thetrading.toolsSubscribe
StocksBi-monthly · FINRA publishes ~2 weeks after settlement

Short Interest & Squeeze Watch: Who the Market Is Betting Against

Every US-listed stock's short interest from FINRA's consolidated bi-monthly filings — the actual standing bets, not daily short-sale flow. First the whole market: how much of All US stocks, the S&P 500, the Nasdaq-100 and every Nasdaq-listed stock is sold short, at every settlement since 2017. Then each stock: percent of shares short, days-to-cover and its percentile across the market, and a frozen study answering the question every squeeze thread skips: do high-short-interest stocks actually squeeze? Cross-check live action on Unusual Volume and the options tape.

Latest settlement

At the October 30, 2026 settlement, $2.04 trillion of US-listed stock was sold short, 2.22% of the market value of the 4,637 stocks counted. That ranks in the 95th percentile of 213 settlements since December 2017; the high was 2.48% on June 15, 2026. Since July 31, 2026 it has fallen 0.06 points. The median stock had 4.3% of its shares sold short. FINRA publishes each settlement about two weeks after it.

Sources, methodology & freshnessFINRA consolidated short interest (all US exchanges, Rule 4560), joined to our own price and share-count data · Bi-monthly settlements (15th and month-end), published ~2 weeks later; checked daily in the pipelineData as of 2026-11-20 · Open ↓
Source
FINRA consolidated short interest (all US exchanges, Rule 4560), joined to our own price and share-count data
Methodology
SI% = shares short ÷ split-consistent shares outstanding; days-to-cover as published by FINRA; tables floored at $300M market cap; frozen quintile study with $300M/$5 floors
Updates
Bi-monthly settlements (15th and month-end), published ~2 weeks later; checked daily in the pipelineData as of 2026-11-20
Maintained & reviewed by Yuriy Matso — methodology shown on the page.
Short Interest · whole marketsettled 2026-10-30
2.22%
of US market value sold short

$2.04 trillion short across 4,637 US-listed stocks, 95th percentile since December 2017. Since July 31, 2026 it has fallen 0.06 points.

UniverseOf mkt valueMedian stockPctile
All US2.22%4.3%95th
S&P 5001.92%3.0%97th
Nasdaq-1001.65%2.5%88th
Nasdaq-listed2.29%4.5%76th

Short value over market value, summed across each universe. Median stock: shares short over shares outstanding. Percentile among 213 settlements since December 2017.

01

Short interest across the market — four universes

At the October 30, 2026 settlement, All Nasdaq-listed stocks carried the most short interest for their size (2.29% of market value) and Nasdaq-100 members the least (1.65%). In every universe the median stock is shorted more heavily than the universe as a whole, because the largest companies, which dominate a sum of market value, carry the least. Dollars short and market days to cover need no share count; the percentages divide by shares outstanding.

UniverseStocksDollars shortOf market valuevs priorvs a year agoMedian stockOver 10%Days to cover
All US stocks4,637$2,036B2.22%−0.02 pts+0.29 pts4.26%18.0%3.04
S&P 500 members501$1,299B1.92%+0.03 pts+0.34 pts3.02%5.0%2.67
Nasdaq-100 members100$646B1.65%+0.08 pts+0.31 pts2.54%1.0%2.20
All Nasdaq-listed stocks2,683$1,006B2.29%+0.01 pts+0.25 pts4.46%21.2%2.77
Of market value: dollars sold short over market value, summed across the universe. Median stock and over 10%: shares short over shares outstanding. Days to cover: dollars short over dollars traded on an average day (FINRA's average daily volume). Changes are in percentage points.
02

The record since December 2017

Every twice-monthly settlement, one line per universe. Tap a name in the legend to hide its line. All US stocks reached 2.48% of market value on June 15, 2026, its high in the record, and read its low of 1.49% on January 14, 2022.

