This tutorial replaces the Portfolio Aggregator tool that used to live on this site. The same analysis now runs in JSK TradeProof, inside your own spreadsheet.
Even a robust strategy has losing streaks. The classic answer is to run several strategies together, so that one’s bad month is another’s good month.
It works, but only when the strategies really are different. Two strategies that lose at the same time don’t diversify anything. They just double the size of the same bet.
This article covers how to combine strategies and how to check that the combination actually helps.
Two ways to combine strategies
1. Overlapping: run everything. Every trade from every strategy is taken, even when two strategies are in the market at the same time. Each strategy keeps its own capital, so the portfolio’s capital is the sum of theirs.
This is how most traders run several systems in separate accounts or with separate allocations.
2. Non-overlapping: one position at a time. Strategies are ranked by priority. A trade is taken only if no other strategy’s trade is open at that moment; otherwise it’s skipped. One pool of capital is shared.
This suits a single account where you can’t, or don’t want to, hold several positions at once. The order of priority matters: the top strategy gets every trade it wants, and lower ones fill the gaps.
The numbers that tell you whether it helped
Don’t judge a portfolio by its combined profit. Profit simply adds up. Risk doesn’t, and that’s the whole point.
Combined drawdown
Compare the portfolio’s maximum drawdown with each strategy’s own. If the strategies offset each other, the combined drawdown is smaller than you’d expect from adding them together. If it’s roughly the sum, the strategies are losing at the same times.
Correlation
A correlation matrix shows how closely each pair of strategies’ results move together, from +1 (always together) to −1 (always opposite). Low or negative correlation is what you want. Two strategies on the same market with similar logic will often be highly correlated, however different their rules look.
Diversification ratio
The diversification ratio compares the volatility you’d expect from the strategies on their own with the volatility of the combined account. Above 1 means combining them reduced risk; the higher, the more it helped. Close to 1 means you’ve gained very little.
Months when everything lost
The most practical check of all: in how many months did every strategy lose at the same time? Those are the months that hurt. A portfolio with many of them isn’t diversified, whatever the correlation matrix says.
Common mistakes
- Combining near-copies. The same idea on two timeframes, or two indicators that measure the same thing, usually move together.
- Judging on a short history. Correlations change. Strategies that look independent in a calm market often lose together in a crash.
- Ignoring capital. In an overlapping portfolio, check that you could actually fund every position that was open at the same time.
- Optimising the mix. Picking weights that made the past look best is just overfitting at the portfolio level.
How to test it in a spreadsheet
JSK TradeProof runs this analysis in Excel or Google Sheets:
- Load each strategy’s trades. Put each strategy on its own sheet and name it, or keep everything in one range and split it by symbol or by a tag column.
- Choose the portfolio analysis. TradeProof measures the strategies as one account.
- Read the results: the combined equity curve and drawdown, the correlation matrix, the diversification ratio, and the count of months in which every strategy lost at once.
A free account covers portfolios of up to 3 strategies; Pro has no limit.
Go deeper
Combining strategies is Part IV of my book Seven Tests Before You Go Live, alongside testing a strategy across market phases. Before you combine anything, make sure each strategy passes the robustness tests on its own.