Skip to main content
All insights

Share Dilution: Who Doubled Their Share Count, and What Their Stocks Did

Equibles Research · Original study 12 min read
Cite this study & download tables

Equibles Research. Share Dilution: Who Doubled Their Share Count, and What Their Stocks Did. 2026-11-19. https://equibles.com/research/share-dilution-who-doubled-their-share-count-and-what-their-stocks-did

Methods, sample dates, limitations, and source references are described in the study below. Each table's CSV download contains its published cells, not the underlying raw dataset. Spreadsheet-formula text is escaped for safety.

Please cite this study and retain its source attribution when sharing a figure or table.

Summary

We compared the cover-page share counts of 3,692 US-listed companies a year apart. The median company's count rose 0.8%, but 278 doubled theirs, and their median stock fell 77.9% over the same year. Large companies that issued stock to pay for an acquisition were the exception.

On this page

In October 2025, Devon Energy (DVN $41.96 -3.45%) had 627.3 million shares outstanding. A year later it had 1.14 billion, 82% more. On May 7, 2026, Devon completed its merger with Coterra, and each Coterra share was converted into 0.70 Devon shares. Devon's stock rose 57.6% over the same twelve months.

Tonix Pharmaceuticals (TNXP $2.82 -4.08%) grew its count by about two-thirds over a similar year, from 11.8 million shares to 19.4 million, without buying anyone. It sold stock through an at-the-market program, a standing agreement that lets a broker sell new shares into the market a few at a time; its 10-Q for the third quarter of 2026 reports 6.3 million shares sold that way in the first nine months of 2026, for about $81.7 million of net proceeds. Its stock fell 71.9%.

Both companies printed a lot of new stock. To see how common that is, and what came of it, we took the share count every US company states on the cover of its 10-K and 10-Q and compared each company's latest count with its count a year earlier: 3,692 companies listed on Nasdaq or the NYSE. Most barely moved. A large minority, almost all of them small, issued so much stock that their share count doubled, and those stocks mostly collapsed.

The data behind this piece

Explore the primary data Equibles carries for DVN:

At a glance

What we measured Result
US-listed operating companies compared 3,692
Median change in share count over about a year +0.8%
Count down 1% or more (net buybacks) 26.3% of companies
Count up 20% or more 18.7%
Count at least doubled 278 companies (7.5%)
Count up tenfold or more 89 companies
Median stock return over the same year, smallest third by trading volume, count doubled -82.3% (76.8% at least halved)
The same, smallest third, count changed by less than 1% +6.7%
Smallest third by trading volume: median change in share count +6.1%
Largest third: median change in share count -0.6%

What we counted

Every 10-K and 10-Q starts with a cover page, and the cover states how many shares of common stock were outstanding on a recent date. Companies tag that number in their XBRL filings, so it can be compared across companies. For each company we took its latest cover count, dated from May to November 2026, and the cover count closest to one year before it. The median pair runs from October 31, 2025 to October 30, 2026.

Two adjustments matter. A stock split changes the count without changing anyone's stake, so we put both counts on the same basis using the split records in our database. We applied only splits that the company's own counts, or a filing stating the split ratio, support. And a company with several classes of common stock, such as class A and class B, is counted as the total of its common classes, never preferred stock or depositary shares.

The comparison is descriptive. A stock that falls and a company that sells more stock feed each other: a company short of cash sells shares because its business needs money, and each sale at a lower price takes more shares to raise the same amount. The numbers below show which companies printed shares and what their stocks did over the same year. They don't show that the printing caused the fall.

Most companies barely changed

The median company ended the year with 0.8% more shares than it started with. A quarter of the companies, 26.3%, finished with at least 1% fewer shares, which takes buybacks larger than the stock handed out to employees. About as many changed by less than 1% in either direction.

The distribution has a long right tail. Almost a third of the companies, 30.6%, grew their count by 5% or more, 18.7% by 20% or more, and 278 by 100% or more. Of those 278, 89 ended the year with more than ten times as many shares as they started with, after adjusting for splits.

A small-company habit

We split the companies into thirds by how much their stock trades, using the median daily dollar volume over the 90 days before the first count. The smallest third traded less than $2.1 million a day; the largest traded more than $33.8 million.

Companies by change in share count over a year, by trading volume
Trading volume Companies Median change in count Grew 20% or more Doubled or more Shrank 1% or more
Smallest third 1,208 +6.1% 38.0% 234 8.5%
Middle third 1,208 +0.8% 13.3% 29 25.3%
Largest third 1,208 -0.6% 4.0% 4 45.5%

The two ends look like different markets. In the largest third, 45.5% of companies bought back enough stock to shrink their count by 1% or more, and only 4 of 1,208 doubled it. In the smallest third, 38.0% grew their count by a fifth or more and 234 doubled it; only 8.5% shrank. Of the 278 companies that doubled, 234 were in the smallest third.