Range:

Short value as a share of market value

1.1%1.9%2.6%201820192020202120222023202420252026SPY (top pane, log)7562.22%1.92%1.65%2.29%
Dollars sold short over market value, summed across each universe, at each FINRA settlement. Large companies weigh most. S&P 500 and Nasdaq-100 are today's members carried back.
Range:

The median stock: shares short as a share of shares outstanding

1.3%2.9%4.5%201820192020202120222023202420252026SPY (top pane, log)7564.26%3.02%2.54%4.46%
The middle stock of each universe at each settlement, so a few very large companies do not set the reading. S&P 500 and Nasdaq-100 are today's members carried back.
Range:

Dollars sold short

103.8B1197.8B2291.8B201820192020202120222023202420252026SPY (top pane, log)7562036B1299B646B1006B
FINRA short positions times the price as traded on the settlement date, in billions of dollars. Rises with prices as well as with positions. S&P 500 and Nasdaq-100 are today's members carried back.
Range:

Market days to cover

12.84.7201820192020202120222023202420252026SPY (top pane, log)7563.042.672.202.77
Dollars sold short over dollars traded on an average day: how many sessions of normal volume it would take to buy every short position back. S&P 500 and Nasdaq-100 are today's members carried back.
03

Where the dollars are

The ten stocks with the most dollars sold short at the October 30, 2026 settlement, and their share of the $2.04 trillion total across US-listed stocks. Together they hold 14.1% of it. Large companies lead in dollars even when only a small share of their stock is sold short.

SymbolDollars shortShare of totalOf its shares
NVDA$57.2B2.8%1.16%
AAPL$49.9B2.5%1.00%
MSFT$31.8B1.6%0.82%
TSLA$26.4B1.3%1.67%
GOOGL$24.5B1.2%1.49%
AMZN$22.7B1.1%0.87%
MU$20.7B1.0%2.07%
AMD$19.5B1.0%2.35%
AVGO$16.9B0.8%1.09%
GOOG$16.6B0.8%1.07%
04

Does market-wide short interest time anything?

The test: at each settlement, take the change in short value over market value across the last 6 settlements (about three months) and rank it against the changes before it. Then grade the universe's index fund over the next 63 sessions, starting on the day FINRA published the settlement, never on the settlement itself. The level is not tested: it rose for most of the record, so ranking it would mostly mark the calendar. Settlements from March 31, 2020 to July 31, 2026. The widest gap: after shorts built fastest on Nasdaq-listed, QQQ's median quarter was 3.2 points weaker than after every settlement, on 7 independent runs, too few to rely on. S&P 500 and Nasdaq-100 are left out because their members are today's.

Universe · fundSettlementsCount (runs)Median next 63 sessionsA 10% fall within
All US · VTIShorts building fastest (top fifth)45 (7)+3.9%11%
Shorts covering fastest (bottom fifth)11 (5)+4.1%18%
Every settlement153+4.9%12%
Nasdaq-listed · QQQShorts building fastest (top fifth)53 (7)+3.0%19%
Shorts covering fastest (bottom fifth)20 (5)+6.9%10%
Every settlement153+6.3%19%
A run is a group of settlements no more than 6 apart; consecutive settlements share most of their forward window, so runs are the closer count of independent cases. A 10% fall is a close at least 10% below the starting close within the window. Price returns. All US today: the 20th percentile of past three-month changes. Nasdaq-listed today: the 20th percentile of past three-month changes.
05

How these readings are built

  • Short positions are FINRA's consolidated short interest: every broker's short positions in every stock on every exchange, settled on the 15th and the last business day of each month and published about two weeks later.
  • All US stocks are the common stocks and ADRs listed on a US exchange on each settlement date, by a dated listing record (funds, preferreds and over-the-counter shares are left out). Nasdaq-listed is the same rule on Nasdaq's own markets, by FINRA's listing code that day, which is what the Nasdaq Composite holds. Both are counted point in time.
  • S&P 500 and Nasdaq-100 are today's members carried back: we hold no membership history, so their past readings are for today's companies and feed no study.
  • A stock counts only with one of our price files and a close within 7 days of the settlement. Those files hold few companies that later failed, and heavily shorted companies fail more often, so the older readings run somewhat low.
  • Dollars use the price as traded on the settlement date, since FINRA reports positions in the shares of that day; later splits are taken back out of our adjusted closes.
  • Shares outstanding come from each company's SEC cover page for the quarter, so past percentages use the share count of their time. ADRs and companies with several share classes use today's count from a market data vendor, which ignores issuance and buybacks since. The percentages divide by shares outstanding; many vendors divide by the float, which is smaller, so their figures read higher. At the latest settlement 4 stocks read above 100% of their shares and were left out of the percentages.
06

Do high-short-interest stocks actually squeeze?

