Drawdowns: why duration matters more than depth
intermediate
Automated material · TradeAlmanac editorial deskPrepared by a language model from our stored data and checked by an editor.
Drawdown is the fall from a previous peak to a trough. Maximum drawdown answers the question of what would have had to be endured, and it is the most honest risk measure among the simple ones.
The arithmetic of recovery
It is asymmetric, and that is the main thing to know about it. Recovering from a fall of a third requires a rise of a half. The deeper the hole, the disproportionately harder the climb out.
But depth is not what breaks people
Time does. Five years underwater destroy discipline more reliably than one sharp crash: a crash is lived through as an event, while years of going nowhere feel like a verdict. It is in the fourth year that people sell at the bottom.
What to do about it
Look at the drawdown before entering a position rather than after. The question "could I withstand this" is asked in advance — later it is answered by something other than reason.
Diversification reduces a drawdown only while assets do not fall together. In a crisis correlations converge on one, and a portfolio of thirty securities from a single market falls like a single security.
Related instruments
Prepared by a language model from our stored data and checked by an editor.
How we use language modelsSimilar articles
- Hidden concentration: when twenty securities are one betA portfolio can look varied and depend on a single factor. That is established by calculation rather than by looking.
- Portfolio liquidity risk: how long an exit would takeWhat has to be assessed is not only a position's return but the time it takes to close without losses.
- Correlation: why diversification sometimes stops workingThe link between assets is not constant and rises precisely in a crisis — that is, when it was being counted on.
- Beta: how closely a security repeats the marketA coefficient of sensitivity to the market. Useful for understanding the structure of risk and useless as a forecast.