Short Interest Explained: Days to Cover, Percent of Float and How Short Squeezes Start

October 6, 2026

Short interest is the number of a company’s shares that have been sold short and not yet bought back. A short seller borrows shares, sells them, and hopes to buy them back later at a lower price before returning them to the lender. Until that buyback happens, the position is open, and it shows up in the short interest count.

Investors watch the figure for two reasons. It tells you how many professional skeptics are betting against a stock. And it tells you how much forced buying could hit the stock if those skeptics all try to exit at once, which is how short squeezes start. This page explains the main short interest metrics, how and when the data is reported in the US as of October 2026, what borrowing costs reveal, and what actually happened in the two most famous squeezes.

The core metrics

Raw short interest (say, 40 million shares) means little on its own. Three ratios put it in context.

Metric	Formula	What it tells you
Short interest % of float	Shares sold short ÷ float	How much of the tradable supply is already sold short
Short interest % of shares outstanding	Shares sold short ÷ total shares outstanding	Same idea, using all shares including insider holdings
Days to cover (short interest ratio)	Shares sold short ÷ average daily volume	How many days of normal trading it would take shorts to buy back

Float is the number of shares available for public trading. It excludes shares held by insiders, controlling shareholders, governments and other locked-up holders. Because float is smaller than shares outstanding, percent of float is always the higher of the two numbers, and it is the one most traders quote.

Rough rules of thumb for large US stocks: short interest under 5% of float is ordinary; 10% to 20% is high; above 20% is crowded. These are conventions, not thresholds anyone enforces.

Days to cover is about liquidity. A stock with 15 million shares short and 5 million shares of average daily volume has a days-to-cover of 3. If the same short position sat in a stock trading 500,000 shares a day, days to cover would be 30. That second stock is far more dangerous for short sellers, because there is not enough everyday selling for all of them to buy back without pushing the price up.

Short interest can exceed 100% of float. That sounds impossible but isn’t: a borrowed share that is sold short ends up in a buyer’s account, and that buyer’s broker can lend it out again. The same share can be counted on the long side more than once. GameStop reached this level in 2021.

How and when short interest is reported

In the US, short interest data comes from broker-dealers, not from short sellers themselves.

FINRA Rule 4560 requires member firms to report short positions in all equity securities, exchange-listed and over-the-counter, twice a month:

Settlement dates: the 15th of each month (or the business day before, if the 15th isn’t a settlement date) and the last business day of the month.
Reporting deadline: 6:00 p.m. Eastern on the second business day after the settlement date.
Publication: FINRA publishes aggregate short interest per security on the seventh business day after the settlement date, free on its website, and exchanges publish their own versions on a similar timetable.

The practical result is lag. By the time a figure is public, it is roughly one to two weeks old, and it shows only two snapshots a month. A lot can change in between; during a squeeze, the published number may describe a market that no longer exists.

In May 2026 FINRA filed a proposal (SR-FINRA-2026-012) to move to weekly reporting, with publication five business days after each settlement date and more detailed data. FINRA later withdrew that filing. As of October 2026, the twice-monthly schedule above still applies.

FINRA also publishes daily short sale volume, which is a different thing. It counts how many shares were sold short in trades reported to FINRA facilities that day. A large share of daily volume is marked short because market makers short routinely to fill customer buy orders, then cover within the day. A “50% short volume” headline does not mean half the stock is shorted. Short interest (open positions) and short volume (daily flow) should not be mixed up.

Rule 13f-2 and Form SHO: the stalled new layer

In October 2023 the SEC adopted Rule 13f-2, which would require institutional investment managers to file a monthly Form SHO when their short positions cross set thresholds: for a company that reports to the SEC, an end-of-day gross short position of $10 million or more, or a monthly average of 2.5% or more of shares outstanding; for non-reporting issuers, $500,000 or more on any day. The SEC would then publish aggregated, anonymized short position data per stock, including daily changes.

It has not taken effect. Compliance was first due in January 2025, then pushed to January 2, 2026. On August 25, 2025, the Fifth Circuit Court of Appeals sent Rule 13f-2 back to the SEC, along with its companion securities lending rule (Rule 10c-1a), finding the agency had not properly weighed the combined economic impact of the two. The SEC then moved compliance to January 2, 2028, with first Form SHO filings due February 14, 2028. Whether the rule arrives in its current form, or at all, depends on how the SEC answers the court.

Borrowing: locates, cost to borrow and utilization

Under SEC Regulation SHO, a broker must “locate” shares that can reasonably be borrowed before accepting a short sale. That is why the borrowing market matters as much as the short interest figure.

