Why a Second Strategy Can Reduce Your Drawdown
Posted: Thu Oct 29, 2026 8:27 am
Why a Second Strategy Can Reduce Your Drawdown
Strategy A's monthly results in R over six months: plus 3, minus 2, plus 4, minus 3, plus 2, plus 1. Strategy B's: minus 1, plus 2, minus 1, plus 3, minus 1, plus 1.
Totals. A: 3 minus 2 plus 4 minus 3 plus 2 plus 1 is 5R. B: minus 1 plus 2 minus 1 plus 3 minus 1 plus 1 is 3R.
Running totals for A: 3, 1, 5, 2, 4, 5. The largest drop from a peak is from 5 to 2, a 3R drawdown.
Running totals for B: minus 1, 1, 0, 3, 2, 3. Each dip is small: from the start to minus 1, from 1 to 0, and from 3 to 2. The worst drawdown is 1R.
Combine them month by month: 2, 0, 3, 0, 1, 2. Total 8R. Running totals: 2, 2, 5, 5, 6, 8. No drawdown at all.
Why? A's bad months, the second and fourth, coincided with B's good months. The two strategies are negatively correlated in this sample, so combining them smooths results.
Caution. Six months is a tiny sample, and the offsetting pattern may not persist. Correlation changes with market conditions.
Costs and complexity rise when running two systems. Risk should be allocated so that combined exposure stays within limits.
Still, the exercise shows the mechanism: different strategies lose at different times.
Try this: line up the monthly results of two approaches you use and look for offsetting patterns.
Strategy A's monthly results in R over six months: plus 3, minus 2, plus 4, minus 3, plus 2, plus 1. Strategy B's: minus 1, plus 2, minus 1, plus 3, minus 1, plus 1.
Totals. A: 3 minus 2 plus 4 minus 3 plus 2 plus 1 is 5R. B: minus 1 plus 2 minus 1 plus 3 minus 1 plus 1 is 3R.
Running totals for A: 3, 1, 5, 2, 4, 5. The largest drop from a peak is from 5 to 2, a 3R drawdown.
Running totals for B: minus 1, 1, 0, 3, 2, 3. Each dip is small: from the start to minus 1, from 1 to 0, and from 3 to 2. The worst drawdown is 1R.
Combine them month by month: 2, 0, 3, 0, 1, 2. Total 8R. Running totals: 2, 2, 5, 5, 6, 8. No drawdown at all.
Why? A's bad months, the second and fourth, coincided with B's good months. The two strategies are negatively correlated in this sample, so combining them smooths results.
Caution. Six months is a tiny sample, and the offsetting pattern may not persist. Correlation changes with market conditions.
Costs and complexity rise when running two systems. Risk should be allocated so that combined exposure stays within limits.
Still, the exercise shows the mechanism: different strategies lose at different times.
Try this: line up the monthly results of two approaches you use and look for offsetting patterns.