Drug and biotech companies made up 98 of the 278, more than a third. Of the 535 drug and biotech companies in our sample, 18.3% doubled their count in a year, against 6.5% of software and IT services companies and 4.3% of banks, insurers and real estate companies. A company with no product revenue pays for its trials by selling stock, and selling stock at a lower price each time is how a share count doubles.

What the stocks did

We measured each stock's return over the same window as its two share counts, using prices adjusted for splits and dividends. The median stock in the sample rose 0.8%. The table compares companies by how much their count changed, within each size third, because small stocks behave differently from large ones whatever their share count does.

Change in share count Smallest third Middle third Largest third
Shrank 5% or more +14.3% (35) +9.7% (113) +1.3% (175)
Shrank 1% to 5% +24.2% (68) +13.8% (193) +5.2% (375)
Within 1% +6.7% (283) +13.2% (342) +12.3% (345)
Grew 1% to 5% -8.3% (197) +1.2% (245) +4.5% (160)
Grew 5% to 20% -23.2% (165) +13.5% (154) -5.8% (105)
Grew 20% to 99% -37.9% (222) -9.4% (132) -2.3% (44)
Doubled or more -82.3% (220) -38.2% (28) (4 companies)

Median return over the year, with the number of companies in brackets.

Median stock return over the same year, smallest third by trading volume

Among small companies the pattern is steady: the more stock a company issued, the worse its stock did, step by step. The median small company whose count stayed within 1% rose 6.7%. The median small company that doubled its count fell 82.3%, and 76.8% of those stocks lost at least half their value.

In the middle and largest thirds the link was weaker and less regular. A mid-sized company whose count grew 5% to 20% had a median return of 13.5%, level with one that held its count flat (13.2%), while growth of 20% to 99% came with -9.4%. Larger companies often issue stock for other reasons, such as paying for an acquisition or settling employee stock awards. Doubling the count was rare there and still went badly: the mid-sized companies that doubled had a median return of -38.2%.

Buybacks didn't line up with strong returns either. In the largest third, companies whose count shrank 5% or more had a median return of 1.3%, against 12.3% for those that held it flat. A company with a falling stock price can buy back more shares for the same money, so the arrow can run either way here as well.

Paying for a company or raising cash

A share count grows for two broad reasons. A company issues stock to pay for an acquisition, or it sells stock for cash: through an at-the-market program, an underwritten offering, a private placement, or by converting debt or warrants. The SEC's Form 8-K separates some of these. Item 2.01 reports a completed acquisition or disposition of assets. Item 3.02 reports an unregistered sale of equity, such as a private placement or a share issuance to settle debt.

We took the 692 companies whose count grew by 20% or more. We called a company a deal company when it filed an Item 2.01 8-K in the same quarter as its largest quarterly jump in shares, a jump of 10% or more. We put the rest into two groups: those that filed an Item 3.02 8-K at some point in the window, and those that filed neither.

Count grew 20% or more Companies Median return At least halved
Deal: Item 2.01 8-K in the quarter of the biggest jump 134 -17.9% 35.2%
No deal, at least one Item 3.02 8-K 348 -64.9% 63.8%
No deal, no Item 3.02 8-K 210 -17.7% 33.2%
Median stock return over the same year, companies whose share count grew 20% or more

The companies that sold unregistered stock and bought nothing did worst: a median return of -64.9%, and nearly two-thirds lost at least half their value. In the smallest third that group's median was -68.6%. The group with neither filing, a median of -17.7%, mixes registered offerings, at-the-market sales, warrant exercises and debt conversions; a sale under an at-the-market program needs no 8-K of its own.

Size changes the deal picture too. Among the smallest third, deal companies did nearly as badly as the rest, a median of -57.8%. Among the largest third, the 18 deal companies had a median return of 12.2%, 12 of them rose, and none of them halved. The best-known names show why:

In each case the new shares went to the owners of a business that came along with them. When a company sells stock for cash, what the new shares buy is the cash, and the question is what the company does with it. Some of the hardest falls among the larger companies were at those that sold stock to buy cryptocurrency. "We fund our purchases of bitcoin primarily from proceeds of our offerings of our class A common stock and Preferred Stock instruments," Strategy (MSTR $90.32 -6.40%) says in its latest 10-Q, which reports 108.8 million class A shares sold through at-the-market programs in the first nine months of 2026. Its common count grew 47.0% over the year, from 287.4 million to 422.4 million shares, and its stock fell 42.6%. DeFi Development (DFDV $2.61 -5.43%), which runs a digital-asset treasury strategy and also sells stock at the market, grew its count 27.6% while its stock fell 57.1%.

