AnalysisMarkets

The $196mn Nvidia put trade raises a question about timing that nobody can yet answer

An options position opened on 12 October has gained roughly 24 times its cost as Nvidia fell. The profit is documented; the trader's identity, motive and information are not.

Watch in 55 seconds

Key points

  • A net outlay of about $8mn on 12 October is now worth around $196mn, but nothing public shows who made the trade or why.
  • The structure profits from a fall to between $100 and $150, so it points to a view on size as well as direction.
  • The trade matters because Nvidia faces a smuggling probe, export-licence changes and questions over its leadership, all at once.

The Financial Times reported on Friday that an unidentified trader opened a position in Nvidia options on 12 October that has since gained about $196mn on a net outlay of roughly $8mn. The trader bought 180,000 puts struck at $150 and sold 280,000 puts struck at $100, when the shares were at $226. They now trade near $166, down more than a quarter in a month, which puts the position close to its most profitable zone.

A profit that proves little on its own

The size of the gain invites an obvious conclusion, but the facts do not support it yet. Nobody knows who the trader is. It is also unclear whether this was a single speculative bet, one leg of a larger book, or a position already sold. Time decay costs the holder about $880,000 a day, so closing early to bank the gain would be rational, and public data would not necessarily show it.

A large paper profit is also not evidence of foreknowledge. Short positions in a falling stock make money whether or not the person holding them knew anything unusual. What separates luck, analysis and information is what the trader knew on 12 October and how they came to know it. Only an investigation with access to account records could establish that.

What the structure suggests

The construction is more revealing than the size of the gain. By selling more $100 puts than the $150 puts it bought, the trader cut the upfront cost to about $8mn but accepted a risk. At expiry on 15 January the position makes money if Nvidia finishes between $114.60 and $150.46, peaks at about $892mn near $100, and loses money if the shares fall well below $114.60.

Someone certain of a collapse would not normally cap the upside in this way. The shape of the trade expresses a view that Nvidia would fall substantially but stop short of a rout. That is consistent with a trader holding specific information about a limited event. It is equally consistent with a fund that sold downside insurance it considered safe and happened to own the cheaper protection above it. We cannot tell which from the structure alone.

The bearish thesis was also not secret. In May, Culper Research, a short seller that disclosed its own position, published a 40-page report on Nvidia's China exposure through Southeast Asian intermediaries. Those were allegations, not established findings. They show, though, that a negative case existed in public months before the trade. What a public thesis cannot explain is why the trade was placed on 12 October. The link between that date and what became public afterwards is the point any inquiry would test, and the available reporting does not settle it.

Why it matters beyond one position

The trade has drawn attention because of what has followed. Global Eye 24 has reported a smuggling investigation, scrutiny of Nvidia's terms for allocating GPUs, and unconfirmed rumours about Jensen Huang's position, alongside a sell-off in AI-linked stocks and Asian markets. Accounts of Huang's status differ, and it should be treated as unresolved. Meanwhile, The Wall Street Journal reported that Commerce is preparing licence requirements for advanced chips shipped to Malaysia, Singapore and Indonesia, the same region at the centre of the intermediary allegations.

Against that background, a single account earning 24 times its premium becomes a test of market integrity as well as a curiosity. If the trade was informed, the question is where the information came from, given that several of these developments depended on officials and company insiders. If it was not, it is a reminder of how much of the gain came from a position sized and shaped before the news. No regulator has been reported as examining the trade, and that absence is not evidence either way.

There is also a market-structure point. Options on a stock of Nvidia's weight let a modest outlay carry very large exposure. The net notional exposure of this position has grown from about $90.6mn at entry to nearly $1bn. Such leverage means a small number of participants can end up with outsized stakes in how a single story unfolds.

What to watch

The first marker is whether any supervisor or exchange confirms a review of Nvidia options activity around 12 October. Without that, the trade stays a market anecdote. The second is the Commerce licensing plan for Southeast Asia, since the details will determine how much of Nvidia's shipments are affected.

The administration is also expected within a week to name Nvidia, Amazon, Microsoft, OpenAI and Anthropic as national champions in AI. That would sit awkwardly beside a smuggling probe and open questions about Nvidia's leadership, and how the company is treated in the announcement will be telling.

Finally, there is the price itself. Nvidia was indicated slightly higher before Friday's open, at about $168. If it holds between $114.60 and $150.46 into mid-January, the position would stay profitable if held to expiry. A move back toward $226 would erase most of the gain. The trader's incentive is to sell well before then, which means the clearest evidence of who they are, if any appears, may come from the unwinding.

Analysis by the Global Eye 24 desk, with AI assistance. Spot an error? Write to corrections@globaleye24.com.