How EquiReason Works: From Market Event to the Companies It Moves
Most market tools hand you a wall of disconnected scores. EquiReason is built around one idea: a market event and the companies it moves should stay connected, from the headline to the name. This is how that pipeline works.
The pipeline, and what it checks about itself
EquiReason measurement ·
A system that reads this much data is only as good as its willingness to catch its own errors. These are both sides of the current build — what it processes, and what it verifies before publishing:
- 2,484
- companies scored
- 12,804
- filings analysed
- 56
- live trends
- 13 / 13
- data invariants passing, 0 violations
The invariant checks are the part worth explaining. Each one asserts something that cannot be true of real data — a negative enterprise value, a net margin above 100 percent, a price/earnings ratio computed from negative earnings — and a value that fails is withheld rather than displayed. A separate freshness contract checks that every data source is current: 28 of 28 passing at the time of measurement. Both run continuously, not at release.
Derived from SEC filings via EquiReason's own processing pipeline. How this is measured.
The problem: scores without a chain
A number like "7.8/10" is only useful if you can see why. Without the chain of reasoning behind it — what event drove it, through which economic mechanism, to which company — a score is just noise you have to trust blindly. EquiReason's design goal is the opposite: every name is explained through the same decision language so nothing hides behind a single number.
Step 1 — Detect the event
Dozens of data jobs run continuously: market news, SEC filings (8-K, 10-K, 10-Q), insider transactions, 13F institutional flow, short interest, macro indicators and prices. Each is a different lens on the same market, and each can surface a signal the others miss.
Step 2 — Map the impact
A raw event isn't actionable on its own. The system maps it to an economic mechanism — the cause-and-effect path by which the event changes a company's revenue, costs or risk. That mechanism then points to the industry areas, and the companies within them, that are actually exposed.
Step 3 — Confirm the trend
A single headline is news, not a trend. The system checks whether an event is corroborated across sources and confirmed by the market's own behaviour before it is treated as a trend. Signals that are still building are tracked separately as forming trends.
Step 4 — Select and explain the company
For each confirmed trend, candidate companies are ranked by their actual exposure to the mechanism — not by cheap price momentum. The best name is then explained through a consistent set of dimensions:
- Trend support — is there a real, confirmed trend behind this name?
- Economic mechanism — the path from event to this company's fundamentals
- Financial-statement quality — is the underlying business sound?
- Evidence tier — how strong and how trusted is the supporting evidence?
- Macro gate — does the broader environment support or fight the thesis?
- Conviction tier — the overall strength of the case, with the market's move checked against it
Why this matters
Because the chain is explicit, you can audit any conclusion. You see the event, the mechanism, the evidence and the conviction — so you can agree, disagree, or dig deeper, instead of trusting a black box.
Frequently asked questions
What data does EquiReason use?
Market news, SEC filings (8-K, 10-K, 10-Q), insider transactions, 13F institutional flow, short interest, macro indicators and price data — all feeding one connected research workflow.
Is EquiReason investment advice?
No. EquiReason is a research and market-intelligence tool, not brokerage or investment advice. It surfaces and explains signals; investment decisions remain yours.