Adaptive Asset Allocation (AAA)
Adam Butler, Michael Philbrick, Rodrigo Gordillo, and David Varadi's Adaptive Asset Allocation, in its momentum and inverse-volatility form. Each month it holds the five strongest of ten global asset classes by six-month return, weighted so that calmer assets get more, by their volatility over the last 60 days.
Designed by Adam Butler, Michael Philbrick, Rodrigo Gordillo, and David Varadi, 2012. Implemented and tracked by Tactfolio.
| Jun 2007 to Nov 2026 | Strategy | SPY |
|---|---|---|
| Annual return (CAGR) | 9.7% | 10.5% |
| Worst drawdown | −21.9% | −55.2% |
| Sharpe ratio | 0.88 | 0.60 |
| Volatility | 11.2% | 19.7% |
| Annual return since publication (Aug 2012) | 8.0% | 14.3% |
| Jun 1970 to Nov 2026 | Strategy | SPY |
|---|---|---|
| Annual return (CAGR) | 12.5% | 11.5% |
| Worst drawdown | −21.9% | −55.2% |
| Sharpe ratio | 1.21 | 0.72 |
| Volatility | 10.2% | 17.2% |
| Annual return since publication (Aug 2012) | 8.0% | 14.3% |
Adaptive Asset Allocation (AAA) is a monthly strategy that holds the five strongest of ten global asset classes and gives the calmer ones more money. Adam Butler, Michael Philbrick, Rodrigo Gordillo, and David Varadi introduced it in 2012. Momentum chooses what to own, and recent risk decides how much of each.
Key points#
- Adaptive Asset Allocation updates both what it holds and how much, every month.
- It keeps the top five of ten funds by six-month return.
- Calmer funds get bigger weights, so no single wild fund drives the risk.
- Since mid-2007 its worst drop was about 22%, less than half the S&P 500’s.
- It trailed the S&P 500 in 11 of the 13 years from 2012 through 2024.
What is Adaptive Asset Allocation?#
Adaptive Asset Allocation is a portfolio that adapts to the market as it is now, instead of using fixed weights. Its authors argued that the classic fixed-weight portfolio relies on long-run averages that rarely hold in practice.
A traditional portfolio, such as a 60/40 mix or an equal mix of asset classes, sets its weights once. It assumes that returns and risks stay stable. The AAA authors showed that recent returns and recent volatility forecast the near future better than long-run averages do. So a portfolio can do better by updating both every month.
What two ideas does AAA combine?#
AAA combines two well-documented effects: momentum and risk balance.
- Momentum is the tendency of assets that have done well over the last several months to keep doing well for a while.
- Risk balance means sizing positions by their recent volatility. Volatility is how much a price swings up and down. This keeps one volatile asset from dominating the portfolio’s risk.
The primer’s fullest version also weighs how the assets move together.
How does Adaptive Asset Allocation work?#
At the close of the last trading day of each month, AAA ranks, trims and weights its funds in three steps.
- Rank ten asset classes by their six-month return. The ten are listed in the table below.
- Keep the top five.
- Weight them by inverse volatility. Measure each fund’s volatility over the last 60 trading days. Give it a weight in line with one divided by that volatility, so the calmer funds get more.
| Asset class | Fund |
|---|---|
| US stocks | SPY |
| European stocks | VGK |
| Japanese stocks | EWJ |
| Emerging markets | EEM |
| US real estate | VNQ |
| International real estate | RWX |
| Intermediate Treasuries | IEF |
| Long Treasuries | TLT |
| Commodities | DBC |
| Gold | GLD |
Treasuries are bonds issued by the US government. Commodities are raw goods such as oil, metals and crops.
Here is a made-up example of step 3 with two funds. Fund A swings 10% a year and fund B swings 20%. One divided by each gives 0.1 and 0.05, so A gets twice B’s weight. That works out to about 67% in A and 33% in B.
This is the primer’s momentum and inverse-volatility version (Exhibit 4). Its headline version (Exhibit 5) goes a step further. It holds the same five in the mix with the lowest expected volatility, which also counts how they move together. Tactfolio does not offer that minimum-variance weighting, so this page tests the simpler step.
What the backtest shows#
Figures quoted in this article's text were measured on prices through October 9, 2026. The chart and tables are recomputed after every market close.
Over the ETF era, from mid-2007, AAA returned about 10.1% a year, against 10.9% for the S&P 500 and 8.1% for a 60/40 portfolio. It beat both on risk-adjusted return, meaning return for each unit of risk.
Its worst drawdown was about 22%, less than half the S&P 500’s. A drawdown is how far a portfolio falls from its last high point. It gained 8.5% in 2008, when the S&P 500 lost more than a third.
