FinFeedAPI Blog - Building an Earnings-Event Monitor with SEC Filings, Stock OHLCV, and Prediction Market Odds

Every earnings announcement creates an information shock.

A company publishes its quarterly or annual results. Investors compare those numbers with expectations. The stock reprices. Analysts revise forecasts. Prediction markets adjust their implied probabilities as uncertainty disappears.

Most monitoring systems only capture one part of that process.

Some watch SEC filings and trigger an alert when a new 10-Q or 8-K appears. Others focus on stock charts and notify users when the price moves. A growing number monitor prediction markets to understand how expectations evolve.

Each approach provides useful information—but none tells the complete story.

A filing explains what the company disclosed.

A stock chart shows how investors reacted.

A prediction market reveals what participants believed before the announcement and how those beliefs changed afterward.

Together, these three datasets provide a much richer view of an earnings event than any single feed can offer.

The Three Layers of an Earnings Event

Think of every earnings announcement as three connected timelines.

The first is the official disclosure.

The second is the market reaction.

The third is the expectation curve.

Each answers a different question.

Data Layers	Question It Answers
SEC Filings	What did the company officially disclose?
Stock OHLCV	How did investors react?
Prediction Markets	What did the market expect before the announcement?

The real insight appears when those timelines are aligned.

Suppose a company reports earnings that exceed analyst estimates.

The filing itself doesn't tell you whether the result surprised investors.

The stock chart tells you the price increased by 8%.

Still, you don't know whether that move was unexpected.

Now add prediction market odds.

If the probability of a positive earnings surprise had already climbed from 48% to 82% during the previous week, much of the optimism was already priced in.

The same earnings report now tells a very different story.

That's why a modern earnings monitor should combine all three layers instead of treating them independently.

Step 1: Detect the Event with SEC Filings

Every monitor starts with the official disclosure.

For U.S. public companies, that means the SEC's EDGAR system.

EDGAR contains annual reports, quarterly reports, current reports, exhibits, XBRL filings, and many other regulatory documents.

It's the authoritative source of company disclosures.

Unfortunately, it wasn't designed as a developer-friendly data platform.

Parsing raw filings, locating specific sections, extracting filing metadata, or converting XBRL into usable application data quickly becomes a project of its own.

FinFeedAPI's SEC API removes much of that complexity by providing structured access to SEC filings, filing metadata, source documents, clean text extraction, full-text search, and XBRL converted into JSON.

For an earnings-event monitor, the workflow is straightforward.

Watch the companies you care about.

Filter filings by relevant form types such as 10-Q, 10-K, and 8-K.

When a new filing appears, capture information including:

filing timestamp,
company identifier,
form type,
reporting period,
accession number,
filing metadata,
extracted text,
XBRL data where available.

These become the anchor point for the remainder of the event timeline.

Understanding the Different Filing Types

Although many SEC forms can influence markets, three are particularly useful for earnings monitoring.

10-Q

The quarterly report.

It provides an updated view of financial performance, management commentary, and operational developments.

Because companies publish it every quarter, it's often the filing that officially documents recently reported earnings.

10-K

The annual report.

Unlike the 10-Q, the 10-K is audited and considerably more comprehensive.

It includes financial statements, business descriptions, risk factors, legal proceedings, and detailed management discussion.

For long-term analysis, it's often the richest disclosure available.

8-K

Current reports.

Unlike scheduled quarterly or annual filings, 8-Ks communicate significant corporate events as they happen.

They may include earnings-related materials, exhibits, investor presentations, or other announcements outside the regular reporting cycle.

A well-designed earnings monitor should track all three because each contributes different pieces of the overall event.

Step 2: Measure the Stock's Reaction

Once the filing has been detected, the next question becomes obvious.

How did the market respond?

This is where stock OHLCV data becomes valuable.

OHLCV condenses raw market activity into time-based candles containing:

Open
High
Low
Close
Volume

Instead of processing every individual trade, developers can quickly measure how prices evolved before and after the disclosure.

Around earnings announcements, several metrics are particularly useful:

opening gap after the filing,
intraday trading range,
abnormal trading volume,
multi-session returns,
volatility expansion.

One of the simplest measurements compares the final closing price before the announcement with the first opening price afterward.

A large overnight gap often indicates that the market rapidly repriced new information.

Volume provides additional context.

A sharp price movement accompanied by unusually high trading volume tells a different story than the same movement occurring on relatively light participation.

FinFeedAPI's Stock API provides T+1 historical equity market data across more than 40 global exchanges, including OHLCV time series, trades, Level 1 quotes, Level 2 depth, Level 3 order book data, administrative messages, and exchange system events.

Step 3: Add the Missing Layer—Expectations

Even after combining SEC filings and stock prices, one important question remains unanswered.

Was the outcome actually surprising?

Financial markets don't react to good or bad news.

They react to news relative to expectations.

This is where prediction markets become extremely valuable.

Unlike stock prices, prediction markets measure belief rather than valuation.

In many binary markets, the "Yes" price can be interpreted as the market's implied probability.

A contract trading at 0.72 suggests participants are pricing roughly a 72% probability of the event occurring.

