MLB Monte Carlo Sim

MLB Monte Carlo game simulations

This tool simulates every MLB game on the slate thousands of times and reports the full distribution of outcomes rather than a single projected score. For each game you get a win probability for both sides, an over and under probability against the posted total, a run line cover probability, and the gap between each of those and the price the sportsbook is offering. It is free with an account.

The inputs are the ones that actually move a run total: the starting pitchers and how deep each is likely to go, the batting order and its platoon split against that starter, bullpen state, park factor, first pitch weather, the umpire behind the plate and the opposing catcher's framing. Those interact, which is the reason to simulate rather than add them up.

Why simulate instead of projecting a score

A single projected score throws away the thing you actually need. Baseball outcomes are a distribution, and two games with the same projected total of 8.5 runs can have very different chances of going over, depending on how the runs are likely to be distributed between the teams and the innings.

Simulation keeps the shape. Running a game many times produces the full range of outcomes, which is what lets you price a run line properly. Winning by exactly one run is a common baseball result, so the difference between a moneyline and a run line is a question about distribution rather than about who is better.

It also handles the interactions that a spreadsheet cannot. Bullpen availability, park, weather and lineup order all affect each other, and the only practical way to see the combined effect on a run distribution is to play the game out repeatedly.

Reading an edge without overreacting to it

The number worth acting on is the difference between the simulated probability and the price implied by the market. A small edge is inside the model's own error bar and is not a bet. A large edge usually means the model is missing something the market knows, most often a lineup or an injury, so check before assuming you have found something.

The market is a strong baseline. Treat a simulation as a second opinion that identifies where to look more closely, rather than as a number that overrides a price set by everyone betting into it.

What the History tab is for

Every pick the model publishes is written down at the price it was made and graded against the real final score. The History tab shows the running record split by bet type, a cumulative profit and loss line on a flat one unit stake, and a breakdown by how big the edge was when the pick was made. Break even at standard -110 juice is 52.4 percent, so that is the line any record has to clear before it means anything.

Frequently asked questions