was a sector event rather than a market event. A stock can be dragged by the company it keeps.
Narrative, and results measured against a stated target
Scotiabank’s 7.02% day on August 25 is the counterexample that completes the set. Its adjusted diluted earnings per share of $2.28 beat the $2.08 consensus, and revenue of $10.54 billion came in above a $9.97 billion consensus. The release called the quarter “a record quarter for the Bank”, and adjusted return on equity was 14.2% against the bank’s own 14% target.
That last detail is the one worth studying. A company that has publicly committed to a target creates a second scoreboard alongside the analyst consensus, and clearing it is a distinct piece of information. Scotiabank’s chief executive made the point directly in the release. “Q3 was a record quarter for the Bank, as all business lines reported strong results and we exceeded our medium-term objectives in the period,” said Scott Thomson, President and CEO of Scotiabank. “In particular, we exceeded our 14% return on equity target this quarter, highlighting the improvements that we have made across the bank to increase margins and fee income.”
We are not claiming those words caused the 7.02%. We are pointing out that the quarter contained something beyond the earnings line: a public objective met. Where the earnings beat was the expected outcome, that was not necessarily priced.
The mechanics underneath all of it
Every move described in this guide reaches the price the same way. Buy orders and sell orders arrive, they do not match evenly, and the price shifts until they do. Information matters only because it changes who is willing to trade and at what price. There is no separate mechanism by which good news lifts a stock; the news simply changes the orders. Our guide to how the stock market works takes apart that order book and shows exactly who is on the other side of your trade.
One practical consequence of that plumbing. The same dollar amount of buying moves a thinly traded stock far more than it moves a bank whose shares change hands by the million every day. In a stock with few willing sellers at any given price, a modest order has to reach much further up the queue to be filled. Size of move and size of news are not the same thing.
The mistake this costs people money
The expensive version of misunderstanding all this is short and specific: reading a headline beat as a buy signal, and acting on news after the price has already reacted to it.
Price the real case. Suppose that on Tuesday, August 25, 2026, you watched BMO and Scotiabank report strong quarters and reasoned as follows. Every big bank has been beating estimates. National Bank reports tomorrow morning. National Bank will probably beat too. So you bought at Tuesday’s close of $222.53.
Both halves of that reasoning were correct. The banks were beating, and National Bank did beat, on every headline line. The stock fell 4.2% the next day.
The trade lost because the reasoning was already public. The 2.52% the stock gained on Tuesday was the market pricing in exactly the conclusion you reached, and you bought after it. A forecast that everyone shares is not an edge. It is the price.
The smaller mistake in the same family is the basis error described above. BMO’s reported $2.38 looked like a large miss against a $3.74 consensus it was never being measured against. A reader who saw only those two numbers would have concluded the bank had a terrible quarter on a day it beat by 22 cents and closed up.
What to take away
Six ideas, in the order they matter.
A price is a balance point, not a grade. It sits where buyers and sellers currently agree, and it moves when that agreement breaks.
Only new information moves it. Anything already known or widely assumed is in the price already, which is why a strong quarter can produce a flat or negative day.
Expectations are the benchmark, and they move continuously. The consensus drifts between reports as analysts revise. National Bank’s fiscal-year consensus rose from $12.86 to $13.09 over ninety days, with 10 raises and 2 cuts in a single month, all of it before the company said a word.
Read past the headline number. Provisions, margins, the dividend decision, guidance and adjusting items are where the part that was not already assumed usually sits.
Check the basis. Adjusted versus reported earnings turned two of these six beats into apparent misses for anyone comparing the wrong pair of numbers.
Company news is one of several channels. Buybacks and other corporate actions change the supply of shares. Sector and macro news reprices whole groups at once, as August 19 did to every Canadian bank without a single company announcement.
The habit that follows from all of it is simple to state and hard to keep. Before you react to a piece of news about a company you own, ask what the market already assumed. If your conclusion is the same one everybody else reached from the same public facts, the price has probably reached it first.
Following an individual holding over time is its own skill, and the Ontario Securities Commission covers it without anything to sell you in its investor-education material on how to monitor stock performance.