The honest test, frozen before this page shipped: at each of 201 settlement dates, stocks above $300M market cap and $5 were bucketed into short-interest quintiles, and each bucket's forward 63-session returns were graded. The answer cuts both ways. The typical high-SI stock does not squeeze — the top quintile's median forward return is 0.1% with a 50% win rate, a coin flip. But the squeezes live almost entirely there: 5.8% of high-SI names gained +50% within the quarter, versus 1.7% of low-SI names — and that fat right tail drags the top quintile's average (3.4%) above every other bucket. High short interest buys squeeze exposure, not expected return.

SI quintileAvg fwd 63sWin rate+50% squeeze rate
lowest SI2.06%53.9%1.74%
Q22.82%55.4%1.71%
Q32.95%54.6%2.22%
Q43.15%52.9%3.53%
highest SI3.36%50.1%5.78%
201 settlements since 2017; SPY averaged 3.3% over the same windows. Bi-monthly windows overlap and stock-quarters are not independent samples; the study is frozen and only re-run deliberately.
07

Most shorted stocks — SI % of shares

The 30 highest short-interest names above $300M market cap at the latest settlement. Readings above ~50% usually mix directional bets with arbitrage positions (convertibles, merger spreads) — check the story before reading pure conviction.

WOLF112.0%
DTC 8.9$557M$497M
ONDS63.8%
DTC 6.2$925M$5.3B
NUAI62.2%
DTC 3.1$126M$431M
FCEL49.4%
DTC 3.7$158M$474M
AVTX48.1%
DTC 2.1$60M$301M
EOSE47.6%
DTC 5.5$163M$5.9B
USAR43.8%
DTC 4.8$364M$3.2B
GOSS43.5%
DTC 7.5$14M$664M
SERV42.7%
DTC 10.1$65M$987M
SOUN42.3%
DTC 12.4$427M$4.6B
CAPR41.0%
DTC 7.4$142M$1.3B
HTZ40.4%
DTC 5.7$140M$1.8B
ALT40.4%
DTC 12.2$56M$561M
SLS40.1%
DTC 6.7$780M$768M
PCT38.4%
DTC 10.7$109M$2.2B
NTLA38.3%
DTC 11.1$444M$1.8B
WHR37.5%
DTC 3.5$520M$5.0B
TE37.2%
DTC 4.5$87M$1.8B
TNGX35.8%
DTC 9.7$1.05B$1.7B
SOC35.6%
DTC 9.0$27M$1.8B
UWMC34.8%
DTC 5.9$99M$1.6B
OCGN34.5%
DTC 16.4$83M$515M
SEI34.4%
DTC 5.9$919M$2.8B
SMR34.4%
DTC 3.6$327M$5.8B
BKKT34.4%
DTC 8.2$20M$396M
NNE34.0%
DTC 10.1$156M$1.8B
BTDR33.7%
DTC 6.0$177M$3.3B
NKTR33.3%
DTC 10.3$204M$743M
CABO33.3%
DTC 2.4$4M$470M
LEU33.1%
DTC 6.4$472M$5.5B
08

Slowest exits — days to cover

Short interest divided by average daily volume: how many normal sessions of buying it would take for every short to cover. Squeezes need this — a crowded short with huge daily volume can unwind quietly.

