Skip to main content
Correlated risk occurs when positions that look separate respond to the same underlying factor. If they are likely to lose together, the account’s effective concentration can be much larger than any one ticket suggests.

Why per-trade limits are not enough

Consider three positions, each with a maximum planned loss of 1%:
  • long EUR/USD;
  • long GBP/USD;
  • long AUD/USD.
The platform shows three symbols and 3% total stop-based risk. Economically, all three include short U.S. dollar exposure. A broad dollar rally can push every position toward its stop at the same time. The total planned loss is still approximately 3% before costs, but the probability of losses occurring together is higher than it would be for independent trades. That clustering matters for daily loss limits, drawdown, and liquidity.

Sources of common exposure

Measuring correlated risk

Historical correlation is a useful starting point, not a complete answer. It depends on the sample period and can change during stress. A practical framework combines:
  1. Stop-based aggregation: total potential loss if every current stop is reached.
  2. Factor grouping: positions classified by common currency, sector, macro driver, direction, and strategy.
  3. Scenario testing: estimated loss under a shared shock, gap, or volatility increase.
  4. Concentration limits: maximum risk per symbol, factor, direction, and trade idea.
  5. Portfolio methods: covariance, beta, value-at-risk, or expected shortfall when reliable data and expertise are available.
No model removes uncertainty. A conservative rule is useful when precise dependence is difficult to estimate.

Example control

Assume the account allows:
  • 0.5% risk per position;
  • 1.0% risk per trade idea;
  • 1.5% total open risk;
  • a lower cap before a major scheduled release.
Two positions already use the full 1.0% dollar-weakness budget. A third related signal should be reduced, substituted, or rejected even if its stand-alone 0.5% limit is available.

Stress behavior

Correlations can rise when investors de-risk and liquidity deteriorates. Stops may also execute beyond their planned prices. The risk review should therefore ask both:
  • “How have these positions moved together?”
  • “What event could make them lose together?”

Risk per trade idea

Convert shared exposure into an operating limit.

One-side betting

Detect repeated directional bets across symbols.

Hidden-risk case study

Follow a simplified multi-position example.

EmoGuardian

Review account-level exposure controls.
Correlation estimates are backward-looking and assumption-dependent. They should inform risk limits, not be treated as guarantees.