Trading Edge Stress Test | VSC Capital & Advisory

What this stress test answers

One question: if this trading edge is roughly real, how difficult could the journey still become? You supply a win rate, an average win and an average loss in R. The tool runs that edge 10,000 times as a Monte Carlo simulation over the horizon you choose, then reports the range of endpoints, the drawdowns and the losing streaks those runs contained.

R is one unit of risk — whatever you stake on a single trade. A winner worth twice its risk is +2R; a full stop-out is −1R. Working in R keeps every figure independent of account size.

Why the runs differ

Every run uses identical assumptions, and they still end up in different places for two separate reasons.

How many wins you get

Each run draws its own wins and losses, so the count varies from run to run. That sampling variance is what moves the endpoint. Over 200 trades at a 45% win rate, some runs land 80 winners and some land 100.

What order they arrive in

Two runs with the same number of wins can still feel completely different. Sequencing is what determines how deep the drawdowns get and how long the losing streaks run.

Trading expectancy per trade: E = (Win rate × Avg win) − (Loss rate × Avg loss)

How the numbers are produced

Simulated percentiles

The endpoint range, the drawdown and the losing streak are percentiles across all 10,000 runs, under the assumptions you entered. They are reported as the 5th, 50th and 95th percentile rather than the best and worst run, because a single extreme run is unstable — it gets more extreme simply by simulating more.

An exact probability

The model-implied chance of finishing below 0R is not estimated from the runs. Because every win and loss has a fixed size, the endpoint depends only on how many trades won, which follows a binomial distribution — so the probability has a closed form and is computed exactly. It is conditional on the entered constant win rate and fixed payoffs.

What the model leaves out

Each trade is an independent draw at a constant win rate, and every win is exactly your average win while every loss is exactly your average loss. The model does not represent variation in individual trade sizes, tail losses, gaps, slippage, costs, changing market regimes, or an edge that decays over time. Real results may differ materially from these figures, and may be more extreme in either direction.