ISOU5.3%
DTC 33.3$16M$640M
LUXE2.7%
DTC 31.3$38M$1.2B
TY1.6%
DTC 28.7$26M$1.7B
OTTR11.1%
DTC 27.3$356M$3.7B
EVCM1.0%
DTC 25.2$12M$2.3B
DMAC9.6%
DTC 24.8$41M$434M
IMO3.0%
DTC 24.6$1.68B$49.3B
ZBIO21.1%
DTC 24.1$271M$1.1B
LUCK1.5%
DTC 24.0$9M$1.3B
RYTM17.5%
DTC 23.3$1.31B$7.2B
PROK17.2%
DTC 22.5$25M$309M
DAO0.5%
DTC 20.8$11M$1.3B
RLAY25.2%
DTC 20.7$703M$1.3B
MAIN10.2%
DTC 20.3$504M$5.7B
FTK14.3%
DTC 19.8$93M$531M
09

Biggest changes vs the prior settlement

The largest swings in short interest among names already carrying 5%+ SI — building bets and covering in progress, per FINRA's own change calculation.

NMZ7.0%
DTC 2.9+1669% chg$1.2B
HGV16.3%
DTC 11.2+82% chg$4.0B
STNE5.7%
DTC 1.9+66% chg$4.2B
BHE5.8%
DTC 4.2+58% chg$1.8B
FTAI9.4%
DTC 4.9+56% chg$30.1B
KDK7.9%
DTC 1.7+50% chg$1.7B
APAM6.2%
DTC 3.6+49% chg$3.1B
EYPT13.3%
DTC 1.0-48% chg$1.4B
TARA6.3%
DTC 2.9-47% chg$317M
HTO7.9%
DTC 2.4+46% chg$1.9B
ABVX5.2%
DTC 4.6-43% chg$9.7B
S10.2%
DTC 2.9+42% chg$4.8B
RGR7.0%
DTC 1.8-39% chg$597M
SFNC5.7%
DTC 4.1+39% chg$3.1B
SGRY6.9%
DTC 1.1-38% chg$2.0B

How Short Interest & Squeeze Watch Works

  1. 1
    Pull every bi-monthly settlement from FINRA
    FINRA member firms must report short positions in all equity securities twice a month (Rule 4560), with settlements on the 15th and the last business day, published about two weeks later. We pull the consolidated dataset — all exchanges, not one venue — for every settlement since 2017.
  2. 2
    Join to our own price and share data
    Each snapshot is joined to our US common-stock universe. Short interest percent = shares short ÷ shares outstanding, using split-consistent share counts derived from market cap and adjusted prices. Days-to-cover is FINRA's own figure: shares short ÷ average daily volume.
  3. 3
    Add it up for four universes
    At every settlement we sum dollars sold short and market value for All US stocks, S&P 500 members, Nasdaq-100 members and Nasdaq-listed stocks, and record the median stock's short interest, market days to cover and the share of stocks above 10% and 20%. All US and Nasdaq-listed are counted point in time from a dated listing record; the two indexes are today's members carried back. Dollars use the price as traded on the settlement date, and past share counts come from each company's SEC cover page for that quarter.
  4. 4
    Grade the squeeze folklore, then freeze it
    At each settlement date, stocks above a $300M market cap and $5 price are bucketed into short-interest quintiles; forward 63-session returns, win rates and +50% squeeze frequencies are graded per bucket. The study is frozen — it re-runs only deliberately, never silently.

Who Uses Short Interest & Squeeze Watch

Squeeze-narrative skeptics
When a ticker trends as "the next big squeeze," look it up: its actual SI%, days-to-cover and percentile are here from FINRA filings rather than social-media screenshots — and the settlement date tells you how stale the number is.
Single-name traders
The top-SI and biggest-change tables surface where bearish positioning concentrates and where it is building or covering fastest, twice a month. A large SI drop alongside a rally is covering; a build into weakness is conviction.
Market-regime readers
The market section shows whether shorts are leaning in or backing off across the whole market and in the large-cap indexes, side by side. Our test found no timing signal in it, so read it as positioning, not as a forecast.
Squeeze-lottery hunters
The study quantifies the trade-off: the typical high-SI stock went nowhere over the next quarter (median ~0%, coin-flip win rate), but the odds of a +50% move were about three times a low-SI stock's. High short interest buys squeeze exposure, not expected return.

