Dan Loeb: The Lost Art of Short Selling, and Why Stock Picking is Back

Key points

• **The headline call: the short side is live again, and it's a picker's market.** *'The lost art of short selling has come back and it's absolutely critical. Doesn't matter what you do, you have to be really selective. People talk about stock pickers market. This is a bond and credit pickers market.'* **This is the bear-side complement to the rest of the cohort's bull tape** — where [Laffont frames a $4T AI IPO wave](/issues/2026-06-07) and [Friar defends $100B+ compute spend](/issues/2026-06-07), Loeb is saying the dispersion that lets you short single names has returned for the first time in years. The actionable read: alpha is back in selection, not index beta.
• **Short discipline is NOT a valuation short — the Reddit-squeeze lesson.** *'One thing that we've avoided is kind of evaluation, solely valuation based approach. I've just seen too many people get run over by shorts that have dumb valuations, but they get captured on Reddit or one of these other things.'* He name-checks *'some of these space companies right now that there's no rhyme or reason.'* **The craft note worth keeping: a high multiple is a reason to be careful, not a thesis** — the meme-squeeze mechanism punishes pure valuation shorts. Pairs with his own warning below that crowded megacap shorts (Nvidia) are dangerous for the same structural reason.
• **The live short: home builders, on a structural-impairment thesis.** *'We had a really strong view on home builders... the home building industry was first structurally impaired because of the way that they were all pretending to be NVR, which is all pretending to be asset light, but they had massive commitments to these land pools which in things that they said were options but they were really very committed.'* Layered on a post-Covid *'inventory disruptions and pricing that really made no sense,'* with buyers squeezed by costs and financing. **A rare named, dated short thesis** — and a CRE-relevant one: the tell is balance-sheet reality (committed land capital) hiding behind asset-light optics.
• **The contrarian long: Nvidia is undervalued, and the short case is mechanical not fundamental.** Asked directly if Nvidia is undervalued: *'Yeah, absolutely. On earnings over the next two or three years.'* Why it screens as a crowded short: *'And the long short pods are structured such that they have to be short something. So Nvidia feels like a safe short. By the way, Google was a safe short. Amazon was a safe short.'* **This is the load-bearing trade** — Loeb is calling the most-shorted megacap a long and explaining the crowding as a structural artefact of pod-shop construction. (Transcript renders the size as '$5 billion'; context — *'multi trillion dollar companies', 'the largest entity that's ever existed'* — makes clear he means trillions; an AssemblyAI error, so not quoted as a figure.) Directly contra [Ackman](/issues/2026-06-07) and the [$4T-IPO bubble framing](/issues/2026-06-07).
• **Distribution is the symmetric agony — two named, quantified misses.** *'We were, we were private investors in Palantir and I think we sold all our stock in the 20s. Huge mistake.'* And on Enphase: *'we sold some stock on the IPO and then took a tax hit and I think sold it under a dollar. And the stock, I think had we stayed on would have made $4 billion.'* **The honest counterweight to the short thesis: getting *out* is as hard as getting in** — *'markets are brutal.'* The lesson he draws is structural, not analytical: *'we learned not to go on boards anymore because it restricts your ability to be liquid.'* Useful for anyone holding concentrated private/IPO positions into this week's [IPO wave](/issues/2026-06-07).
• **Where the human edge actually survives: trust and read, not analysis.** *'the human element, I think this is true for everyone. You have here the element of the social component, the human network of knowing people, being able to capture opportunities, work with people, interact. Like that's never going away.'* And the sharper bit: *'And there's a thing that I think the agents of the AI will never really be able to look in your eye and assess all the things that you've expanded.'* **A narrower claim than the usual 'humans-still-matter' line** — Loeb concedes the analytical layer to machines and locates the durable edge in relationship capital and management assessment. The natural pairing with [Satya Nadella's full-stack-builder thesis](/issues/2026-06-07) and the cohort's commoditisation debate: as analysis commoditises, the residual is the social layer.
• **The quality screen now turns on adaptable management — assessed by pattern recognition, not a rubric.** On moats: *'we can't really just look at a product or a technology and say, oh, this is going to be it forever. So we really look for a management team that we think will be adaptable.'* He explicitly disclaims a system: asked if assessing management is quantifiable, *'No, it's still very subjective, qualitative... after 30 years there's like a pattern recognition.'* Frames it via **Chamath's 'time bounded value'** — *'what are the companies that are going to be around 7 to 10 to 20?'* **The honest caveat is self-aware:** *'I think we deluded ourselves earlier... if you ask people about the moat around IBM or AOL or Yahoo, you'd say the same thing.'* Moats are harder to underwrite than the cohort's defensibility talk admits.
• **The DNA: event-driven + credit, reverse-engineered from the best.** Loeb traces Third Point to event-driven investing — *'Takeovers, spinoffs, risk or arbitrage, bankruptcies, privatizations, demutualizations'* — built by studying mentors at Jefferies and Goldman: *'I was like a Chinese corporation that was copying and reverse engineering and taking everything in and creating my database of knowledge and my own operating system.'* The platform today spans the hedge fund (long/short credit + equity), a CLO business, private credit (*'direct lending and workouts, which is very important'*), and a wholly-then-half-owned insurance vehicle capturing the investment-grade book. **The structural point: the firm is now a credit-and-event platform, which is *why* he can credibly call this a 'credit pickers market'** — it's where his actual edge sits, not a throwaway line.
• **Persona arc: from the 'original troll' to institutional activism — and the through-line is the same.** Loeb says of himself *'I was the ot, the original troll'* — on Silicon Investor chat boards *'long before Reddit,'* uncovering 90s frauds (Act Trade, a factoring company at *'some large multiple of book value'* dressed in fake 'TAD' technology). On modern activism: *'Activism without proxy contests is like Catholicism without hell'* — landing the same weekend his Nestle campaign was in the news. **The continuity matters for the short thesis: the edge is forensic skepticism of narrative-vs-reality**, whether the target is a 90s factoring fraud or a 2026 asset-light homebuilder.
• **The macro frame is unchanged from last week — quote Jesse Livermore, not the Fed.** *'I always quote this Jesse Livermore line. There's nothing new under the sun.'* The mechanics he hunts haven't changed in 30 years: management incentivised to *'sandbag their numbers'* into option grants and excess supply, then beat. **Cross-ref: this is the craft layer beneath [last week's ILB macro frame](/issues/2026-05-31)** — there Loeb collapsed the world to oil + AI; here he shows the bottom-up mechanism (incentive-driven mispricing) that he's actually trading. Same investor, two altitudes.

Notable quotes

Themes

• Return of single-name short selling
• Valuation-vs-narrative short discipline
• Nvidia as crowded-short contrarian long
• Distribution / selling-discipline problem
• Human edge as relationship capital