Short Interest Metrics Explained

ORTEX publishes a family of short interest metrics that together describe how heavily a stock is shorted, how expensive and difficult it is to short, and how those conditions are changing. This guide explains what each metric means and how to interpret it. For where the underlying data comes from — and why so much of it is delayed or partial — read How Short Interest Is Reported Around the World first.

📘
Key takeaways

ORTEX Estimated Short Interest is an estimate of total shares short, anchored to official figures and updated daily and intraday with securities lending market activity — always read it together with its confidence bands.

Days to Cover translates short interest into time: how many days of normal trading volume it would take shorts to exit.

Cost to Borrow and Short Availability describe the borrowing market: what shorting costs and how much lendable supply remains.

Short Score condenses several short-positioning dimensions into a single 0–100 comparison across all stocks.

No single metric tells the whole story. Elevated short interest with cheap, plentiful borrow is a very different situation from the same short interest with expensive, scarce borrow.

ORTEX Estimated Short Interest

What it is. ORTEX's estimate of the total number of shares sold short in a stock — published daily, and updated live intraday as securities lending activity is recorded through the day. It is published as a share count, as a percentage of free float, and as a value in the stock's currency.

Why an estimate? Official short interest is either delayed (a US figure is one to three weeks old by the time it's published) or incomplete (European disclosures only capture large positions). The estimate starts from the most recent official figure and moves it forward day by day using observed activity in the securities lending market — the market where shares are borrowed before being sold short. When the next official figure arrives, the estimate is recalibrated against it.

Confidence bands. Every estimate is published with an upper and lower bound. The bands widen as time passes since the last official anchor and narrow when the estimate is well-calibrated for that stock. Treat the band, not the center line, as the honest answer: "estimated short interest is 43M shares, plausibly between 38M and 48M" is the correct reading.

Things to know.

Expect a visible adjustment the day after a new official figure is published — the model course-correcting against ground truth.
Share counts are adjusted across stock splits, so the history remains comparable through corporate actions.

Not all borrowing is short selling

Securities lending activity is a powerful signal, but it has a well-known impurity: shares are borrowed for reasons other than shorting. The clearest example is dividend arbitrage. Around a dividend's ex-date, some investors temporarily lend their shares to a counterparty who is taxed more favorably on the dividend; the two share the tax saving, and the shares flow straight back after the ex-date. The result is a large, short-lived spike in lending that has nothing to do with anyone betting against the stock — Canadian stocks show this pattern especially strongly, with massive lending peaks around ex-dividend dates.

Left uncorrected, these spikes would masquerade as sudden jumps in short interest. ORTEX identifies dividend-driven lending and factors it out of the estimation, so the Estimated Short Interest isn't incorrectly inflated simply because a stock paid a dividend.

Official Reported Short Interest

The authoritative figure published by each market's regulator — FINRA's twice-monthly print in the US, daily figures in Australia, weekly in Hong Kong, and so on. In European and UK stocks the equivalent is flagged shorts: the sum of publicly disclosed positions above the 0.5% threshold, which is a floor on true short interest rather than a complete count. Schedules, delays, and caveats for every market are covered in the reporting regimes guide.

On ORTEX you'll see both the official history and the daily estimate — the official prints are the anchor points; the estimate is the line connecting and extending them.

Days to Cover (DTC)

What it is. Short interest expressed as time: the number of days it would take all short sellers to buy back their positions if they consumed the stock's average daily trading volume. Also known as the short interest ratio.

Days to Cover = Estimated Short Interest ÷ Average Daily Volume

Why it matters. A large short position in a heavily traded stock can be unwound quickly and quietly. The same position in a thinly traded stock takes far longer to exit — which is what makes crowded exits violent. DTC captures this: higher values mean shorts would need more time (and likely more price impact) to cover.

The four windows. Average daily volume depends on the period you average over, and a single unusual high-volume day can flatter the figure. ORTEX therefore publishes DTC against four volume windows — 1 week, 2 weeks, 1 month, and 3 months. If the 1-week DTC is far below the 3-month DTC, recent volume has surged (each day of that volume retires more short interest); if it's far above, volume has dried up. Values are capped at 999 to keep extreme illiquidity from distorting charts.

