How we think about backtests

Backtests are not predictions. They're tools for understanding how a strategy would have behaved historically—and what risks it might carry into the future.

The metrics

What we measure

CAGR

Compound Annual Growth Rate. The annualized return that would produce the final value from the starting value.

Maximum Drawdown

The largest peak-to-trough decline. Shows the worst loss you'd have experienced before recovery.

Volatility

Standard deviation of returns. Measures how much the portfolio value fluctuates over time.

Sharpe Ratio

Risk-adjusted return. Higher values mean better return per unit of risk taken.

Worst Year

The calendar year with the poorest performance. A gut-check for psychological tolerance.

Recovery Time

How long it took to recover from the worst drawdown. Tests patience and time horizon alignment.

The purpose

Why backtesting matters

🔍

Stress-test the strategy

See how rules performed during 2008, 2020, and other crisis periods. Would you have stayed the course?

📊

Quantify the risks

Abstract fears become concrete numbers. A -35% drawdown means something specific.

⚖️

Compare alternatives

How does DCA + DEEP compare to pure DCA? To lump-sum? Numbers answer what intuition can't.

🎯

Calibrate expectations

An 8% CAGR with 15% volatility sets different expectations than 12% with 25% volatility.

Illustrative example

Sample backtest output

This is a hypothetical example for illustration purposes only.

DCA + DEEP Strategy (2010–2023) Example Only

DCA + DEEP

Pure DCA (baseline)

CAGR +9.2%

Max Drawdown -22.4%

Volatility 13.8%

Sharpe Ratio 0.67

Worst Year -14.1%

A note on methodology

Backtests in Zyra are calculated using historical price data, documented FX rates, and realistic assumptions about trading costs. We include fund expense ratios and typical brokerage fees in our calculations.

However, backtests have inherent limitations. They assume perfect execution at historical prices, which may not reflect real-world slippage. They also can't account for future market conditions that may differ from historical patterns.

We present backtests as tools for understanding—not as predictions or promises. The goal is to build confidence in the methodology, not to guarantee outcomes.

Backtests are not guarantees. They are tools for understanding risk, stress-testing strategy rules, and calibrating expectations. Past performance does not predict future results.