How does keeping cash and rebalancing change a gold allocation?
Gold is represented by SPDR Gold Shares (GLD), not spot bullion. Fund expenses are reflected in its market price. Sessions follow America/New_York. Cash is held in USD with no interest.
$10,000 starting capital · Jan 1, 2021 – Sep 30, 2026 · Daily observations
Invested weight band: 68.0–92.0%. Corrections aim back at 80% invested.
Fixed fee: 0.2% per order, charged on initial buys and every correction.
Change a slider to recalculate using the same historical data.
Includes daily values, effective target weights and skipped-order reasons. 0 orders skipped.
Maximum drawdown measures the largest daily-close fall from a previous portfolio peak.
Buy & hold $20,887
Buy & hold 108.87%
Buy & hold 13.67%
Buy & hold 26.40%
Buy & hold $19.96
Buy & hold 1
With 20% target cash, a 15% relative allocation threshold, and 0.20% fees, rebalancing finished $2,174 behind buy-and-hold. Its maximum drawdown was 2.22 percentage points smaller.
The comparison includes both the effect of holding cash and the effect of rebalancing. It does not isolate a rebalancing premium. A lower drawdown and a higher return are different outcomes.
Calendar-year returns after modeled fees. Partial years cover only the available window.
| Year | Buy & hold | Rebalancing | Difference |
|---|---|---|---|
| 2021 | -6.24% | -4.99% | 1.25 pp |
| 2022 | -0.77% | -0.61% | 0.16 pp |
| 2023 | 12.69% | 10.01% | -2.68 pp |
| 2024 | 26.66% | 21.53% | -5.12 pp |
| 2025 | 63.68% | 53.60% | -10.07 pp |
| 2026 partial | -3.90% | -3.50% | 0.40 pp |
Signals use the previous daily close. Fills use the next available trading session’s open. The initial purchase is included.
| Session date | Order | GLD quantity | Price | Order value | Fee | Cash after | Invested after |
|---|---|---|---|---|---|---|---|
| Jan 4, 2021 | buy initial | 43.893061 | $181.97 | $7,987.22 | $15.97 | $1,996.81 | 80.00% |
This gold backtest uses SPDR Gold Shares (GLD), an exchange-traded gold proxy, rather than spot bullion or futures. GLD market prices reflect its expenses and can differ from the price of physical gold. Compare the impact of holding cash and restoring a target allocation.
These two strategies have different exposure: buy-and-hold starts fully invested; rebalancing starts with your selected cash reserve. Any performance difference includes both allocation and trading effects. Neither strategy is promised to win.
Both portfolios start with $10,000 at the first day's open. Buy-and-hold invests its full budget; rebalancing uses your selected fixed allocation or the dynamic rule’s starting target. Initial holdings are sized to their targets after the fixed 0.2% fee, within the $10,000 budget. Dynamic cash uses a fixed $10,000 reference, increasing linearly from 5% cash at −30% return to 25% at +30% by default. The rule determines targets before invested-asset drift is checked.
At each session’s close, use the Rebalio app’s evaluator to compare invested weight with target × (1 ± relative threshold). At an 80% target and 10% threshold, sell above 88% invested or buy below 72%; equality does not trigger. Cash never triggers an order. Size the full correction back to the target at the closing valuation, then fill that frozen quantity at the next session’s open.
Buys must be fully funded, including fees; unfunded orders are skipped rather than resized. Overnight price changes and fees can leave the resulting weight away from target. A final-day signal has no fill inside the window.
Equity is remaining cash plus units × daily closing price. There is no leverage, cash yield, added capital, withdrawal, or forced sale at the end. Maximum drawdown uses daily closes; intraday losses can be larger.
Daily OHLCV prices are downloaded from Yahoo Finance using yahoo-finance2. Crypto has one session per UTC day; US stock indices and GLD exclude weekends and exchange holidays, including exceptional closures. Missing sessions fail validation rather than being filled with invented prices. Each release starts January 1, 2021 and ends at a completed calendar quarter. The start date stays fixed when the dataset is refreshed. The displayed window and publication date identify the bundled release. Releases update manually after a completed quarter, and every market in a release is validated before publication.
We share the app’s decision and funding logic, using full fills at opening prices with a fixed 0.2% fee. Live approval expiry and 1% price-change rejection are not modeled. Historical exchange quantity steps, minimum orders, spread, slippage, liquidity constraints, taxes, and cash interest are excluded. Index levels are treated as hypothetical tradable units for comparison only; dividends, fund fees and tracking differences are excluded. Cash and crypto quotes use USD. Fees apply to initial buys and all executed corrections, and the ending portfolio is not liquidated.
CAGR annualizes the total return over the actual window using 365.2425 days per year. Calendar-year returns use the previous year's final equity; the first partial year starts from $10,000. Maximum drawdown includes the starting capital as the first peak.
Rebalio's product requires a trade preview and explicit approval. This simulated daily strategy is research, not an automated trading feature or individualized investment advice. Historical performance does not guarantee future results.
Dataset SHA-256: a4c7426d0a42dc2c18d8471f9d40ba21e266a7e1d25238ece75399648b1ff6bd
Retrieved: 2026-10-08T10:55:21.816Z · Release: 2026-09
rebalio