Do classic strategies still work after publication?
We track 55 classic investing strategies from the month after each one was published. Then we compare each one with the S&P 500 over the same dates. Here is how they did.
Key findings#
- Few beat the S&P 500 on return. 2 of 55 strategies earned more a year than the S&P 500 after they were published. The typical one earned 7.1% a year, 6.8 points a year behind the S&P 500.
- Most fell less than the S&P 500. 43 of 55 had a smaller worst drop over the same years. The typical worst drop was −20.7%, against −33.7% for the S&P 500.
- Few earned more for the risk. 9 of 55 had a higher Sharpe ratio than the S&P 500. The Sharpe ratio is the return a strategy earned for each unit of ups and downs.
- The edge faded after publication. In the years of ETF prices before publication, the typical strategy was 0.5 points a year ahead of the S&P 500. After publication it was 6.8 points a year behind the S&P 500.
These are backtests: hypothetical results on past prices, after trading costs. They are not money anyone made.
Every strategy after publication#
Best Sharpe ratio first. In each pair, the first figure is the strategy and the second is the S&P 500 over the same dates.
| Strategy | Published | Return a year | Worst drop | Sharpe ratio |
|---|---|---|---|---|
| HAA Simple: Hybrid Asset Allocation with one asset | Mar 2023 | 14.7% vs 19.5% | −11.2% vs −18.8% | 1.27 vs 1.26 |
| TrendYCMacro (Quantpedia’s market timing strategy) | Sep 2023 | 9.3% vs 17.6% | −8.9% vs −18.8% | 1.00 vs 1.14 |
| Protective Asset Allocation (PAA) | May 2016 | 7.8% vs 14.5% | −11.0% vs −33.7% | 0.98 vs 0.85 |
| Growth-Trend Timing (Philosophical Economics’ GTT) | Feb 2016 | 13.8% vs 14.9% | −21.5% vs −33.7% | 0.95 vs 0.88 |
| Volatility-Managed Portfolio (Moreira and Muir) | May 2016 | 9.6% vs 14.5% | −17.0% vs −33.7% | 0.94 vs 0.85 |
| 200-Day Moving Average Strategy (S&P 500) | Mar 2016 | 11.0% vs 14.8% | −20.4% vs −33.7% | 0.93 vs 0.87 |
| Tactical Permanent Portfolio (GestaltU / ReSolve) | Sep 2012 | 5.2% vs 14.2% | −11.1% vs −33.7% | 0.90 vs 0.88 |
| Harry Browne’s Permanent Portfolio | Jan 2007 | 7.1% vs 10.6% | −17.7% vs −55.2% | 0.90 vs 0.61 |
| Golden Butterfly Portfolio | Oct 2015 | 7.7% vs 14.6% | −18.9% vs −33.7% | 0.90 vs 0.86 |
| Growth-Trend Timing with job growth | Feb 2016 | 14.4% vs 14.9% | −33.7% vs −33.7% | 0.88 vs 0.88 |
| US Risk Parity Trend-Following (The Trend is Our Friend) | Feb 2016 | 4.5% vs 14.9% | −10.6% vs −33.7% | 0.83 vs 0.88 |
| Trinity Portfolio (Meb Faber) | Feb 2017 | 6.9% vs 14.4% | −17.9% vs −33.7% | 0.81 vs 0.83 |
| Leverage Rotation Strategy 2x (LRS with SSO) | Mar 2016 | 17.9% vs 14.8% | −37.4% vs −33.7% | 0.81 vs 0.87 |
| Lethargic Asset Allocation (Keller’s LAA) | Jan 2020 | 9.3% vs 14.1% | −19.7% vs −33.7% | 0.81 vs 0.76 |
| Adaptive Asset Allocation (AAA) | Aug 2012 | 8.0% vs 14.3% | −21.9% vs −33.7% | 0.80 vs 0.89 |
| Hybrid Asset Allocation (HAA) | Mar 2023 | 8.5% vs 19.5% | −14.6% vs −18.8% | 0.79 vs 1.26 |
