When the share count outruns the last annual report
340 of the 2,154 US companies below three hundred million dollars carry a share count more than half again above, or more than a third below, the one in their own last annual report, which is 15.8% of the band. At ten billion and above it is 2.7%.
The comparison is between the market value divided by the price, which gives the shares trading today, and the diluted count the filer printed in its last annual report. When today’s is more than one and a half times the filed count, or under two thirds of it, we decline to publish the ratio. Nothing below is a percentage of anything issued.
How often the two counts disagree
- Below three hundred million15.8% of 2,154 companies in that band.
- 340
- Ten billion and above2.7% of 912. The same test, the same night.
- 25
Notes on this count
The marker behind both rows is the catalogue’s own, and so is its definition: Held back: the last annual report's share count does not match today's, as after a split. It is applied to every price ratio in the table, on every company, on the same pass, so the two rows above are not two measurements with a story told between them.
GET /api/public/companies. Catalogue built Oct 16, 2026. Prices of Oct 15, 2026.
And which figures go dark when it does
- Price to book
- 306
- Free cash flow yield
- 139
- Price to earnings
- 21
- Dividend yield
- 7
- Enterprise value to EBIT
- 7
GET /api/public/companies. Catalogue built Oct 16, 2026. Prices of Oct 15, 2026.
Why this costs the band its best ratio
Price to book is the one valuation measure most of these companies do publish, and it is also the one withheld most often here: 306 times. Assets and equity are the numbers a loss-making small company still has, so the ratio everybody falls back on is exactly the ratio a changed share count takes away. A vendor that keeps printing it is dividing last year’s book value by this year’s shares.
What this count is not
It is not a measure of dilution, and three different events land inside it: an issue of new stock, a forward split, and a reverse split. Those mean opposite things to an owner and nothing in this comparison tells them apart, so no sentence here claims to. A company’s own cover page states the shares outstanding on a given date, and reading that across filings is the only way to answer the question properly.
Nor is the absence of a mark a clean bill of health. A raise of a third of the equity leaves both counts well inside the threshold and never appears above. The threshold exists to stop a ratio being printed over nonsense, not to flag a financing, and reading it as the second thing would be reading a guard as a signal. Every formula and every refusal is written out at methodology.
What is watched on a company you hold a thesis on
Once you subscribe, every night:
- Debt and financing. A guardrail you set on debt to equity or the current ratio, re-tested every night against the newest filed accounts.
- Cash going out faster than it comes in. A guardrail on free cash flow yield or the Altman Z-score, re-tested the same way.
- The price you paid, against the value you wrote down. A guardrail on your margin of safety against your own target or saved DCF, read against the latest close.
- An auditor resigning or being replaced. An 8-K reporting an auditor change (item 4.01) is emailed to you. If you wrote a guardrail on auditor changes in the last twelve months, that guardrail decides the email instead.
- Accounts the company says can no longer be relied on. An 8-K saying earlier statements should no longer be relied on (item 4.02) is emailed to you. A guardrail on these notices in the last twelve months decides the email instead, where you wrote one.
- A delisting notice, or a listing standard missed. An 8-K reporting a delisting notice or a failure to meet a listing rule (item 3.01) is emailed to you.
- Shares sold outside a registered offering. An 8-K reporting an unregistered sale of equity (item 3.02) is emailed to you.
- A material agreement, a deal closed, new debt, a shareholder vote or other company news. An 8-K under item 1.01, 1.02, 1.05, 2.01, 2.03, 2.05, 2.06, 3.03, 5.01, 5.07, 7.01 or 8.01 goes to your Alerts with the item title and a link to the filing. These are not emailed, and the filing is not read beyond its item codes.
- A director or officer leaving. An 8-K reporting a director or officer change (item 5.02) goes to your Alerts and is never emailed, because routine board elections file under the same item.
- Shares issued between annual reports. A guardrail on the share count from each filing’s cover page, or on how far it moved since the last filing, over about a year or about three, re-tested every night.
- Insiders buying or selling. A guardrail on net insider buying or selling over three or twelve months, or on how many insiders sold in three months, read from open-market Form 4 filings.
- How long an exit would take at today’s volume. A guardrail on average daily dollar volume or on the share of sessions with no trade. The company page also turns a position size you type into days of volume.
- Cash runway, in months. A guardrail on months of cash at the burn of the newest filed twelve months, re-tested nightly against the newest filed accounts. A business whose operations brought cash in has no runway figure, and the page says why.
- A going-concern paragraph in the audit opinion. The newest annual or quarterly report is read for a statement of substantial doubt about the company continuing as a going concern. A guardrail on that doubt is re-tested nightly, and its email quotes the sentence and cites the filing.
- A reverse split. A confirmed reverse split puts one line in the inbox of a subscriber holding the company or with a thesis on it. A guardrail on reverse splits in the last year is emailed like any other.
- A shelf registration or an at-the-market sale. A new registration statement (S-3, S-1, F-3) or a prospectus supplement that sells (424B5, 424B4) puts one line in the inbox of a subscriber holding or watching the company. A guardrail on new financing filings sends it by email, citing the filing.
- Warrants and convertibles. A guardrail on the shares tagged warrants and convertible notes could add, as a percentage of the shares outstanding on the same date, read from the company’s own XBRL facts and re-tested every night.
- One customer carrying the revenue. A guardrail on the largest single customer’s share of revenue, as the newest annual report tags it, re-tested every night and emailed with the filing cited.
A guardrail we could not read that night is reported as untested, never as holding.
Not watched yet operating milestones. None of these is read for you today, so silence about them means nothing.
What this page cannot tell you
Nothing here is about a company you hold, and the threshold above is deliberately blunt: it catches the share counts that have already moved a long way and says nothing about the quarter they started moving. That quarter is the one that matters, and it arrives in a filing, usually with no announcement a holder would see.
Run the companies you own through the audit at no charge and it names every ratio it will not print for them, with the reason. Professional turns that into a standing question: you record what has to stay true about a holding, and it is tested again every night instead of against your memory of the month you bought.
Every guardrail you have saved is re-tested every night against newly filed accounts, and you get one email the first night one breaks, naming the company, the reading and the filing it came from. What to do about it is yours. It comes with Professional.
The check runs every night at 23:30 UTC. A filing is picked up the night after it is published.
The audit needs no account. A new account starts with 30 days of Professional, no card.
Meeting a test is not a reason to buy. Every count above is read from the public endpoints when the page is built, never typed into it: the whole table is at the company catalogue and every formula is at methodology.
No company is named above, and no count here is evidence about any single filer.