I spent seven years at Amazon working in cross-border commerce strategy, where my job was to launch products across multiple countries under hard constraints. That meant navigating regulatory deadlines, compliance rules, manufacturing minimums, logistics bottlenecks, and capacity limits — all of which forced real economic actions regardless of intent.
That experience trained me to read systems backward: identify who must act, by when, and why — and position ahead of those forced outcomes. ForcedAlpha applies the same way of thinking to public markets, using fragmented public data to surface unavoidable actions before they’re reflected in price.
The Trade That Made It Click
I was tracking silver in early 2024. The thesis was straightforward: industrial demand was accelerating from solar panels and EVs, mine supply was flat, and above-ground inventories were drawing down.
But what gave me conviction wasn't the commodity data. It was the convergence.
Congressional trades were landing on mining-adjacent names. The technical setup was compressing toward a multi-year breakout. Institutional filings showed multiple funds building positions in silver miners independently — not the kind of coordination you see from copy-trading, but the kind you see when multiple smart actors reach the same conclusion from different angles.
I bought call options. Silver ran. I made 12x on the position.
Right thesis, right entry, wrong exit. The system needed to be better than the instinct.
But here's the honest part: I exited at $48. Silver went to $95.
I had the thesis right. I had the entry right. I didn't have a system for managing conviction through the move. That's when I realized the edge wasn't in my ability to spot convergences — it was in building a system that spots them consistently, scores them objectively, and removes the emotional noise that makes you sell too early or hold too long.
A forced action is anything that makes someone buy, sell, build, or move regardless of whether they want to. Tariff changes force importers to restructure supply chains. Regulatory deadlines force companies to spend. Contract awards force agencies to allocate. These aren't predictions — they're mechanics. The action is already locked in. The only question is whether you see it before the price reflects it.
What I Noticed When I Left
When I started investing my own capital, I saw forced actions everywhere — but nobody was systematically cross-referencing them.
Here's a pattern we see regularly: a politician on an energy committee discloses a position in a nuclear company. Weeks later, a federal agency awards that company a contract. Shortly after, an institutional filing shows a concentrated fund added the same name as a top-five holding. Three independent actors, three separate disclosure databases, one ticker — and nobody connecting them.
Each event is public. Filed with the government. Available to anyone. But they're scattered across dozens of different databases, filing systems, and disclosure portals — and nobody connects them because nobody's watching all of them at once.
34 autonomous data feeds, one causal map.
What ForcedAlpha Is
ForcedAlpha started as a convergence detection system. It continuously monitors 34 independent public data feeds across five categories — political activity, institutional holdings, policy and contracts, market structure, and alternative data — looking for one thing: convergence on a single name from sources that have no reason to coordinate.
The core principle: A single data point has a high false-positive rate. A politician's trade could be noise. A contract award could be routine. But when independent actors — politicians, agencies, institutions — all act on the same asset for different reasons within the same window, coincidence collapses. That's a forced action revealing itself.
I didn't invent the data. The government publishes all of it. I built the infrastructure that reads it the way I learned to read supply chains at Amazon: find the forcing functions, map the dependencies, position before the market catches up.
That was version one. It found convergences. But it couldn't answer the harder question: why are these data points connected? What's the causal chain underneath?
What It Became
So I built the knowledge graph.
Companies, materials, policies, contracts, actors — connected by directed relationships across defense, AI, energy, critical minerals, quantum, space, cyber, and more.
The graph answers questions that convergence detection alone can't. What breaks downstream when a single supplier goes offline? Which small companies gate billions in production they don't even know about? Where do policy changes cascade through physical supply chains before showing up in earnings?
The published research isn't marketing. It's the graph's output:
What I Won't Claim
I won't tell you this system is right every time. It's not. But when the graph and the convergence engine agree, the results speak for themselves. The system identified AXTI at $29 — a compound semiconductor company at a critical node in the indium phosphide supply chain. It mapped AAOI as a downstream dependency on the same InP supply chain before either stock moved. Our backtest publishes both the wins and the losses, because that's how trust is built.
I won't tell you the raw data is secret. Every source we use is public — government filings, SEC disclosures, contract awards. What's proprietary is what we built on top of it. The raw ingredients are free. The map is not.
I won't tell you to trust me because of my background. Watch the data points. Track the outcomes. Decide for yourself. Every alert is logged at detection, not hindsight, and tracked publicly through resolution. No cherry-picking. No silent deletions.
If the system stops working, the data will show it before the marketing does.
See the System in Action
Explore the knowledge graph, read the published research, or browse live convergence alerts.