Reverse splits hide the scale

Many of the companies that doubled also did a reverse split, which cuts the count by a fixed ratio, often to keep the price above an exchange's $1 minimum. That makes the raw count look calm. Of the 278 companies that doubled their count, 134 did at least one reverse split during the window. A company that did a 1-for-20 reverse split and ended the year with the same number of shares on its cover had twenty times as many shares as a year earlier, on a split-adjusted basis. That is why we put both counts on the same split basis before comparing them.

What to check

If you own a small company, the cover of its 10-Q is the quickest dilution check there is.

  • Compare the cover count with the one on the 10-Q a year earlier. A rise of 20% or more in a year is common among small companies; in our sample a rise of 20% to 99% came with a median return of -37.9% in the smallest third, and doubling with -82.3%.
  • Look for a reverse split in the year. If there was one, multiply the old count by the split ratio before comparing.
  • Read the equity note in the latest 10-Q for an at-the-market program: how much remains available, and how much was sold in the quarter. Programs that are mostly unused can still be used.
  • Look for Item 3.02 8-Ks, which report private placements and stock issued to settle debt, often at a discount and with warrants attached.
  • When the count jumped because of an acquisition, judge the deal, not the dilution. The new shares came with a business.

Method

  • Companies: issuers with a 10-K or 10-Q cover count (dei:EntityCommonStockSharesOutstanding) dated May 1 to November 16, 2026, and a second cover count dated within 45 days of one year earlier, listed on Nasdaq or the NYSE according to the SEC's company ticker and exchange file. We left out funds, business development companies and blank-check companies (SIC 6221, 6726 and 6770, and BDC elections), 3,692 companies remaining.
  • Counts: the latest filing for each cover date. For companies with several classes of common stock, the sum of the common classes, used only when every cover in the window lists the same classes; otherwise the undimensioned total. Before the other filters, 129 issuers whose cover classes changed during the year were left out.
  • Splits: applied when effective between the filing dates of the two covers, and only when the company's own share counts confirm the split, an SEC filing near the date states its ratio, or (for a reverse split) the filings near the date describe a reverse split. Price adjustments for spin-offs, which our split records also hold, were not applied.
  • Data checks: we left out 23 companies whose count fell by half between two consecutive covers with no recorded split, and three whose count jumped more than 500-fold between two covers, a scale error in the tagging. We dropped the stock returns of 18 companies whose price history jumped on a split date, or whose count jumped by the same factor that their price fell in a single day, because an unrecorded split and a deeply discounted offering look alike in the data.
  • Returns: split- and dividend-adjusted closing prices from the first cover date to the second, for 3,612 companies with at least 200 trading days in the window.
  • Size thirds: median daily dollar volume in the 90 days before the first cover, 3,624 companies with volume data.
  • Deal companies: an 8-K reporting Item 2.01 filed in the quarter that holds the company's largest quarter-to-quarter rise in shares, when that rise is 10% or more. Item 2.01 also covers asset sales, and an acquisition paid in cash can coincide with a stock sale, so this is an approximation. Every company named in this article was checked against its own filing.

Limitations

  • Survivorship. A company that went bankrupt or was delisted during the year has no second cover and is not in the sample. Those are among the heaviest diluters, so the returns of the companies that doubled are, if anything, better than the full group's.
  • Same-window comparison. The share counts and the returns cover the same year. A falling stock can cause more issuance as easily as issuance can push a stock down, and a company in distress does both.
  • Cover dates differ. Each company's window runs between its own two cover dates, so the years don't line up exactly. Most run from October or November 2025 to October or November 2026.
  • The deal flag is approximate. It relies on 8-K items and timing, and it misses deals whose completion 8-K we don't hold.
  • Splits. Our split records come from several sources. We applied only splits that a filing or the company's own counts support, and we checked the largest cases by hand, but an unrecorded split could still put a company in the wrong group.

Conclusion

Most US companies barely changed their share count over the past year, and many large ones shrank it. Among small companies, issuing stock is the norm: the median small company grew its count by 6%, and nearly a fifth of them at least doubled it. Those that doubled saw their stocks fall 82.3% at the median. Large companies that issued stock to buy another business were the exception, and their stocks mostly rose. The cover of a 10-Q, compared with the one a year before, is a quick way to tell which kind of company you own.

Sources

Found this useful? Share the analysis.
Free MCP server

Ask your own questions about Devon Energy — in ChatGPT or Claude

Connect the free Equibles MCP server and your AI can inspect SEC filings, earnings calls, institutional holdings and market data directly — with citations and no copy-paste.

Ask ALVIS about DVN — free, no setup

ALVIS is our own analyst, running on the same tools — click a question and it answers here, or copy it for ChatGPT or Claude.

Server URL
https://mcp.equibles.com/mcp
Or browse all of DVN’s data on Equibles →

More from Equibles Research