Monte Carlo Simulation

Monte Carlo simulation is a technique that re-runs a strategy’s backtested trade results thousands of times in randomized orders or random samples, producing a distribution of possible outcomes instead of a single equity curve. A backtest shows one historical sequence of wins and losses; Monte Carlo asks what the same trades could have produced if luck had dealt them differently.

The two common flavours are trade-order shuffling (same trades, random sequence) and resampling with replacement (random draws, so a bad trade can repeat). Each simulated run rebuilds the equity curve and records metrics like final equity and max drawdown; across 5,000–10,000 runs those metrics form distributions.

Example

A NIFTY options strategy backtests to a −12% maximum drawdown over 400 trades. Shuffling those trades 10,000 times might show a median drawdown of −14%, a 95th percentile of −22%, and worst runs near −30%. A trader sizing a ₹5,00,000 account should plan around the −22% figure (about ₹1,10,000), not the backtest’s flattering −12% (₹60,000). If “ruin” is defined as −40% and 3% of runs breach it, that probability of ruin is the most decision-relevant number in the report. Hypothetical figures throughout.

Why it matters

Position sizing based on one backtest’s drawdown is sizing based on one shuffle of the deck. Monte Carlo turns “what happened” into “what could plausibly happen”, which is the question capital allocation actually depends on — especially for leveraged F&O strategies where an underestimated drawdown can breach margin.

In INDfolio AI, Monte Carlo analysis runs directly on your backtest’s trades — see Monte Carlo.