Since the primer was published in 2012 it has returned about 8.5% a year, with a risk-adjusted return a little below its full-period figure. It trailed the S&P 500 in 11 of the 13 years from 2012 through 2024, the long run of US stock leadership. Then it beat the S&P 500 in 2025.
Its worst drawdown, about 22%, ran from November 2021 to September 2022. Stocks and bonds fell together then, and momentum offered few safe places. It lost 15% in 2022 and recovered its high in July 2024.
With simulated history, the test starts in mid-1970 and returns about 12.6% a year. That is about a point more than the S&P 500, with a much better risk-adjusted return.
Read the years before September 2004 with care. International real estate (RWX) is then modelled as 30% US real estate and 70% developed-market stocks. That stand-in grew about 5% a year faster than RWX over the years the two can be compared, so those simulated years may flatter AAA.
All of these are hypothetical backtests on past prices, after trading costs. They are not a track record. Our guide to reading a backtest explains what such numbers can and cannot show.
When does Adaptive Asset Allocation struggle?#
AAA struggles when most asset classes fall at once or when a trend turns suddenly.
- Stock and bond sell-offs together. With both falling, the top five by momentum can still all be losing.
- Trend reversals. Six-month momentum is slow to leave a leader that has just turned.
How to try Adaptive Asset Allocation on Tactfolio#
The live strategy at the top of this page runs these rules at each month-end close. Copy it to try a different lookback, a different number of holdings, a different volatility window, or your own set of ten asset classes. The visual builder lets you change each of these and rerun the test.
If you want a simpler momentum rule to compare, Dual Momentum holds one fund at a time. More ideas are in our classic strategies.
Year by year
| Year | Strategy | SPY |
|---|---|---|
| 2026* | 10.3% | 7.9% |
| 2025 | 21.1% | 17.7% |
| 2024 | 9.3% | 24.9% |
| 2023 | 10.4% | 26.2% |
| 2022 | −15.0% | −18.2% |
| 2021 | 21.7% | 28.7% |
| 2020 | 15.8% | 18.3% |
| 2019 | 10.2% | 31.2% |
| 2018 | −3.6% | −4.6% |
| 2017 | 20.1% | 21.7% |
| 2016 | 7.8% | 12.0% |
| 2015 | −5.9% | 1.2% |
| 2014 | 4.0% | 13.5% |
| 2013 | 12.7% | 32.3% |
| 2012 | 9.1% | 16.0% |
| 2011 | 9.7% | 1.9% |
| 2010 | 16.5% | 15.1% |
| 2009 | 17.3% | 26.4% |
| 2008 | 8.5% | −36.8% |
| 2007* | 15.7% | −1.9% |
| Year | Strategy | SPY |
|---|---|---|
| 2026* | 10.3% | 7.9% |
| 2025 | 21.1% | 17.7% |
| 2024 | 9.3% | 24.9% |
| 2023 | 10.4% | 26.2% |
| 2022 | −15.0% | −18.2% |
| 2021 | 21.7% | 28.7% |
| 2020 | 15.8% | 18.3% |
| 2019 | 10.2% | 31.2% |
| 2018 | −3.6% | −4.6% |
| 2017 | 20.1% | 21.7% |
| 2016 | 7.8% | 12.0% |
| 2015 | −5.9% | 1.2% |
| 2014 | 4.0% | 13.5% |
| 2013 | 12.7% | 32.3% |
| 2012 | 9.1% | 16.0% |
| 2011 | 9.7% | 1.9% |
| 2010 | 16.5% | 15.1% |
| 2009 | 17.3% | 26.4% |
| 2008 | 8.5% | −36.8% |
| 2007 | 23.6% | 5.1% |
| 2006 | 20.5% | 15.8% |
| 2005 | 6.9% | 4.8% |
| 2004 | 16.4% | 10.7% |
| 2003 | 39.6% | 28.2% |
| 2002 | 9.6% | −21.6% |
| 2001 | 3.5% | −11.8% |
| 2000 | 8.1% | −9.7% |
| 1999 | 25.2% | 20.4% |
| 1998 | 11.6% | 28.7% |
| 1997 | 13.2% | 33.5% |
| 1996 | 16.4% | 22.5% |
| 1995 | 23.7% | 38.0% |
| 1994 | 1.6% | 0.4% |
| 1993 | 13.3% | 9.7% |
| 1992 | 5.8% | 7.6% |
| 1991 | 11.6% | 30.3% |
| 1990 | −2.0% | −3.2% |
| 1989 | 17.1% | 31.5% |
| 1988 | −1.1% | 16.4% |
| 1987 | 15.6% | 5.1% |
| 1986 | 51.8% | 18.6% |
| 1985 | 38.5% | 31.6% |
| 1984 | 6.2% | 6.2% |
| 1983 | 15.6% | 22.4% |
| 1982 | 25.0% | 21.6% |
| 1981 | −5.4% | −4.9% |
| 1980 | 3.7% | 32.4% |
| 1979 | 40.4% | 18.3% |
| 1978 | 23.6% | 6.5% |
| 1977 | 7.8% | −7.2% |
| 1976 | 11.7% | 23.8% |
| 1975 | 5.2% | 37.1% |
| 1974 | 3.0% | −26.5% |
| 1973 | −1.4% | −14.7% |
| 1972 | 32.1% | 18.9% |
| 1971 | 7.8% | 14.1% |
| 1970* | 13.1% | 28.9% |
* Partial year.
