How the Prop Simulator Works
The prop simulator uses Monte Carlo simulation — running 10,000 random samples from a statistical distribution based on your projection and variance inputs. By counting how many simulated games exceed the prop line, it calculates the true over probability from your model, independent of what the book prices it at.
Why Monte Carlo for Props?
Simple expected-value math assumes a normal distribution. Player props often aren't normally distributed — passing yards skew right (big games happen, very low games less so), touchdowns follow Poisson distributions, and variance compounds across position and matchup type. Monte Carlo simulation handles all these distribution shapes accurately.
NFL Prop Betting: Passing Yards
For passing yards, use the log-normal distribution (right-skewed). Set your projection as the median game expectation based on matchup data, weather, pace, and injury reports. Set variance at approximately 20-25% of projection. A QB projecting 275 yards might have standard deviation of 60-70. The simulator then tells you the true probability of hitting the book's line.
NBA Prop Betting: Points and Rebounds
NBA props are among the most exploitable markets because books set lines based on season averages, while sharp bettors adjust for rest, pace of opponent, lineup changes, and minutes projections. Use normal distribution for most NBA props. For three-pointers made, use Poisson (it's a counting distribution for rare discrete events).
Reading the Results
Compare your simulated over probability against the book's implied probability. If your simulation gives Over 58% and the book prices it at -110 (implied 52.4%), that's a +5.6% edge — positive expected value. If the book prices it at -145 (implied 59.2%) and your simulation gives 58%, pass — no value.