Cost to Borrow (CTB)

What it is. ORTEX's estimate of the annualized fee a short seller pays to borrow the stock, expressed as a percentage. It is derived from lending rates observed across institutional securities-lending activity combined with retail brokerage borrow rates.

All vs. New. Two variants are published: CTB All — the weighted average rate across all currently outstanding loans — and CTB New — the weighted average rate on loans initiated that day. CTB New is the marginal price: what it costs to open a new short position today. When CTB New runs well above CTB All, borrowing is getting more expensive; existing shorts are grandfathered at older, cheaper rates while new entrants pay up.

Reading it. For most large, liquid stocks CTB is well under 1% — borrowing is a non-issue. Elevated CTB signals scarce borrow supply and/or heavy demand to short, and it is a direct, ongoing cost that erodes a short position's profitability: a 50% annualized borrow fee means a short seller needs the stock to fall substantially just to break even over a year.

Yes, it can be negative. Borrowers post collateral against loans, and depending on the collateral arrangements and prevailing interest rates, the net rate can go below zero — effectively the lender paying the borrower. It's uncommon but not an error.

Short Availability

What it is. A measure of how much lendable supply remains relative to current borrowing demand — how much room is left to short, expressed as the percentage of additional supply available.

Reading it. This metric is most informative when it is low. A value near zero means the available lending supply is nearly exhausted: the stock is hard to borrow, new shorts are difficult to establish, and conditions can tighten quickly. In that range, day-to-day changes are meaningful.

At the other end, don't over-read large values. Easily borrowed stocks routinely show availability in the thousands of percent, and the difference between 5,000% and 30,000% carries no practical information — both simply mean "easy to borrow." The metric earns its place on your screen when it heads toward zero, not while it's large.

Securities Lending Volume

What it is. The total daily activity in the stock's lending market: the sum of shares newly lent and shares returned to lenders on that day.

Reading it. This is a measure of turnover in the borrowing market, not a level of short interest — new loans and returns both add to it, so a spike tells you the borrowing market is busy without telling you the direction on its own. Read it alongside the Estimated Short Interest: rising lending volume with rising estimated short interest suggests new short positioning being built; rising volume with falling short interest suggests covering and returns. Quiet lending volume in a heavily shorted name means positions are simply being held.

Short Seller Gains/Losses

What it is. A daily estimate of the aggregate profit or loss of short sellers in the stock, in the stock's currency, based on the day's close-to-close price move applied to the estimated short position.

Reading it. On a down day shorts gain; on an up day they lose. The value is mark-to-market — what the day's move did to open positions — not realized profit, since most positions remain open. It's most useful cumulatively: a stock whose short sellers have absorbed sustained, large mark-to-market losses is one where covering pressure may be building, especially when borrow is expensive and Short Availability is low — the classic preconditions of a squeeze.

Short Score

What it is. A single 0–100 score summarizing how elevated a stock's short positioning is relative to all other stocks, combining several dimensions of short interest and borrowing conditions into one comparable number. The exact inputs and weights are proprietary.

Reading it. A score around 50 means the stock is close to the market average across the underlying dimensions. Scores well above 50 mean the stock is elevated across multiple dimensions simultaneously — which is precisely what makes it interesting: a stock can have high short interest but cheap, plentiful borrow, or expensive borrow with modest short interest. A high Short Score means several of these flags are raised at once. It's a screening and comparison tool, not a prediction.

US-only official datasets

These come directly from US regulators and are available alongside the metrics above for US stocks.

FINRA Short Volume and Short Exempt. The aggregate volume of trades executed and marked as short sales each day. This is a flow, not a position — and it is routinely much larger than short interest, because it includes vast market-making activity (a market maker selling stock it hasn't yet received marks the trade as short). FINRA itself cautions that daily short sale volume "does not — and is not intended to — equate to" reported short interest. A stock trading 60% daily short volume is not "60% shorted." Use it as a gauge of short-selling activity, never as a substitute for short interest.