Cost to borrow (borrow fee): the annualized fee a short seller pays the lender, quoted as a percentage of the position’s value. Most large stocks are “general collateral” and cost well under 1% a year. Hard-to-borrow stocks can cost 20%, 50% or more than 100% a year. A fee climbing fast is a sign that lendable supply is running out.
Utilization: shares on loan ÷ shares available to lend. Utilization near 100% means almost every lendable share is already out. New short sellers can’t get shares, and existing ones risk having their loans recalled.
Hard-to-borrow lists: brokers flag stocks where locates are scarce and may refuse new shorts entirely.
Threshold securities: Regulation SHO also tracks stocks with persistent delivery failures, which trigger mandatory close-out rules for the brokers involved.

Short interest tells you how big the bet is. Borrow cost and utilization tell you how much pressure is on it right now. A stock at 25% of float with a 0.5% fee is crowded but calm; a stock at 15% of float with a 60% fee and 98% utilization is much closer to trouble.

How short squeezes start

A short squeeze happens when a rising price forces short sellers to buy, and their buying pushes the price higher, which forces more buying. Shorts face open-ended losses, since a stock can rise without limit, and they pay borrow fees and margin calls while they wait.

The usual ingredients:

High short interest relative to float.
High days to cover, so exits can’t happen quietly.
Tight borrow: high fees, high utilization, recalled loans.
A catalyst: earnings, a takeover, an index change, a large buyer, or a crowd of retail traders.
Options activity. Heavy call buying forces market makers to buy shares to hedge (a gamma squeeze, in Wall Street trading slang), which adds to the upward pressure.

High short interest alone predicts very little. Many heavily shorted stocks are shorted for good reasons and keep falling. The squeeze needs the catalyst and the closed exit together.

Case study: Volkswagen, October 2008

The Volkswagen squeeze shows what happens when float disappears.

On Sunday, October 26, 2008, Porsche disclosed that it held 42.6% of Volkswagen’s ordinary shares directly plus cash-settled options over another 31.5%, a combined 74.1%. The German state of Lower Saxony held about 20%. That left less than 6% of ordinary shares in free circulation. Hedge funds had been short roughly 12% to 13% of the stock, betting on a crisis-era decline.

The math did not work. Short positions were about twice the size of the available float. VW ordinary shares had closed near €211 on Friday, October 24. Over the next two sessions they spiked above €1,000, and on October 28 Volkswagen briefly became the most valuable listed company in the world by market capitalization. Porsche then said it would release some of its option position to ease the pressure, and the price fell back sharply within days. Losses for short sellers were estimated at more than €20 billion.

Days to cover was not the issue here. The issue was that shares to buy back barely existed.

Case study: GameStop, January 2021

GameStop is the better known case and the more misunderstood one.

At the end of 2020, GameStop’s short interest stood at about 109% of float, and it peaked near 123% in January 2021. The stock started January below $20 a share. Retail traders, organized partly on Reddit, piled in through shares and short-dated call options. GameStop closed at a record $347.51 on January 27 and hit $483 intraday on January 28. From its early-January low to that intraday high, the stock gained roughly 2,700%.

The SEC staff report on the episode, published October 14, 2021, found that short covering was real but made up only a small fraction of buying. Buying from accounts with large short positions picked up from around January 22, yet the rally lasted for weeks because of positive sentiment and new buyers, not mainly because shorts were forced out. The squeeze lit the fuse; the crowd kept it burning.

Two more details from the episode matter for how squeezes play out:

Clearing margin. On January 27, the NSCC (the clearinghouse for US stock trades) issued intraday margin calls of $6.9 billion across 36 member firms. Several retail brokers restricted purchases of GameStop and other meme stocks the next day, largely to meet those deposit demands.
Fund damage. Melvin Capital, one of the best-known shorts, lost 49% in the first quarter of 2021 and took a $2.75 billion capital injection from Citadel and Point72.

Using short interest in practice

A few working rules:

Check the date. Short interest data is stale on arrival. Pair it with borrow fees and utilization, which update daily through brokers and data vendors.
Read the ratios together. Percent of float shows crowding; days to cover shows how hard the exit is; borrow cost shows urgency.
Know why the shorts are there. Heavy short interest in a company with weak finances often reflects real problems, and short sellers are right more often than squeeze stories suggest. It can also be a hedge: convertible bond arbitrage and merger arbitrage both create short positions that have nothing to do with a bearish view.
Respect the mechanics after the spike. Both VW and GameStop gave back most of their gains once the forced buying ended.

If you trade squeeze candidates, expect extreme moves and trading pauses: single-stock volatility halts are common in these names; see how circuit breakers and LULD pauses work. Intraday margin also matters more since FINRA replaced the pattern day trader rule with new intraday margin standards. For a sense of whether crowded shorts are a market-wide theme or confined to a few names, check market breadth indicators alongside the VIX. Definitions of float, locate and margin call are also in the Referently financial markets glossary.