Methodology

Overview

This page outlines how we evaluate the accuracy of mock drafts and beat writer content. We currently offer a detailed explanation of our methodology for Analyst Mock Draft grading. Explanations for Beat Writer Reliability Ratings and Team-Specific Mocks are coming soon. Each area uses a rigorous, data-driven framework to ensure that evaluations are as fair and consistent as possible.

Beat Writer Reliability Ratings

This feature is coming soon. We're evaluating beat writers based on the accuracy and reliability of their tweets and articles. We evaluate all of their content with a focus on actionable areas e.g. game-day inactive reporting, roster projections, etc.

Analyst Mock Drafts

All analyst mock drafts are evaluated based on their performance in the first round only, even if the mock includes multiple rounds. This ensures that comparisons are always apples-to-apples across mocks.

Each mock draft is graded using four primary metrics:

  • Exact Matches: Credit for correctly predicting both the pick number and the team that selects a specific player.
  • Player-Team Matches: Credit when a mock draft predicts the correct team for a player, regardless of slot.
  • Captured Capital: Proprietary. Measures how well a mock draft captures the order in which players were actually drafted.
  • Positional Accuracy: Proprietary. Measures how well a mock draft approximates the NFL's distribution of first-round capital across positions (e.g. QB, WR, EDGE).

The final score is a blend of these four metrics.

Beat Writer Mock Drafts

Team-specific mock draft scoring is coming soon. These will evaluate how well each mock predicts a single team's selections.

FAQ

The following questions apply specifically to our evaluation of mock drafts (both league-wide and team-specific).

Do you account for hybrid-position players?

Yes. To ensure our Positional Accuracy metric is as precise as possible, we carefully allocate positional weights for hybrid players. This applies to both the actual draft and mocks. For example, a player like Travis Hunter may be split between CB and WR depending on context.

For actual drafts, we consider how the teams announce their selections and any relevant quotes about their selected players' expected roles that were made around the time of the draft.

For mock drafts, we consider how the author labels each player's position and what the blurb says about each player's expected role.

Why is your mock draft grading system better than other grading systems?

Our grading system was carefully designed to measure what really matters in a mock draft. Other grading systems that use arbitrary point systems or other similar approaches often fail to capture the magnitude of errors. For example, other grading systems may treat a small miss like Keon Coleman (pick 33) being mocked 32nd overall the same as a larger miss like Adonai Mitchell (pick 52) being mocked 32nd overall. Our metrics measure errors precisely and differentiate between small misses and big misses.