Pro Tips

01
Squeezes need both ingredients
Days-to-cover above ~10 means even modest covering pressure meets thin volume — a real squeeze setup pairs high SI% with high DTC. One without the other is a weaker setup.
02
Read the change column first
A 20%+ jump in short interest at one settlement is fresh conviction; a 20% drop into strength is a squeeze already in progress. The level tells you where the crowd is; the change tells you what it is doing.
03
The print is always stale in a fast squeeze
Settlement data publishes ~2 weeks late, so during a squeeze the posted number describes the world before it started. Treat it as positioning context and use live volume for timing.
04
Interrogate readings above ~50%
Extreme SI% usually involves convertible arbitrage, merger spreads or share-count drift alongside directional bets — check the situation before reading it as pure crowd conviction.

Common Issues & Solutions

The SI% here differs from my broker's number▾
We divide by shares outstanding; most broker sites divide by float, which is smaller and produces higher percentages. Neither is wrong — they are different denominators, and ours is stated on every table.
A stock shows an implausibly high SI%▾
The stock tables take share counts from our market-cap snapshot, made split-consistent, so recent offerings or buybacks can drift a name's denominator until the next snapshot; readings above 150% are excluded from those tables. The market section uses SEC cover-page counts and leaves any stock above 100% out of its percentages.
A thin stock shows no days-to-cover▾
FINRA reports 999.99 as a sentinel for near-zero-volume names; we treat those as missing rather than ranking them as extremes.
A famous squeeze never appeared in the data▾
The series is bi-monthly: a position opened and covered between settlements never prints. That is a limitation of the source, and it is why the page pairs with live volume and options tools.

Frequently Asked Questions

How much of the stock market is sold short?▾
A low-single-digit share of its market value. The answer card at the top of this page gives the latest figure for All US stocks, the S&P 500, the Nasdaq-100 and Nasdaq-listed stocks, each summed as dollars sold short over market value, with where it ranks among every settlement since 2017. The median stock carries a higher share than the market as a whole, because the largest companies are shorted least in proportion to their size.
Does rising short interest predict a market decline?▾
Not in our test. The settlements where market-wide short interest rose fastest over three months, and those where it fell fastest, were followed by quarters close to the ordinary one on the broad market, on a handful of independent runs. The table on the page shows the counts and returns from the day FINRA published each settlement.
What is short interest?▾
The total number of shares investors have sold short and not yet covered, reported to FINRA by member firms twice a month for every US equity. Expressed as a percent of shares outstanding, it measures how heavily the market is positioned against a stock.
What counts as high short interest?▾
The median US common stock runs a low-single-digit percent of shares short. Above 10% is elevated (roughly the top sixth of the market), above 20% is heavily shorted (top ~5%), and readings above 50% are rare and usually involve arbitrage positions or share-count quirks alongside directional bets.
What is days-to-cover and why does it matter?▾
Short interest divided by average daily volume — how many typical sessions of buying it would take for every short to exit. High days-to-cover means covering is slow and crowded; squeezes need it, because high SI% with huge daily volume can unwind without fireworks.
Do high-short-interest stocks actually squeeze?▾
Occasionally — and that is the honest answer. In our study across every settlement since 2017, the typical highest-quintile stock returned about zero over the next quarter with a coin-flip win rate, but its chance of a +50% move was roughly triple a low-SI stock's. High short interest buys squeeze exposure, not expected return.
How often is short interest data updated?▾
Twice a month. Settlements fall on the 15th and the last business day, and FINRA publishes the consolidated data roughly two weeks later; this page updates the same day each new settlement publishes.
Is this the same as the daily short volume data?▾
No. Daily short-sale volume files count shares sold short during a day (most of it market-making flow, closed intraday); this dataset is open short positions outstanding at settlement — actual standing bets. The bi-monthly positions series is the one squeeze analysis needs.

Explore Other Tools