How biotech hedge funds learn whether a clinical trial will fail before the scientists do, from prediction models trained on public data to expert calls billed at over $1,000 an hour, and the largest insider trading case in American history.

Biotech hedge funds are a strange business: who can know in advance whether a drug will work in humans? Yet the three largest each manage over $15 billion.
Part of the answer is surprisingly mundane: one platform predicts Phase II outcomes from simple inputs like target choice and trial design, and hit 79 percent accuracy across 14 resolved trials. The more expensive part is talking to experts at $500 to $1,000 an hour.
In 2008 the portfolio manager Mathew Martoma learned from an 81-year-old professor that an Alzheimer's drug was failing, liquidated a position worth roughly $700 million and went short, avoiding losses and making $275 million. He was convicted, but the subtler versions of the same conversation seem never to have stopped.


Curated from high-quality sources, with concise summaries and key takeaways.