Sean Westropp from Deep Sail Capital is back for another Skull Sessions chat. He’s been actively posting on Twitter and on his new Substack, Short and Squeeze Corner, about his short selling strategies.
Given my past history of writing research exposing pump & dumps and China Frauds at Geoinvesting, I couldn’t wait for this one.
We dug into where Sean is seeing opportunity right now, what he’s avoiding, and what he’s learned the hard way running a long-short book for years.
Pixel Research and Macro Mind, Micro Market co-hosted alongside me. Both are definitely great stock pickers to follow. If you’d ever like to cohost a skull session with us, just let us know in the comments. Thank you.
Sean runs roughly 80–90% net long, so shorting is a tool he uses carefully, not a constant exposure. A lot of this conversation centered on discipline: the mechanics that make shorting fundamentally different from being long: borrow costs, timing risk, market cap and liquidity constraints, and how easily a short that looked like a winner on paper can actually have lost money once you account for everything.
We also worked through his read on the current bubble environment. He has clear views on quantum, where the first leg has already deflated and he’s recently re-opened positions on IonQ (IQNT) and Rigetti (RGTI). He also walked us through how he’s thinking about the AI supply chain: the optical, semiconductor, and memory names that supply the hyperscalers… where real revenue growth makes timing any unwind extremely difficult right now. The parallels to the 2022 EV bubble came up several times. One idea kept resurfacing through the whole conversation: shorting is almost entirely a timing game, and identifying overvalued companies is only step one.
Toward the end we touched briefly on the Andrew Left’s criminal conviction and what it might mean for both the short selling environment and long-side investors who write publicly about their positions. Right after my chat with Sean, we recorded an Expert Insights session with securities attorney John Sutter, where we cover that case in much more depth. Get alerted on the availability of this conversation by subscribing, for free, to this Substack;
Sean also got into the secondary benefit of spending so much time studying the short side: occasionally a low-quality company sets up for an asymmetric squeeze that’s playable from the long side. Sean writes about both halves of that on his Substack. We talked through how he thinks about those setups, what makes some workable, and why most of them don’t fit his criteria.
Again, a special thanks to Lukas Milosic from Pixel Research and JEV from Macro Mind, Micro Market for joining the conversation and pushing on several of these topics.
Skull Sessions is a collaboration with Geoinvesting.com, a full-stack microcap research platform and MS Microcaps LLC , home of the Microcap Quality Index (MSMqi)
Key Takeaways
Shorting is almost entirely a timing game. Identifying overvalued companies is only step one. The hard part is finding the exact right moment when the market turns on them.
Borrow costs can quietly destroy short returns. Many “winning” shorts actually lose money once you do the math, which is why Sean caps annualized borrow at 30% before he’ll take a position.
The AI supply chain has real fundamental growth, which makes the timing of any short hard to call. Most names in the cycle won’t be shortable until at least 2027 or later, when hyperscaler spending growth eventually reverses.
The quantum bubble has deflated once and is starting to inflate again. The 2022 EV bubble is the closest parallel… it trailed out over multiple years rather than crashing in weeks.
Spending time studying shorts occasionally surfaces asymmetric long opportunities through squeezes, but they’re rare (one every couple of months) and involve low-quality companies, not high-conviction longs.
The best squeeze candidates combine extended drawdowns (~90% over three years) with public name recognition. No-name microcaps rarely move enough to play.
Activist short reports vary wildly in quality. Hindenburg-style research that engages directly with companies is closer to a real edge; many others are smash-and-grab. Even well-substantiated reports sometimes don’t move the stock at all.
Stocks Discussed
IonQ (NYSE: IONQ): Quantum computing name Sean has recently re-opened a short position on, following the bubble’s initial deflation and a partial rally back.
Rigetti Computing (Nasdaq: RGTI): Another quantum name in Sean’s current short basket, treated as part of the same industry bubble setup.
Avis Budget Group (Nasdaq: CAR): Cited as an example of a recent short squeeze in a low-quality, low-float setup.
Beyond Meat (Nasdaq: BYND): Discussed as another squeeze-prone name fitting the “down significantly with public recognition” profile.
Virgin Galactic (NYSE: SPCE): Squeeze setup Sean wrote about and traded, though noted he was a day or two late on the move.
Groupon (Nasdaq: GRPN): Used as the canonical example of a battered, well-recognized name that can run dramatically on small narrative shifts.
Energous (Nasdaq: WATT): Example of a name with stubborn short interest that eventually pivoted into a real operating business.
New Era Energy (Nasdaq: NUAI): Discussed as an example of a “fake” AI company, built around repurposing a small oil and gas footprint into an AI data center narrative.