The rules as implemented
This is the exact tree Tactfolio runs monthly, rebalancing every run, with signals and trades at the close. Open it to inspect or copy it.
Make it yours
Copy Adaptive Asset Allocation and change one rule
Swap a fund, change a lookback or pick another rebalance day. Then compare your backtest with this one, and run your version on paper.
Needs Plus, $12 a month or save 25% paid yearly. Plus and Pro start with a 7-day free trial. 14-day money-back guarantee.
Sources and caveats
- Adam Butler, Michael Philbrick, Rodrigo Gordillo, and David Varadi, Adaptive Asset Allocation: A Primer (2012)
- ReSolve, Adaptive Asset Allocation whitepaper (2015 revision)
- Follows the primer's Exhibit 4: the top 5 of 10 by six-month (126-session) return, weighted by inverse volatility, here over the last 60 daily returns. The primer's headline version (Exhibit 5) holds the same five in their minimum-variance mix, which also weighs how they move together; Tactfolio does not offer minimum-variance weighting.
- VGK stands in for EZU as European stocks.
- Signals and trades use the close of the last trading day of each month, as in the source.
Common questions#
What is Adaptive Asset Allocation?#
It is a strategy from Butler, Philbrick, Gordillo, and Varadi (2012). Each month it holds the five strongest of ten global asset classes by six-month return. The calmer ones get more weight.
Is this the minimum-variance version of AAA?#
No. It weights the top five by inverse volatility over their last 60 days of returns, the primer’s Exhibit 4. The primer’s headline version (Exhibit 5) holds them in their minimum-variance mix, which also accounts for how they move together. Tactfolio does not offer minimum-variance weighting, so that version cannot be built here.
Which ETFs does Adaptive Asset Allocation use?#
It uses SPY, VGK, EWJ, EEM, VNQ, RWX, IEF, TLT, DBC, and GLD. VGK stands in for EZU as the European stock fund.
How often does AAA rebalance?#
AAA rebalances monthly, at the close of the last trading day. To rebalance means to reset each holding to its target weight.
What is inverse volatility weighting?#
Inverse volatility weighting gives each fund a weight based on one divided by its recent volatility. A fund that swings half as much gets twice the weight. The aim is for each holding to add a similar amount of risk.
Can I automate Adaptive Asset Allocation?#
Yes. Copy the live strategy and run it in Trade, on a Tactfolio paper account or in your own Alpaca account, paper or live. Trade runs it on a daily schedule and only trades when the monthly rule calls for a change. It is long only and holds US-listed ETFs, and you choose how much it starts with.
Keep exploring
Run it automatically
The funds it holds
- SPY ETF: fees, returns, drawdowns and holdings
- VGK ETF: fees, returns, drawdowns and holdings
- EWJ ETF: fees, returns, drawdowns and holdings
- EEM ETF: fees, returns, drawdowns and holdings
- VNQ ETF: fees, returns, drawdowns and holdings
- RWX ETF: fees, returns, drawdowns and holdings
- IEF ETF: fees, returns, drawdowns and holdings
- TLT ETF: fees, returns, drawdowns and holdings
- DBC ETF: fees, returns, drawdowns and holdings
- GLD ETF: fees, returns, drawdowns and holdings
How its funds compare
- IEF vs TLT: fees, holdings and returns compared
- SPY vs VOO: fees, holdings and returns compared
- SCHH vs VNQ: fees, holdings and returns compared
- GLD vs IAU: fees, holdings and returns compared
Learn the ideas behind it
- Momentum strategy: types, rules and an example
- Momentum stocks: what they are and how to trade them
- Tactical asset allocation vs strategic, with backtests
Check the numbers
- Do classic strategies still work after publication?
- CAGR calculator and reverse CAGR calculator
- Monte Carlo simulation: will your money last?