V6

V6 Monte Carlo Performance

10,000 game simulations combining V4 Bayesian + V5 ML predictions with full probability distributions

Compare to V3Compare to V4Compare to V5

Overall Performance

Performance by Stat Type

StatTotal PropsWith ActualsAccuracyStatus

Model Comparison

How does V6 compare to other model versions?

Stat TypePropsCurrentComparisonAdvantageWinner

Edge Calibration (Backtesting)

When V6 predicts X+ points over/under the line, what % actually hit?

Edge ThresholdTotal BetsCorrectAccuracyOver BetsOver Hit %Under BetsUnder Hit %Calibrated?

Betting Simulation

Simulate P&L if you bet on every V6 recommendation above a threshold.

Probability Calibration

When V6 predicts X% probability of over, what % actually hit?

Probability RangeExpectedActualDiffSampleQuality

Line Movement Correlation

How does V6 perform when the market agrees vs disagrees?

Pace × Edge Performance Matrix

Win rates by game pace and edge threshold. Fast (>102), Moderate (98-102), Slow (<98)

Pace≥1.0≥1.5≥2.0≥2.5≥3.0≥4.0≥5.0

Accuracy Over Time

Weekly V6 prop betting accuracy with cumulative running total

WeekBetsCorrectWeekly %Cumulative %Avg EdgeTrend

Player-Level Analysis

Filter by stat type to see which players V6 predicts best in each category

Granular Prop Data

V6 Model Details

How V6 Works

  • Ensemble: Combines V4 Bayesian (45%) + V5 ML (55%) predictions
  • Monte Carlo: 10,000 simulations per game accounting for variance
  • Correlations: Player stats within same game are correlated
  • Distributions: Full probability distributions, not just point estimates
  • Game Scenarios: Generates 4 likely game outcome scenarios

Key Metrics

V4 Weight:45%
V5 Weight:55%
Simulations:10,000 per game
Scenarios:4 per game
Stat Types:PTS, REB, AST, FG3M, STL, BLK + combos