What Is Risk of Ruin?

Risk of ruinis the estimated probability that an account’s equity falls to a critical threshold at some point within a given horizon — say, the next 200 trades. The threshold does not have to be zero. For most traders, “ruin” is the level at which trading effectively stops: a 30% or 50% drawdown, a margin floor, or simply the point where the required recovery gain becomes unrealistic.

Two things make the concept useful. First, it looks at the pathof equity, not just the destination — a strategy can be profitable on average and still pass through a fatal drawdown on the way. Second, it converts a vague worry (“what if I hit a bad run?”) into a number that responds to the one input you fully control: how much you risk per trade.

Every risk-of-ruin model, from a closed-form formula to a full simulation, is built from the same small set of inputs:

The first two inputs describe the edge; the last three describe exposure to bad sequences. Realistic values come from a meaningful sample of your own closed trades, not from a guess — which is one reason reviewing your own MetaTrader history per strategy matters before trusting any calculator output.

The horizon deserves a closer look, because it only works in one direction. Ruin is an absorbing event: once the threshold is hit, the run is over, so the probability can only accumulate as more trades are added. An estimate of 5% over 200 trades is always higher over 1,000 trades with the same inputs — surviving a quarter is not the same claim as surviving a career.

The engine inside every ruin estimate is the arithmetic of consecutive losses. If each trade loses independently with probability q, the chance of n losses in a row is q multiplied by itself n times:

With a 45% win rate, q is 0.55, so five straight losses have probability 0.55⁵ ≈ 5%. That sounds comfortably rare — but it is the chance for one specific window of five trades. Across a 200-trade sample there are many overlapping windows, and in simulation the chance of at least one five-loss run somewhere in those 200 trades comes out near 99%. Eight in a row (0.55⁸ ≈ 0.8% per window) remains a realistic event over a trading year.

Losses compound geometrically: each one is taken from an already-reduced balance, and the damage accelerates as the per-trade fraction grows. The same eight-loss streak that a conservatively sized account absorbs as a routine drawdown can push an aggressively sized account through its ruin threshold — with the win rate, payoff ratio and strategy all unchanged.

This is why risk of ruin reacts so violently to the sizing input. Doubling risk per trade does not double the ruin probability; it can multiply it many times over, because it shortens the streak needed to reach the threshold while streak probabilities fall off exponentially with length.

The table below comes from a simple illustrative simulation: 45% win rate, 1.5 payoff ratio (a positive edge of +0.125R per trade), fixed-fractional sizing, a 200-trade horizon, and ruin defined as equity dropping 30% below the starting balance. Only the risk per trade changes between rows.

The shape is the lesson: the edge is identical in every row, yet the estimate moves from effectively impossible to roughly one chance in three. At 0.5% risk, reaching the threshold takes the equivalent of about 71 net losing trades — so deep that the positive expectancy pulls equity away from the edge of the cliff faster than ordinary variance can push it over. At 5%, seven net losers are enough, which is well inside the range of a normal bad fortnight. You can reproduce this sensitivity with your own inputs in the free Monte Carlo Trading Simulator, which resamples a trade distribution thousands of times and counts how often the threshold is hit.

A risk-of-ruin number is conditional on everything you fed into it. The standard models quietly assume:

• a stable win rate and payoff ratio— but real strategies drift as market conditions change, and a measured 45% can become 38% without notice;
• independent trades— while correlated positions, news events and one-sided market regimes make losses cluster more than coin-flip math suggests;
• clean fills at the modelled risk— slippage, gaps and widened spreads can make an individual loss larger than the planned fraction.

None of this makes the estimate useless — it makes it a sensitivity tool. Comparing the output at 1% versus 2% risk with your own measured inputs is informative even when the absolute number is uncertain. Treating the output as a guarantee of safety is the misuse.

The three concepts form one chain. Your ruin threshold is usually expressed as a maximum acceptable drawdown; your risk per tradeis the position-sizing decision that determines how quickly a normal losing streak approaches that drawdown; and the risk-of-ruin estimate quantifies the link between the two over a horizon. How the sizing decision is actually made — from stop distance and pip value — is covered in the position sizing guide.

In practice, the workflow runs backwards: decide the drawdown you are genuinely willing to sit through, measure your real win rate and payoff ratio from your own MetaTrader trade history, and then solve for the risk per trade that keeps the ruin estimate acceptably small. The free Risk of Ruin Calculator does that arithmetic for any combination of inputs — the quality of the answer depends entirely on how honestly the inputs were measured.