Watching those probabilities before an earnings announcement reveals how expectations evolved leading up to the official disclosure.

Did confidence steadily increase over several days?

Did traders become more cautious just hours before earnings?

Did probability collapse immediately after the filing?

Those movements often explain the stock's reaction far better than the filing alone.

FinFeedAPI's Prediction Markets API provides normalized access to historical and current prediction market data from Polymarket, Kalshi, Myriad, and Manifold, including market metadata, OHLCV history, trades, quotes, and order book snapshots.

Bringing the Three Timelines Together

The real challenge isn't collecting three datasets.

It's aligning them around the same event.

Each source has its own timestamps, identifiers, and update frequency:

SEC filings are published when the company submits its disclosure.
Stock OHLCV summarizes trading activity into time-based intervals.
Prediction markets update continuously as participants buy and sell contracts.

Viewed independently, each dataset answers only part of the story.

Viewed together, they create a complete earnings timeline.

Instead of asking a single question "Did the stock move?"… the monitor now answers several:

What was officially disclosed?
What did investors expect beforehand?
How quickly did expectations change?
How large was the market reaction?
Did price, volume, and probability move together?

That context is often more valuable than the earnings announcement itself.

Example Earnings Monitor Architecture

An earnings monitor doesn't need to be a complex system.

Conceptually, it consists of a few connected components.

Notice that the hardest part isn't downloading data.

It's synchronizing three different information sources around the same event.

Once that timeline exists, every new earnings announcement becomes another structured dataset that can be analyzed automatically.

Measuring the Event

One advantage of combining multiple datasets is that you can build metrics that describe the event from several perspectives.

SEC Filing Metrics

The filing layer provides structured information about the disclosure itself.

Useful metrics include:

Filing detected
Form type
Filing timestamp
Reporting period
Company identifier (CIK)
Filing accession number
XBRL availability
Earnings-related filing sections

Stock Reaction Metrics

The stock layer measures how the market responded.

Typical calculations include:

Pre-event close
Post-event open
Opening gap
Event window return
Intraday range
Volume ratio
Rolling volatility

These metrics separate a brief reaction from a sustained repricing.

Prediction Market Metrics

Prediction markets measure changing expectations.

Useful signals include:

Pre-event implied probability
Post-event implied probability
Probability change
Trading volume
Market activity
Market status
Price volatility
Available liquidity

One important reminder is that probability should always be interpreted alongside liquidity.

Two markets may both imply a 70% probability, but the one supported by deeper trading activity is generally a stronger signal than a thin, inactive market.

Turning Data into an Event Score

Once all three layers have been aligned, it's possible to create higher-level event metrics.

For example, an earnings-event monitor might combine:

filing relevance,
opening price gap,
abnormal trading volume,
prediction market probability change,
prediction market participation,

into a single earnings surprise score.

The exact calculation depends on your application.

Some research teams may prioritize price movement.

Others may emphasize probability changes.

Still others may use machine learning models trained on historical earnings events.

The important point is that these signals become available only after the datasets have been joined into one timeline.

Beyond Earnings Alerts

An earnings monitor doesn't have to stop at sending notifications.

Once historical event timelines have been collected, they become a valuable research dataset.

Developers can use them to:

compare market expectations with actual outcomes,
study how quickly prices absorb new information,
identify delayed market reactions,
analyze abnormal trading volume,
evaluate earnings-related volatility,
build event-driven trading strategies,
generate features for machine learning models,
create dashboards for investors, analysts, or newsrooms.

The same architecture can also be extended beyond earnings announcements to monitor mergers, regulatory filings, management changes, or other material corporate events.

One Financial Data Stack Instead of Four

Traditionally, building an earnings-event monitor meant integrating several unrelated vendors.

One for SEC filings.

Another for stock prices.

Another for prediction markets.

Each with different identifiers, formats, authentication methods, and update schedules.

FinFeedAPI brings those layers together in a single developer-focused ecosystem.

Developers can access:

structured SEC filings,
historical stock market data,
prediction market data,
and FX data,

without stitching together multiple disconnected APIs.

That doesn't eliminate the analytical work—but it significantly reduces the engineering work required before analysis can even begin.

Build Smarter Earnings Monitors

An earnings announcement is more than a regulatory filing.

It's an information event.

The filing explains what the company disclosed.

The stock market shows how investors reacted.

Prediction markets reveal how expectations evolved before and after the announcement.

Viewed separately, each tells part of the story.

Combined, they create a complete event timeline that supports richer dashboards, better alerts, deeper research, and more informed decision-making.

Whether you're building an internal research platform, an investor dashboard, an automated alerting system, or an event-driven analytics engine, combining these three data layers provides a far more complete understanding of earnings events than any single source alone.

Start Building with FinFeedAPI

Building an earnings-event monitor doesn't require stitching together multiple data vendors anymore.

With structured SEC filings, historical stock market data, and normalized prediction market data available through FinFeedAPI, you can focus on analyzing events instead of integrating data sources.

Explore the APIs, experiment with historical events, and start building your own earnings analytics pipeline.

Building an Earnings-Event Monitor with SEC Filings, Stock OHLCV, and Prediction Market Odds