Portfolios of Strategies and Capital Allocation
One strategy is a bet on a single edge. A portfolio is a bet on that diversification being real.
One strategy is a bet on one edge holding up. A portfolio of strategies is a bet on diversification being real — which, as with any diversification claim, is something you have to verify rather than assume.
Combining strategies is genuinely powerful when done for the right reason and genuinely pointless when done for the wrong one.
Diversify the source of edge
The useful kind of diversification is different reasons to make money. A trend-following system on indices, a mean-reversion system on currency pairs, and a carry-biased system are drawing on different market behaviours.
The useless kind is five variants of the same idea with slightly different parameters. They will be highly correlated, they will fail together, and the portfolio will have all the complexity of five systems with none of the benefit.
| Strategy | Risk Profile | Illustrative Target | Core Logic | Min Capital |
|---|---|---|---|---|
Grid Hero Forex | Low Volatility | 3-5% / mo | Grid Oscillation | $5,000+ |
Alpha Trade Gold(XAUUSD) | Medium Volatility | 8-12% / mo | Breakout Trading | $2,000+ |
M1 Sniper Gold(XAUUSD) | High Volatility | 15%+ / mo | Pullback Trading | $1,000+ |
RangeQuantum AUDNOK | Medium Volatility | 5-8% / mo | Pullback Trading | $2,000+ |
These target ranges are illustrative examples, not forecasts or guarantees — no backtest produced them. CFDs are leveraged products: 68-89% of retail investor accounts lose money trading them.
Correlation between strategies
Measure correlation at the level of daily or weekly returns between strategy equity curves, not at the level of instruments. Two strategies on the same instrument can be uncorrelated if their logic differs; two on different instruments can be highly correlated if their logic is the same.
Combined equity curves that draw down at the same time are not a portfolio. They are one strategy with extra steps, and the correlation matrix of returns will tell you so before the drawdown does.
Forex Correlation Matrix
Asset Relationships • D1
| EUR/USD | GBP/USD | USD/JPY | USD/CHF | AUD/USD | USD/CAD | NZD/USD | EUR/GBP | EUR/JPY | GBP/JPY | AUD/JPY | CAD/JPY | CHF/JPY | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| EUR/USD | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| GBP/USD | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| USD/JPY | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| USD/CHF | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| AUD/USD | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| USD/CAD | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| NZD/USD | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| EUR/GBP | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| EUR/JPY | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| GBP/JPY | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| AUD/JPY | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| CAD/JPY | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| CHF/JPY | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
Allocate by risk, not by past profit
Allocating capital to whichever strategy performed best recently is performance chasing, and it reliably buys high and sells low. It concentrates capital in whatever is most likely to mean-revert.
A more robust approach allocates by risk contribution, so each strategy contributes a comparable share of portfolio volatility. Smaller allocations to newer or less-proven strategies is a reasonable refinement until their live record matches their test.
When to retire a strategy
Every strategy will underperform at some point. The difficult problem is distinguishing normal variance from genuine decay, and there is no perfect answer.
A workable rule uses evidence rather than feeling: compare live results against the walk-forward expectation using a statistically meaningful number of trades, and act when performance falls outside what that distribution allows. One bad month is not evidence. Three hundred trades below expectation is.
Key takeaways
- Diversify the source of edge, not the parameters of one idea
- Measure correlation between strategy equity curves, not between instruments
- Allocate by risk contribution rather than by recent performance
- Retire a strategy on statistical evidence, never after one bad month
- Combine strategies across different instruments and timeframes
- Correlation between strategies matters more than individual quality
- Allocate capital by risk contribution, not by past profit
- Retire a strategy on evidence, not after one bad month
Running five variants of the same idea and believing you have five strategies.
Reading is not verification
Take the concept you just read and turn it into explicit rules, then test it. That is the only way to know whether it actually works.
Build a strategy