Days to cover, also called the short interest ratio, is the number of average trading days it would take short sellers to buy back every share they have shorted: reported short interest divided by average daily volume. A reading of 2 says the shorts could, in theory, exit inside two normal sessions. A reading of 10 says it would take two full weeks of every share traded going to them. That is the whole formula. The hard part is knowing what the inputs actually measure, how old they are, and why a big number is not a countdown clock.
Short squeeze talk lives on this metric. So does a lot of bad data. Here is how the number is built, where it comes from, and where it misleads.
The formula, with hypothetical numbers
Two ratios matter, and they answer different questions.
- Days to cover = short interest / average daily volume. It measures crowding against liquidity.
- Short interest as % of float = short interest / float. It measures crowding against the shares actually available to trade.
Take a hypothetical stock, not a real ticker, built only to show the arithmetic:
| Hypothetical input | Value |
|---|---|
| Shares sold short (short interest) | 12,000,000 |
| Average daily volume | 2,000,000 |
| Float | 60,000,000 |
| Days to cover | 12,000,000 / 2,000,000 = 6.0 |
| Short interest % of float | 12,000,000 / 60,000,000 = 20% |
Read together, the two numbers tell a sharper story than either alone. A stock with 20% of its float sold short but huge daily turnover can have a low days-to-cover reading, because the exit door is wide. A stock with a modest short percentage but thin trading can post a high reading, because the door is narrow. Traders who watch for squeeze setups usually want to see both elevated, and then they want a reason for buyers to show up.
Where US short interest numbers come from
In the US, the raw input is reported by broker-dealers, not estimated. FINRA Rule 4560 requires member firms to keep a record of total short positions in all customer and proprietary accounts in equity securities and report them to FINRA. Only short positions from sales that have settled, or reached settlement date, count.
The schedule is twice a month. Per FINRA's equity short interest data page:
- Mid-month: positions held on the settlement date of the 15th, or the prior business day that settles if the 15th does not.
- End of month: positions held on the last business day of the month on which transactions settle.
- Deadline: firms report by 6 p.m. ET on the second business day after the settlement date.
- Publication: FINRA releases the compiled data on the 7th business day after the settlement date.
FINRA's 2026 calendar shows what that looks like in practice:
| Settlement date | Firms report by | FINRA publishes |
|---|---|---|
| September 30, 2026 | October 2, 2026 | October 9, 2026 |
| October 15, 2026 | October 19, 2026 | October 26, 2026 |
| October 30, 2026 | November 3, 2026 | November 10, 2026 |
| November 13, 2026 | November 17, 2026 | November 24, 2026 |
FINRA's published data includes its own days-to-cover field next to the short position and an average daily volume figure, so the ratio does not have to be hand-built for US listings.
Change may be coming. FINRA has filed a proposed rule change with the SEC (SR-FINRA-2026-012) to move short interest reporting from twice a month to weekly and cut the reporting turnaround from two business days to one. If the SEC approves it, FINRA says it will announce the effective date in a Regulatory Notice. Until then, the twice-monthly calendar above is the one in force.
A real reading: $AAPL, two cycles
Here is FINRA's own data for Apple, pulled from FINRA's public short interest dataset:
| Settlement date | Short position (shares) | Avg daily volume (FINRA) | Days to cover (FINRA) |
|---|---|---|---|
| September 30, 2026 | 120,763,539 | 39,858,260 | 3.03 |
| October 15, 2026 | 131,591,865 | 35,498,663 | 3.71 |
Check the math: 131,591,865 / 35,498,663 = 3.71. It ties out.
Now look closer. Between the two settlement dates, $AAPL short interest rose 8.97%. Days to cover rose about 22%, from 3.03 to 3.71. The extra jump came from the denominator: FINRA's average daily volume figure fell about 10.9%. Less trading made the same shorts look more crowded. That is the first lesson of this ratio. It can move a lot with no change in what the shorts are doing.
And by the standard of squeeze chatter, 3.71 is unremarkable for a mega-cap. The number says Apple shorts hold more than three average sessions of volume. It says nothing about whether they are worried.
