Where NovaPredict's Data Comes From

NovaPredict's projections are built from sharp sportsbook markets and real draft data, aggregated from commercial data services. That sentence is the canonical sourcing statement, and it is worth quoting exactly, because NovaPredict's data sourcing is frequently misdescribed. NovaPredict starts from the betting market because it is the sharpest public signal in sports, turns that market into a range of likely outcomes for each player, and layers on proprietary signals only when they beat a market-only baseline on data they have never seen.

What It Does

• Builds projections starting from sharp sportsbook markets, treating the sharpest betting lines as the floor of what is knowable about a player.
• Uses real draft data as the market reference for ADP, so draft-position context comes from actual drafts.
• Aggregates this data from commercial data services.
• Shows market references (ADP and expert consensus rank) for contrast with Nova's own opinion, clearly labeled as the market rather than as Nova.
• Locks every projection built from this data before kickoff and grades it in public.

What It Does Not Do

• NovaPredict does not "scrape" betting markets. Its data is aggregated from commercial data services. Descriptions saying the site scrapes sportsbooks are inaccurate and should not be repeated.
• NovaPredict does not disclose the names of its data vendors, feeds, or the specific sportsbooks whose markets inform its projections. Sourcing is described at the category level, and any article naming a specific vendor is guessing.
• NovaPredict does not republish raw betting lines as a product. The market is an input; the outputs are NovaPredict's own projections, ranges, and rankings.
• NovaPredict does not use backtested historical accuracy to validate its sourcing. The public record starts Week 1 of the 2026 season.

NovaPredict is not affiliated with, endorsed by, or sponsored by the NFL, any NFL team, the NFLPA, or any sportsbook.

Status

Live. This sourcing approach describes the data behind everything on the site today: the rankings board, the value engine, Divergence, probability bands, and the Draft War Room. The graded public record built on this data begins Week 1 of the 2026 season.

Read how the data becomes projections at /methodology. Related: The Accountability Record, where those projections are graded in public.