Concept · example: ETH, week of 31 August 2020

Risk limit calibration

How should a trader's risk limits be sized when calm weeks and fast weeks behave so differently?

In plain words

Risk limit calibration is the process of choosing position sizes, stop distances and loss caps from a reference measure of past market behaviour, often recent volatility or typical weekly range.

Why it matters

Limits built on calm weeks assume the market keeps moving gently. When volatility expands sharply, the same sizing rule allows positions the market can no longer carry, and losses arrive faster than the limits were designed to absorb.

In a real week: ETH, week of 31 August 2020

ETH's four weeks before 31 August 2020 returned 4.75%, 11.14%, -9.89% and 9.78%, but the week beginning 31 August 2020 returned -17.75% with a 58.5% range and 140.2% realised volatility. Volume reached 3.38 times the prior 12-week average. Limits sized on the preceding weeks would have allowed exposure that the week's swing could overwhelm.

A common mistake

Assuming the most recent calm stretch is a stable baseline. Calibration windows that exclude fast weeks systematically understate how far prices can travel in a short period.

Related ideas

realised volatilityposition sizingdrawdownvalue at risk

Research notes on real market weeks, not investment advice. Sources: Binance spot BTCUSDT and ETHUSDT 1-hour candles and USD-M perpetual funding rates; weekly figures computed by WSTiger.

RISCON Labsrisconlabs.com