How GEM works
Global Equities Momentum is the model that made dual momentum famous, introduced by Gary Antonacci. Once a month it asks two questions. First, absolute momentum: has the US stock market beaten cash over the last 12 months? If not, the model holds bonds. Second, relative momentum: if US stocks did beat cash, which was stronger over those 12 months, US stocks or international stocks? The model holds the winner.
This template uses VOO for US stocks, VXUS for international stocks, BND for bonds and BOXX (with Treasury bill history before its launch) for cash. Change any of them in the settings, or switch the lookback to a multi-period blend to see how sensitive the result is to the single 12-month window.
What is dual momentum?
Dual momentum combines two simple ideas. Relative momentum compares assets with each other and holds the ones that have done best recently. Absolute momentum compares an asset with cash, and steps aside into something defensive when it is not even beating a Treasury bill. Relative momentum keeps you in the strongest markets; absolute momentum is what gets you out of long bear markets. You can read more about how momentum is measured and Risk On and Risk Off.
Dual momentum and rotation strategies
Classic dual momentum, like GEM, is a two-horse race: one US stock fund against one international fund, with bonds as the exit. A rotation strategy widens the field. It ranks a larger list of ETFs, often across asset classes such as US and foreign stocks, gold, commodities and Treasuries, and holds the top few. The same two ideas still drive it: relative momentum decides which ETFs to own, and absolute momentum decides whether each one is strong enough to own at all.
Rotation strategies usually differ from GEM in three ways:
- More holdings. Holding the top two or three instead of one spreads the bet, so a single wrong call costs less.
- Blended lookbacks. Scoring on a mix of 1, 3, 6 and 12 month returns reacts faster to turns than a single 12 month window, at the cost of more trading.
- Per-asset cash tests. Each pick has to beat cash on its own, so the portfolio can step down to one third or two thirds defensive rather than switching all at once.
Which works better depends on what you want to avoid. GEM trades rarely and is easy to follow, but it is all in or all out. A rotation spreads risk and adjusts gradually, but trades more, which is why the friction setting matters. The Global Rotation template is a ready-made example across seven ETFs; switch between it and the GEM template to compare them over the same years.
How to use this calculator
Start from a template, then change anything. Every setting updates the backtest immediately, and the share link reproduces your exact configuration.
- Momentum candidates are the ETFs the model chooses between. Hold the top sets how many it holds at once.
- Lookback is the window momentum is measured over: one period, or a blend of several with your own weights.
- Absolute momentum either tests each pick on its own, or uses one ETF (such as VOO in GEM) to decide Risk On or Risk Off for the whole portfolio. The hurdle is cash or zero.
- Risk Off is where money goes when the test fails: a fixed ETF, or the strongest of a basket of bond funds.
- Weighting splits money equally between holdings, or by equal risk, giving calmer ETFs a bigger share.
- Buy and hold sleeve keeps a fixed slice, such as managed futures or gold, alongside the momentum portion.
- Rebalancing resets weights monthly, each January, or only when a holding drifts outside a band.
- Trading friction is on by default and charges every trade, because nobody earns frictionless returns in a real account. See how the friction model works.
History back to 1980
Most dual momentum backtests start when the ETFs launched, which leaves out the 1980s, the 1987 crash, the 2000 to 2002 bear market and, for many funds, 2008. The histories here are extended back to 1980 using the indexes and funds each ETF tracks, so you can see how a rule behaved through several very different markets. See extended returns and where the data comes from.
Reading the results
The backtest reports CAGR, maximum drawdown, volatility, Sharpe and Sortino ratios and the Ulcer Performance Index, side by side with the benchmark you choose. Numbers can change noticeably with the start date; here is why.
Want more than dual momentum?
The strategies on DualMomentumSystems.com go further: Treasury duration limits, smart leverage, credit spread signals and more, tracked live every month. Browse every strategy.
ETFs available in the calculator
Every ETF below can be used as a momentum candidate, a gate, a Risk Off asset, part of a buy-and-hold sleeve or a benchmark. The start year is where the monthly history begins; earlier years for younger ETFs come from a matching index or fund, as explained in where the data comes from. Typing SPY or GLD works too; they map to VOO and SGOL.
