Simulación de Montecarlo
A Monte Carlo simulation runs thousands of random sequences of a strategy's possible trades or returns to estimate the range of outcomes, such as likely drawdowns and the risk of ruin.
Explained
A backtest shows one path; a Monte Carlo simulation reshuffles or resamples the trades to show many possible paths with the same statistics. This reveals how deep drawdowns can get just from the order of wins and losses.
Results are given as percentiles: for example, a 95th percentile drawdown of 28% means 5% of simulated paths had deeper drawdowns.
Example
A backtest of 300 trades shows a max drawdown of 15%. 10,000 reshuffles of the same trades give a median max drawdown of 17% and a 95th percentile of 29%.
Planning for 29% rather than 15% leads to a smaller, safer position size.
Common mistakes
Feeding the simulation optimistic backtest trades.
Ignoring that trades may not be independent (streaks, regimes).
Reading the median as the worst case.
Questions traders ask
Why use Monte Carlo instead of a backtest?
A backtest is a single sequence; Monte Carlo shows the distribution of results the same edge could produce.
How many simulations are enough?
Several thousand runs usually give stable percentiles.