About Scopes
Measurement, not tips.
We're the scorekeeper for prediction markets: an independent scorecard that puts the crowd next to the smart money and tracks which one was right, to resolution.
Scopes puts every prediction-market price (Kalshi, Polymarket) next to an independent benchmark: the sharper, deeper markets and expert models professionals already trust: fed-funds futures, options chains, de-vigged sharp sportsbooks, and forecast models. Where the two disagree is the Scopes Divergence; which one proves right is the record. Think FedWatch, not a newsletter: a calculation with a permalink. We keep the scorecard; we never forecast, recommend, or pick a side.
Scopes is an independent measurement firm based in Hartford, Connecticut. The work is run by a small team drawing on corporate finance, capital markets, and risk-management experience; the methodology is public and the record is mechanical, so the numbers stand on their own rather than on any one person's judgment.
Scopes applies the discipline of investment performance measurement, the field codified in the GIPS standards, to prediction market prices.
How it works
For every tracked event we take the market price (Kalshi, Polymarket) and a Scopes Fair Value: an independent fair value derived without betting money: futures-implied probabilities, statistical nowcasts, realized-volatility models, a de-vigged sharp sportsbook, published forecast models, or a ratings model. The difference is the Scopes Divergence. When a liquid divergence crosses a threshold, the pipeline records it automatically and tracks it to resolution, so the record is mechanical, not editorial.
- Scopes Fair Value :
- an independent probability estimate computed from the desk’s benchmark source: futures-implied odds, options chains, de-vigged sharp sportsbooks, or forecast models.
- Scopes Divergence :
- the difference between the market price and the Scopes Fair Value.
The desks
One method, 8 desks: same product, different benchmark column.
vs options odds
vs NWS forecast
vs NOAA / CSU
- Rate Scope
- Fed-funds-futures–implied probabilities (FedWatch) + CPI/GDP nowcasts Live
- Poll Scope
- Two independent forecast models, shown side by side Live
- Sport Scope
- De-vigged sharp-book lines; an Elo model for two-team finals Live
- Energy Scope
- Options-implied probabilities from oil-ETF options (USO) on the CBOE chain Live
- Stock Scope
- Options-implied probabilities from the CBOE SPX chain (range markets) Live
- Crypto Scope
- BTC/ETH options-implied probabilities Coming
- Weather Scope
- NWS-forecast-implied probabilities (independent of the venue’s settlement source) Coming
- Catastrophe Scope
- Negative-binomial count model on NOAA seasonal outlooks + basin climatology Coming
The scorecard
Every flagged divergence is tracked to resolution and published on a public scorecard: which side, market or fair value, proved right, per desk. Both outcomes carry the same weight. The record is the product: it's how you learn whether an independent benchmark is better calibrated than the crowd, over many resolved flags rather than one.
What we don't do
Scopes does not give investment or betting advice, recommend trades, or predict outcomes. It measures whether the market is right, and keeps the receipts. Research and information only.
Who's behind it
The method is public and the record is mechanical, so the numbers do not depend on who built them. It still helps to know who that is.
Scopes was founded by Jon Shields, a CFA charterholder who has spent his career pricing and hedging real risk: as a divisional CFO at a public company, as CFO of private-equity-backed companies, and running hedging desks at The Hartford through the 2008 financial crisis. He holds an MBA in finance from Wharton, graduated from the U.S. Naval Academy in the top 10% of his class in computer science, and served in the submarine community. Scopes brings the engineering and the markets together: an independent benchmark beside every price, tracked to resolution in public.
For developers & agents
Read the data by API, or let an agent do it.
The same divergences, calibration record, and coverage the site shows are available programmatically. There's a versioned read API (/api/v1) described by an OpenAPI spec, and a Model Context Protocol (MCP) server so AI agents can discover and query Scopes directly: get_divergences, list_outcomes, get_series, get_calibration_summary, get_coverage. Agents can start from /llms.txt.
Free access is delayed and rate-limited; live data and the full archive follow your subscription. Create a key under API keys and send it as a bearer token. Research and information only; see the terms.
Scopes operates only at scopes.com.