Ball's Q3 Miss Sends Shorts to a Year High and the Street Scrambling

Ball Corporation fell 7.1% on November 3 to $63.21, its worst single-day drop in months, arriving on the same morning that packaging peers closed sharply higher, and Q3 earnings land on November 5 with the stock already at a year-high for short interest.

Short interest has climbed 21% over the past month to 3.5% of free float, roughly 9.4 million shares. That is the highest level in at least a year. The build has been methodical: it started in early October, accelerated through late October, and held even as the stock briefly rallied to $66.47 last week before Tuesday's sell-off reversed those gains. The borrowing side of the trade carries no friction. Cost to borrow is 0.49%, essentially flat on the week, and availability is wide at 2,205% of shares short. With more than 175 million shares available to borrow against roughly 9 million already borrowed, there is no mechanical pressure on the short position. Shorts can add, hold, or exit without competing for supply. Options positioning does not contradict that calm: the put/call ratio is 0.63, slightly below its 20-day average of 0.70, meaning options traders are actually less defensively positioned than usual, a mild divergence from the direction the short book has been moving.

The sell-off on November 3 is striking in peer context. Crown Holdings fell 2% on the day but is up nearly 4% on the week. AptarGroup gained 12.5% on the week. Silgan Holdings added 7.8%. Amcor rose 2.2%. Ball gave back 1.4% over the same five sessions. The pattern that two previous ORTEX notes flagged, shorts building while packaging peers rallied, has now resolved into an outright underperformance gap.

Analyst reaction to the print has arrived quickly and mostly in one direction. JP Morgan's Jeffrey Zekauskas downgraded Ball to Neutral this morning, keeping his $65 target unchanged, a move that puts one of the more recently constructive voices on the sidelines. Truist lowered its target from $78 to $76 while holding its Buy rating. Mizuho, by contrast, raised its target from $72 to $75 and kept its Outperform call. The consensus mean target is $71.62 against a current price of $63.21, implying around 13% nominal upside, but that figure will compress as more desks refresh models. The bull case rests on specialty can mix and energy drink volume growth in North America. The bear case centres on aluminum cost pressure, the floating-rate debt burden, and declining carbonated soft drink volumes, which remain a substantial part of Ball's business. At 14.3x trailing earnings and an EV/EBITDA of 10.0x, the multiple has compressed on the day but had been expanding over the prior month as the stock climbed above $66.

One activist-register note worth carrying: Parnassus Investments filed a Schedule 13D/A in April 2025 disclosing a 4.9% stake, down from 5.6% previously. The 13D designation makes it a named activist position on ORTEX's EDGAR register. Parnassus's most recent filing shows it trimmed below the 5% threshold. Under 13D/G disclosure rules, a holder that drops below 5% has no obligation to file again, so the current position is unknown. That said, the filing history suggests a formerly engaged holder that has been reducing.

The alt data layer adds one data point worth noting: Ball's buyback spend in Q2 2026 was $114 million, down 76% year-on-year and the smallest Q2 buyback in two years. That is a corporate-capital-allocation fact, not a revenue signal, but it reinforces the picture of a company husbanding cash rather than returning it at prior rates. Retail attention, as measured by Wikipedia page views and ORTEX page traffic, ran at a z-score of 2.6 relative to the prior 90 days as of October 22, suggesting the stock drew more than usual public interest in the run-up to earnings.

The print itself is due November 5. The specific thing to watch is whether management provides any update on aluminum cost pass-through mechanisms and the floating-debt exposure, the two inputs JP Morgan cited this morning as the basis for its downgrade.