Most investors believe markets reward speed. In biotech, this belief runs deep: FDA announcements drop, clinical trial results go public, and traders rush to react.
But speed is the wrong game.
Biotech is one of the cleanest laboratories we have for studying how markets actually process information. Outcomes are often binary — a drug works or it doesn’t — and stakes are existential. According to the Efficient Market Hypothesis, prices should reflect all available information. And yet, prices consistently move before the news.
This isn’t a behavioral quirk. It’s a structural pattern.
Quantitative research by Glen Carter shows why most headline-driven strategies fail. When thousands of SEC filings are analyzed in aggregate, trading on news produces no meaningful edge. Most filings are administrative. Most “signals” are noise.
The breakthrough comes from adding context.
When biotech companies are segmented by development stage, the picture changes. For pre-revenue, clinical-stage firms, trial-related filings are existential. For commercial-stage companies, they are often operational footnotes. Same news. Different meaning.
But the most revealing insight appears before announcements.
In the days leading up to major positive clinical news, clinical-stage stocks often show a steady pre-event run-up. Not explosive. Not obvious. But consistent. The market isn’t reacting — it’s anticipating.
This doesn’t require illegal insider trading. Institutional investors combine scientific expertise, expert networks, and deep due diligence. By the time news is public, conviction has already formed.
Platforms like Catalyst Ventures attempt to make this anticipatory layer visible. Instead of predicting outcomes, they detect footprints: quiet accumulation, anomalous volume, and patterns consistent with informed belief. Large language models help by filtering context — distinguishing signal from administrative noise at scale.
The lesson extends far beyond biotech.
Markets don’t price information. They price anticipated consequences. Speed matters less than structure. And without context, averages will always hide the signal.
The real edge isn’t in reading the news faster.
It’s in learning to read what happens before the news exists.
