A short squeeze is a violent upward price spike driven by short sellers being forced to buy the stock they are short. A short position profits when price falls, but its loss is theoretically unbounded as price rises — so when a heavily-shorted name climbs, mounting losses, margin calls, and lender stock-recalls force shorts to buy-to-cover. That covering is itself demand, which pushes price higher, which forces still more shorts to cover: a reflexive, self-reinforcing feedback loop. The core tension for a trader is that the same mechanic that makes a squeeze explosive upward makes it unpredictable, untimeable, and prone to a near-total round-trip once the forced buying exhausts itself.

The ingredients

A squeeze needs forced buyers and thin supply. The commonly-cited preconditions:

• High short interest as a percent of float — a large fraction of the freely-tradable shares are sold short. The float (not shares outstanding) is what matters, because it is the actually-available supply. Reported short interest above ~20–30% of float is widely treated as "high"; GameStop's exceeded 100% of float in January 2021 (shares can be shorted more than once when re-lent), an extreme.
• Low float / illiquidity — fewer available shares means cover-buying moves price more per dollar traded.
• High days-to-cover (short ratio) — short interest ÷ average daily volume. A high reading means shorts cannot exit quickly without bidding price up, amplifying the squeeze. (See the Short Interest & Days to Cover sibling node for the metric mechanics.)
• A catalyst — earnings beat, news, an index inclusion, a takeover bid, or coordinated buying — something that starts price rising so the loop can begin.
• Forcing functions — margin calls (the broker demands more collateral or liquidates), and borrow recalls (the share lender demands its shares back, forcing a buy-in). These convert a voluntary short into an involuntary buyer.
• Optional: a gamma squeeze — heavy call-option buying forces option dealers, who are short those calls, to buy the underlying to stay delta-neutral as price rises. This mechanical, non-discretionary hedging adds a second forced-buyer engine on top of short covering, and the two can compound (per SpotGamma, Motley Fool).

How the loop works

1. A catalyst lifts price. 2. Some shorts' losses cross a pain or margin threshold; they buy-to-cover. 3. That buying lifts price further. 4. Higher price triggers more margin calls and recalls; more shorts are forced in. 5. If calls are also being bought, dealer gamma hedging buys more stock alongside. The loop runs until the forced buyers are exhausted (shorts covered, dealers hedged) — at which point the artificial demand vanishes and price typically collapses back toward pre-squeeze levels, often violently.

Famous cases (attributed)

• Volkswagen / Porsche, October 2008. Porsche disclosed control of a large VW stake via options, collapsing the available float against a heavily-shorted stock. Per Wikipedia, VW shares ran from about €210 to over €1,000 in under two days, briefly making VW the world's most valuable company before fully round-tripping. The textbook supply-shock squeeze.
• GameStop, January 2021. Reddit/WallStreetBets-coordinated buying met short interest exceeding 100% of float plus heavy call buying. Per Wikipedia, the stock hit an intraday high of $483 on Jan 28, 2021 (from roughly $17 earlier that month) before largely collapsing. Disputed cause: the SEC's Oct 2021 staff report concluded that positive sentiment, not buy-to-cover, sustained the rally — i.e. it was not primarily a short/gamma squeeze. An ad-hoc academic committee (Mitts, Battalio, Brogaard, et al., SSRN 4030179) contested this, finding a non-trivial share of volume was short covering and gamma effects. Treat the GameStop "mechanism" as genuinely unsettled, not as proof a squeeze is engineerable.

Standing & honest risk framing

Squeezes are real and recurring, but as a trade they are high-variance and gambling-adjacent unless rigorously risk-managed. The honest realities:

• Violent in both directions. The same reflexivity that spikes price up reverses on the way down; intraday swings of tens of percent are normal.
• Almost always round-trips. Forced buying is finite. VW and GME both retraced the vast majority of the move. The late chaser usually buys near the top and rides it back down.
• Untimeable. There is no reliable signal for when a squeeze ignites or peaks. High short interest can persist for months (often because shorts are correct about the fundamentals) and never squeeze.
• Crowded and reflexive both ways. "Short squeeze plays" attract crowds; the exit is as crowded as the entry. Many accounts blow up chasing them.

System relevance

For a swing trade, a heavily-shorted name with a fresh catalyst is a candidate long squeeze setup — and a dangerous place to be short. But it is a high-variance, hard-stop-mandatory play, not a core edge: position-size it small, define the stop before entry, and assume a full round-trip is the base case. Cross-link the Short Interest & Days to Cover node (the screening metrics), and the volatility and position-sizing / risk-of-ruin nodes — a squeeze candidate's realized volatility is extreme, so risk-per-trade must shrink accordingly. Downstream, the Augustus agent should flag a high-short-interest + catalyst name as squeeze-risk-elevated (both as opportunity and as a reason not to be short), never as a high-confidence directional signal.

Sources

• Investopedia / Britannica Money — short squeeze definition, short interest % of float, days to cover.
• Wikipedia, "Short squeeze" — VW/Porsche Oct 2008 (€210 → €1,000+); GameStop intraday $483 on Jan 28, 2021.
• SEC Staff Report on Equity and Options Market Structure Conditions in Early 2021 (Oct 18, 2021) — concluded sentiment, not buy-to-cover, drove the GME rally.
• Mitts, Battalio, Brogaard, Cain, Glosten, Kochuba — Ad Hoc Academic Committee report, SSRN 4030179 — contests the SEC, finds short-covering/gamma evidence.
• SpotGamma; Motley Fool ("GameStop's Gargantuan Gamma Squeeze"); TradingSim — gamma squeeze / dealer delta-hedging mechanics.