Failure to Deliver (FTD). The number of shares that were sold but not delivered by the settlement date, published by the SEC twice a month with a significant delay (each half-month's data arrives roughly two to four weeks later). Fails can result from short selling, but also from long sales, processing errors, and settlement mechanics — the SEC is explicit that FTDs are not proof of naked short selling. FTD figures are snapshots of outstanding fails and must not be summed across dates.

Threshold List. A daily regulatory list of securities with persistently large fails: aggregate FTDs of at least 10,000 shares and 0.5% of shares outstanding for five consecutive settlement days. Inclusion triggers stricter borrowing requirements for brokers, and a stock leaves the list after five consecutive days back under the threshold. The SEC notes that inclusion "should not be interpreted as connoting anything negative" about the issuer by itself.

Free float and shares outstanding

Several metrics above are expressed as a percentage of free float — the shares actually available for trading, which excludes insider holdings, strategic stakes, government holdings, and other locked-up shares. Free float has no industry-standard definition, so figures legitimately differ between data providers; ORTEX uses a float methodology aligned with S&P's, built from regulatory filings.

Percentages of free float use the float as it was recorded at the time — a historical figure, so the percentage history stays truthful as the float changes. Note that before September 2021, float history was captured quarterly; from September 2021 it is captured daily.

Common questions

Why did the estimate move sharply when the price barely did?
The estimate reacts to securities lending activity and official prints, not to price. A recalibration after a new official figure, or a genuine surge in borrowing, can move the estimate on a quiet trading day.

A stock traded 50%+ short volume today — why isn't short interest exploding? Isn't something shady going on?
Neither, in almost every case. Short volume and short interest measure different things: volume is a flow — shares sold in trades marked as short during the day — while short interest is a position — shares still held short at a point in time. A large share of daily short volume comes from routine market making: a market maker filling a customer's buy order often sells shares it hasn't yet acquired, and that sale must legally be marked short even though the exposure is typically hedged or closed within minutes. Those trades never become lasting short positions. It is completely normal for a stock to print 40–60% short volume day after day while short interest barely moves — that pattern is neither short sellers piling in nor evidence of manipulation; it's the mechanics of how markets are made. See FINRA Short Volume and Short Exempt above, including FINRA's own caution that the two datasets are not comparable.

What happened to Utilization, Shares on Loan, and Free Float on Loan?
ORTEX no longer publishes these securities-lending metrics. Lending market activity remains a core input to the Estimated Short Interest, and borrowing conditions are covered by the metrics that replaced them on the platform: Cost to Borrow and Short Availability.

Is a high short interest bullish or bearish?
Neither, by itself. It tells you positioning, not direction: heavily shorted stocks attract sophisticated sellers who are often right — and simultaneously carry squeeze potential if sentiment turns and borrow is tight. Context (Days to Cover, Cost to Borrow, Short Availability, and your own fundamental view) is everything.

Do dividends distort the data?
In some markets, shares are borrowed around dividend dates for tax reasons that have nothing to do with short selling — a practice called dividend arbitrage, seen especially strongly in Canadian stocks. ORTEX identifies these events and factors them out so the Estimated Short Interest isn't inflated by them. See Not all borrowing is short selling above for the full explanation.

Related guides

How Short Interest Is Reported Around the World — the reporting regimes behind the official figures

📘
Coverage and updates
Estimated Short Interest, Days to Cover, Cost to Borrow, Short Availability, Securities Lending Volume, Short Seller Gains/Losses, and Short Score update every trading day, and the Estimated Short Interest additionally updates live intraday. Official reported short interest updates on each regulator's schedule (see the reporting regimes guide). FINRA short volume is daily; SEC FTD data is published twice monthly in arrears.

ORTEX data guides are for information and education only and are not investment advice. Data is provided "as is" and may contain errors or omissions; see our terms of service for details.