A method that plays out the rest of a season thousands of times at random to estimate the probability of each outcome, such as winning the title or being relegated.
A Monte Carlo simulation runs the same uncertain process a large number of times — for football, playing out every remaining fixture at random according to each team’s strength. After ten thousand simulated seasons, the share of them in which a team finishes first, top four or last becomes its probability of that outcome.
It is the standard way to turn a match-by-match model into league-table probabilities, because the maths of who finishes where is far too complex to solve exactly once several teams and many games are involved.
Monte Carlo results are estimates with a small random error: run the same simulation twice and the percentages shift by a point or so. The more simulations you run, the more stable they become.
A five-point lead with four games left is not certain because the trailing team can still win out while the leader drops points — a simulation quantifies exactly how likely that swing is.
Poisson Distribution
The statistical model used to predict football match scorelines by treating goal-scoring as a random process based on each team's expected goals rate.
xG (Expected Goals)
A metric that scores every shot by its probability of resulting in a goal, based on factors like shot location, angle, and assist type.
xPTS (Expected Points)
The number of league points a team would be expected to earn based on their xG and xGA across a series of matches.
For informational and educational purposes only. Disclaimer
