CENX shorts ease but Street targets remain miles above the tape

Short positioning tells a less dramatic story than the price action implies. Bears have actually been trimming. Short interest has fallen from a peak above 10.4 million shares in mid-September to 8.83 million shares, leaving it at 9.5% of free float. That is a meaningful level, but the trend is toward covering rather than piling on. The borrow market is relaxed: availability is running at 843%, meaning there are more than eight shares available to lend for every one already borrowed. Cost to borrow is just 0.51%, well within normal ranges. With borrowing this cheap and this easy, the absence of a more aggressive short build during a 22% price decline is itself a signal that conviction on the bear side is not rising. Options positioning corroborates this: the put/call ratio is 0.61, actually running below its 20-day average of 0.65, nearly 1.2 standard deviations on the call-heavy side. Options traders have been less defensive than usual even as the shares have sold off.

The ownership structure adds context that the short interest alone cannot capture. Glencore International AG holds a Schedule 13D activist stake, last disclosed at 29.99% of shares outstanding, down from 36.6% at its prior filing in November 2025. The 13D designation matters: this is not a passive index position. Glencore's disclosure was filed in March 2026, and per the SEC's event-driven filing rules, the exact current size of the stake is unknown until a new amendment is filed. As always with 13D/G positions, stakes are as last disclosed around the 5% threshold, and a holder dropping below 5% may not file again. Alongside Glencore, BlackRock recently crossed 10.5% and filed a fresh 13G amendment in early October, adding 667,000 shares in the quarter to September. State Street also filed fresh, at 5.6%. The institutional flow among passive and quasi-passive holders has been broadly supportive even as the stock has fallen.

Wikipedia traffic for CENX has spiked well above its own recent history, with the retail attention signal running at a z-score of 2.76 against the prior 90 days. Elevated retail attention during a sharp price decline often reflects search activity driven by the move itself rather than fundamental interest, and there is no measured alt-data series for CENX that has been shown to lead its reported figures, so this reading is colour rather than signal.

The last two earnings prints are worth noting for context. CENX moved 12.9% higher on the day after its August 2026 Q1 report and 9.2% higher after Q2. Both were strong day-one reactions. Both also faded over the following week, with five-day returns of just 1.7% and minus 1.1% respectively. With earnings now 23 days away, the question heading into November 6 is whether a stock that has re-rated 22% lower in a month, with analysts still anchored to targets well above the tape, reflects accumulated pessimism that a decent print could quickly reverse, or whether the move signals something about aluminum market conditions that the Street has not yet fully reflected in its numbers.