Skip to content

PROBLAY • RESULTS CENTER

Results Center

Published, week-by-week records for PROBLAY's stored research populations. Every report names the exact population it measures, uses data frozen before kickoff, shows a deduplicated view beside the raw record, and lists the limitations that apply to that week.

Published reports

  • 2026 NFL Week 2

    Published

    2026 NFL Week 2 Research Results

    Stored Final Pregame-Labeled Model Qualified Over/Under Research

    The published Week 2 record for PROBLAY's stored Final Pregame-labeled Over/Under research, with the deduplicated view, the predicted-versus-observed difference, and the limitations that apply to the sample.

    Raw record
    Raw record: 41 wins, 14 losses, 0 pushes

    Every qualified Week 2 O/U proposition came from alternate-line markets, so the raw propositions are not independent. Two weeks is an early sample and nothing should be presented as an expected future rate.

    Read the full report →
  • 2026 NFL Week 1

    Published

    2026 NFL Week 1 Research Results

    Stored Final Pregame-Labeled Model Qualified Over/Under Research

    The published Week 1 record for PROBLAY's stored Final Pregame-labeled Over/Under research, with the deduplicated view, the predicted-versus-observed difference, and the limitations that apply to the sample. It is an early prospective audit of stored research, not proof of model quality.

    Raw record
    Raw record: 53 wins, 10 losses, 0 pushes

    Every qualified Week 1 O/U proposition came from alternate-line markets, so the raw propositions are not independent. One week is an early sample and nothing should be presented as an expected future rate. This is an early prospective audit of stored research, not proof of model quality.

    Read the full report →

How to read these reports

  • Each week is reported on its own. PROBLAY does not publish a combined, running or cumulative record across weeks.
  • Observed results describe a recorded sample of stored research. They are not an expected future rate.
  • Anytime Touchdown research is evaluated separately and is never blended into these Over/Under records.
  • Records are published unchanged, including a week where the model was less calibrated than the week before it.
  • Alternate-line markets mean the propositions inside one week are correlated rather than independent.

New to the terminology? Read Understanding the Numbers

Research and analysis only. PROBLAY is not a sportsbook and does not accept wagers.

We use Google Analytics to understand how visitors use PROBLAY and improve the experience. See our Privacy Policy.