Every tracked game (post-filter), regardless of whether the favorite flipped or drifted. Use this as the baseline. 7,044 games, 6,535 with known outcomes.
Implied probability vs actual win rate. Negative edge means the market overpriced the favorite (bet against). Positive edge means underestimated.
Did the line move toward the favorite, away from them, or stay stable? CLV is one of the strongest predictors of long-term ROI.
Games where the pre-game favorite (open < 2.0) saw their odds reach ≥ 2.0 at any point during live tracking. 3,013 flip events.
Games where the favorite drifted hugely but came back to win — the most "missed" units.
Games where the pre-game favorite's live odds drifted ≥ 1.0 from open after 10+ minutes of tracking. We track max odds reached, signal odds, and outcome. 3,438 signals.
What you can only see when flips and drift signals share the same pipeline.
Of drift signals at each magnitude, what fraction saw the favorite return below 2.0 by end?