A company files an 8-K.
Minutes later, a market tied to the same event starts moving.
Looking at either dataset alone tells only part of the story.
SEC filings show what a company officially disclosed. Prediction market data can show how market participants are pricing an uncertain outcome. Put the two together, and researchers can build an event-driven data workflow that compares new information with changing market expectations.
For hedge funds, alternative data teams, and AI research systems, this creates a different research question:
What was disclosed, and what changed in the market after people learned about it?
Two Data Layers, Two Different Signals
SEC filings and prediction markets should not be treated as substitutes.
They answer different questions.
| Data Source | Research question | What you can analyze |
| SEC filings | What did the company officially disclose? | Filing metadata, text, specific sections, financial statements, filing timing |
| Prediction markets | What are participants pricing around an uncertain outcome? | Prices, trades, quotes, volume, OHLCV, spreads, liquidity, order books |
FinFeedAPI provides structured access to both.
The SEC Data API provides filing metadata, full-text search, original EDGAR documents, filing extraction, XBRL conversion, and real-time filing notifications.
The Prediction Markets API provides normalized market data across multiple prediction market venues, including market discovery, trades, quotes, OHLCV, and order book data.
The interesting research begins when these datasets are aligned around the same event.
Start With the SEC Event
An event-driven workflow needs a reliable starting timestamp.
For SEC filings, simply knowing the filing date may not be precise enough.
FinFeedAPI filing metadata can include AcceptanceDateTime, which records when the filing was accepted by the SEC.
For event-driven analysis, this provides a more precise anchor for examining what happened around the disclosure.
Researchers can then retrieve the relevant filing and determine what information actually became available.
Depending on the task, the SEC API can be used to:
- query filings by company, CIK, form, or date
- search filing text for specific topics
- download the original EDGAR document
- extract sections from filings such as 10-Ks, 10-Qs, and 8-Ks
- convert XBRL financial data into structured JSON
The accession number can then connect filing discovery with deeper processing such as extraction, document download, or XBRL conversion.
Find the Prediction Market Connected to the Event
The next step is not simply to pull every prediction market.
The research system needs to identify markets that are genuinely related to the disclosure.
Possible matching signals include:
- company name or ticker
- product or event keywords
- regulatory events
- macroeconomic events
- policy topics
- sector-specific developments
For example, an SEC disclosure could contain information relevant to a regulatory approval, acquisition, corporate transaction, or another event that also has an active prediction market.
FinFeedAPI provides the market metadata and normalized prediction market data needed for this research.
The mapping itself belongs to the research workflow.
This distinction matters: FinFeedAPI provides both datasets, but researchers decide which filing and prediction market represent the same underlying event.
Measure What Happened Around the Filing
Once a relevant market is identified, the SEC AcceptanceDateTime can become the center of an event window.
A researcher might compare prediction market behavior:
1 hour before vs. 1 hour after
1 day before vs. 1 day after
7 days before vs. 7 days after
This makes it possible to examine whether expectations began changing before the filing, repriced immediately afterward, or adjusted gradually as participants interpreted the new information.
Historical prediction market OHLCV can show the broader price path.
Trades and quotes add more detail.
Order book snapshots add another important layer: liquidity.
A Price Move Is Not Enough
Suppose a prediction market moves from 45 to 60 around an SEC filing.
That looks significant.
But the price change alone does not tell you how strong the signal was.
- Was there meaningful trading volume?
- Did spreads widen?
- Was there enough liquidity behind the new price?
- Did actual trades confirm the move, or was the market temporarily thin?
Prediction market trades, quotes, OHLCV, and order books can help answer these questions.
That makes market depth especially useful when working with alternative financial data. A large move in a liquid market can carry different information than the same move in a market with very little activity.
Prediction market prices should therefore be treated as market-implied expectations rather than guaranteed probabilities.
Liquidity, fees, market structure, participant constraints, and venue-specific mechanics can all affect the observed price.
From Filing to Research Signal
A combined workflow can look like this:
1. Detect the filing
Use the SEC API to identify a new 8-K, 10-Q, 10-K, S-1, or another relevant filing.
2. Determine what changed
Retrieve the filing, extract relevant sections, search the text, or convert XBRL data into structured financial information.
3. Identify related prediction markets
Match the company, event, topic, or regulatory development against relevant prediction market metadata.
4. Build an event window
Use AcceptanceDateTime as the event anchor and retrieve prediction market data before and after the filing.
5. Measure the reaction
Compare prices, trades, volume, spreads, quotes, and order book depth.
6. Add the result to the research model
The combined dataset can then become an input for quantitative research, event studies, alerts, or AI-driven analysis.
The result is not simply another data feed.
It is a structured way to compare new information with changing expectations.
Where This Research Stack Becomes Useful
Event-driven investment research
Analysts can study whether particular types of corporate disclosures consistently coincide with changes in related market expectations.
Over time, that can create datasets for studying how quickly information is incorporated and which disclosures appear to matter most.
Alternative data research
Prediction markets provide a different signal from conventional prices, analyst estimates, or news sentiment.
Combining them with official SEC disclosures creates an alternative financial data layer grounded in both primary-source documents and observed market activity.
AI research agents
AI agents can use the SEC API through MCP to discover filings, extract relevant sections, and work with structured financial information.
Prediction Markets API access through MCP can provide a second source of context around unresolved events.
An agent could therefore investigate not only:
"What did the company disclose?"
but also:
"Did market-implied expectations change around that disclosure?"
That produces a much richer research workflow than summarizing the filing alone.
Event-Driven Research Needs Context, Not Just More Data
SEC filings provide authoritative corporate disclosures.
Prediction markets provide a separate view into how participants are pricing uncertain outcomes.
Neither replaces the other.
Together, they allow researchers to study the relationship between information and expectation.
With FinFeedAPI, teams can access SEC filing data through REST, WebSocket, JSON-RPC, and MCP, while Prediction Markets API data is available through REST, JSON-RPC, and MCP.
That creates the building blocks for an event-driven research stack spanning filing discovery, document extraction, market matching, historical analysis, liquidity checks, and AI workflows.
Build an Event-Driven Research Stack with FinFeedAPI
Connect structured SEC data with normalized prediction market data to research how disclosures and market expectations evolve around the same events.
Explore the FinFeedAPI SEC API and Prediction Markets API to build event studies, alternative data pipelines, and AI research workflows.