71 ETFs
| Ticker | Description | History from |
|---|---|---|
| ALLW | All Weather | 1980 |
| BIL | 1-3 Month Treasury Bills | 1979 |
| BIZD | Business Development Companies | 2013 |
| BKLN | Senior Loans | 2011 |
| BND | Total Bond Market | 1979 |
| BOXX | 1-3 Month Box Spread | 1979 |
| BUYO | Buyout Strategies | 2025 |
| BWX | International Treasuries | 1990 |
| CAOS | Tail Risk | 1979 |
| CLSE | Long/Short Equity | 1979 |
| DBEF | EAFE Hedged Equity | 1979 |
| DBMF | Managed Futures | 1979 |
| DRN 3x | 3x Real Estate | 1986 |
| EDC 3x | 3x Emerging Markets | 1990 |
| EDV | Extended Duration Treasuries | 1979 |
| EEM | Emerging Markets | 1979 |
| EWJ | Japan | 1979 |
| FAS 3x | 3x Financials | 2008 |
| FTLS | Long/Short Equity | 2014 |
| HFEQ | Equity Long/Short | 2022 |
| HFGM | Global Macro | 1979 |
| HFMF | Managed Futures | 1979 |
| HYG | High Yield Corporate Bonds | 1979 |
| IEF | 7-10 Year Treasuries | 1979 |
| IFRA | U.S. Infrastructure | 2018 |
| IWB | Russell 1000 | 1979 |
| IWD | Russell 1000 Value | 2000 |
| IWM | Russell 2000 | 1979 |
| IWP | Russell Mid-Cap Growth | 1979 |
| IWR | Russell Mid-Cap | 1979 |
| IWS | Russell Mid-Cap Value | 1979 |
| IWV | Russell 3000 | 1979 |
| JAAA | AAA CLOs | 2020 |
| JBBB | B-BBB CLOs | 2022 |
| KMLM | Managed Futures | 2021 |
| LQD | Investment Grade Corporate Bonds | 1979 |
| PDBC | Diversified Commodities | 1979 |
| PEX | Global Listed Private Equity | 2013 |
| PSP | Global Listed Private Equity | 2006 |
| QLD 2x | 2x Nasdaq-100 | 1979 |
| QQQ | Nasdaq-100 | 1979 |
| REET | Global REITs | 2014 |
| SCZ | International Small Cap | 1996 |
| SGOL | Gold | 1979 |
| SHNY 3x | 3x Gold | 1979 |
| SHY | 1-3 Year Treasuries | 1979 |
| SPMO | US Large Cap Momentum Factor | 2015 |
| SRLN | Senior Loans | 2013 |
| SSO 2x | 2x S&P 500 | 1979 |
| TIP | TIPS | 1979 |
| TLT | 20+ Year Treasuries | 1979 |
| TMF 3x | 3x 20+ Year Treasuries | 1979 |
| TQQQ 3x | 3x Nasdaq-100 | 1979 |
| UPRO 3x | 3x S&P 500 | 1979 |
| URTY 3x | 3x Russell 2000 | 1979 |
| VBR | Vanguard Small-Cap Value ETF | 1979 |
| VCSH | Short-Term Corporate Bonds | 1979 |
| VEA | Developed Markets | 1979 |
| VGIT | Intermediate Treasuries | 1979 |
| VGK | Europe | 1979 |
| VGSH | Short-Term Treasuries | 1979 |
| VNQ | U.S. REITs | 1979 |
| VNQI | International Real Estate | 2004 |
| VOO | S&P 500 | 1979 |
| VPL | Pacific Markets | 1979 |
| VTI | Total U.S. Stock Market | 2001 |
| VXUS | Total International Stock Market | 1979 |
| XLE | Energy | 1979 |
| XLK | Technology | 1999 |
| XLP | Consumer Staples | 1999 |
| XLU | Utilities | 1999 |
Backtests are hypothetical and are not a promise of future results. Returns before each ETF's launch are estimates built from comparable indexes and funds. This calculator is educational and is not investment advice.