Short interest vs. daily short sale volume
This is the most common data mistake in the space. FINRA publishes two different short datasets, and they are not interchangeable.
| Short interest | Daily short sale volume | |
|---|---|---|
| What it measures | Open short positions on firms' books | Shares sold short in trades that day |
| Frequency | Twice a month (settlement dates) | Every trade date |
| Coverage | Positions reported by FINRA member firms | Off-exchange trades reported to FINRA facilities, not consolidated with exchange data |
| Use in days to cover | Yes, the numerator | No |
FINRA's May 10, 2019 Information Notice spells out the gap with a simple case: a firm sells short 1,000 shares and buys 1,000 back later the same day. All 1,000 shares appear in the daily short sale volume file. None of it becomes a short position, and none of it shows up in short interest. FINRA also notes that some websites redistribute the daily file and call it "short interest," which it is not.
So a headline that says "60% of today's volume was short" is not saying 60% of anything is short. Some of that selling can be bought back before the close, exactly as in FINRA's example, and never become a position at all.
The pitfalls traders run into
1. The data is old by the time you see it
The October 15, 2026 positions were published October 26, 2026, eleven calendar days later. In a fast-moving name, shorts can cover, or pile in, well before the print lands. Days to cover is a photograph, and FINRA itself calls short interest a snapshot of one moment in time.
2. The volume window changes the answer
There is no single required lookback for average daily volume. Using the hypothetical stock above: with a 2,000,000-share average, days to cover is 6.0. If a longer window that includes a busy stretch averages 3,000,000 shares, the same 12,000,000 short shares produce 4.0. Same shorts, different story. When two sites disagree on days to cover, check the volume window before anything else. The $AAPL table shows that even FINRA's own volume figure shifts from cycle to cycle.
3. Float is not a fixed fact
Data vendors define float differently, for instance in how they treat insider, strategic or locked-up holdings. FINRA's investor guidance notes that different data providers use different methods to calculate and display short sale information. Back to the hypothetical: 12,000,000 shares short against a 60,000,000 float is 20%. If a vendor counts only 48,000,000 shares as float, the same position reads 25%. Neither is "wrong." They are different definitions.
4. A high reading is a condition, not a trigger
A squeeze needs crowded shorts and sustained buying that forces them out. Days to cover only measures the first part. A stock can carry a high ratio for months while the shorts are right and the price drifts lower. The catalysts traders watch for sit outside the ratio entirely: earnings, deal news, a sudden wave of call buying. That is why desks pair short data with flow. Our guide on how to read unusual options activity covers the flow side, and this look at unusual SNAP call volume shows how call activity gets framed around breakout setups.
5. Size the move with options, not the ratio
Days to cover says nothing about how far a stock might move. For that, traders look at what the options market is pricing. The expected move formula turns implied volatility into a price range, which is a cleaner read on how big a swing the market is braced for.
Desk Notes
What the numbers say: Days to cover is short interest over average daily volume. In the US it is built on broker-reported positions under FINRA Rule 4560, published twice a month, and FINRA prints the ratio itself. $AAPL read 3.71 on the October 15, 2026 settlement date.
What they don't: When the shorts entered, at what price, whether they are hedged, or when they will cover. The input is about a week and a half old on release, swings with the volume window, and says nothing about a catalyst.
FAQ
What is a "high" days-to-cover number?
There is no official threshold. The reading is most useful compared with the same stock's own history and with peers of similar size and liquidity. A mega-cap at 3 and a thinly traded small-cap at 3 are very different situations.
How often is short interest updated?
In the US, twice a month: for the mid-month and end-of-month settlement dates, with FINRA publishing on the 7th business day after each one. FINRA has proposed moving to weekly reporting, pending SEC approval.
Is daily short sale volume the same as short interest?
No. Daily short sale volume counts shares sold short in off-exchange trades on a given day, including short sales bought back the same day. Short interest counts open positions on a settlement date. Only short interest belongs in the days-to-cover formula.
Does a high days to cover mean a squeeze is coming?
No. It describes crowding, not timing. A squeeze requires buying pressure that forces shorts to cover, and many heavily shorted stocks never get one. Treat the ratio as a clue, not a verdict.
For more on options flow, volatility and short-side setups in US and Canadian stocks, visit the Wall Street War Room desk.