| Leverage Rotation Strategy (UPRO above the 200-day average) | Mar 2016 | 24.2% vs 14.8% | −51.2% vs −33.7% | 0.79 vs 0.87 |
| Global Tactical Asset Allocation (GTAA 5) | Feb 2013 | 5.1% vs 14.0% | −9.3% vs −33.7% | 0.76 vs 0.87 |
| 12-Month High Switch (Meb Faber) | Jun 2020 | 6.8% vs 16.0% | −18.0% vs −24.5% | 0.76 vs 0.97 |
| Faber’s 10-month moving average timing model | Feb 2013 | 8.6% vs 14.0% | −24.9% vs −33.7% | 0.73 vs 0.87 |
| GTAA Aggressive 6 (Faber GTAA AGG6) | Mar 2013 | 7.7% vs 14.0% | −17.2% vs −33.7% | 0.73 vs 0.87 |
| Ivy Portfolio rotation (Faber’s relative strength model) | May 2010 | 7.5% vs 13.5% | −15.3% vs −33.7% | 0.73 vs 0.83 |
| Baby Bear Portfolio (Muscular Portfolios) | Feb 2018 | 7.3% vs 13.1% | −20.6% vs −33.7% | 0.73 vs 0.74 |
| GTAA 13: Faber’s 13-asset tactical allocation | Mar 2013 | 4.7% vs 14.0% | −9.1% vs −33.7% | 0.72 vs 0.87 |
| Three-Fund Portfolio (Taylor Larimore, Bogleheads) | Jun 2018 | 10.2% vs 14.0% | −28.1% vs −33.7% | 0.72 vs 0.79 |
| Growth-Trend Timing with housing starts and income | Feb 2016 | 8.6% vs 14.9% | −24.9% vs −33.7% | 0.72 vs 0.88 |
| Growth-Trend Timing with unemployment (GTT-UE) | Mar 2016 | 10.9% vs 14.8% | −33.7% vs −33.7% | 0.70 vs 0.87 |
| Global Risk Parity Trend-Following (Clare et al.) | Feb 2016 | 4.3% vs 14.9% | −12.2% vs −33.7% | 0.68 vs 0.88 |
| Tactical Risk Rotation (TRRS) | May 2014 | 7.9% vs 13.3% | −24.9% vs −33.7% | 0.67 vs 0.81 |
| Robust Asset Allocation Balanced (Alpha Architect RAA) | Dec 2015 | 4.8% vs 14.1% | −15.5% vs −33.7% | 0.66 vs 0.84 |
| Defensive Asset Allocation (DAA) | Aug 2018 | 6.0% vs 13.9% | −19.5% vs −33.7% | 0.64 vs 0.78 |
| Three-Way Model (Ned Davis Research) | Jul 2015 | 7.0% vs 13.5% | −20.5% vs −33.7% | 0.64 vs 0.80 |
| All Weather Portfolio (Ray Dalio’s All Seasons) | Dec 2014 | 5.1% vs 13.0% | −23.8% vs −33.7% | 0.64 vs 0.79 |
| GTAA Aggressive 3 (Faber GTAA AGG3) | Mar 2013 | 7.4% vs 14.0% | −18.0% vs −33.7% | 0.61 vs 0.87 |
| Ivy Portfolio (Faber and Richardson, buy and hold) | Feb 2013 | 6.4% vs 14.0% | −27.3% vs −33.7% | 0.61 vs 0.87 |
| Papa Bear Portfolio (Muscular Portfolios) | Feb 2018 | 7.8% vs 13.1% | −20.7% vs −33.7% | 0.60 vs 0.74 |
| Diversified Dual Momentum (Newfound Research) | Feb 2019 | 8.4% vs 15.3% | −31.1% vs −33.7% | 0.59 vs 0.84 |
| Faber sector rotation (the three strongest US sectors) | May 2010 | 6.9% vs 13.5% | −21.5% vs −33.7% | 0.55 vs 0.83 |
| Robust Asset Allocation Aggressive (Alpha Architect RAA) | Dec 2015 | 5.5% vs 14.1% | −25.0% vs −33.7% | 0.55 vs 0.84 |
| Beta Rotation Strategy (utilities vs. the market) | Apr 2014 | 8.2% vs 13.2% | −39.1% vs −33.7% | 0.55 vs 0.81 |
| Vigilant Asset Allocation G12 (VAA-G12) | Aug 2017 | 5.6% vs 14.1% | −29.7% vs −33.7% | 0.54 vs 0.80 |
| Accelerating Dual Momentum (ADM) | Jun 2018 | 7.7% vs 14.0% | −38.5% vs −33.7% | 0.54 vs 0.79 |
| Composite Dual Momentum | May 2012 | 4.3% vs 13.9% | −21.6% vs −33.7% | 0.53 vs 0.87 |
| US Cross-Asset Momentum (stock and Treasury timing) | Jan 2017 | 5.6% vs 14.4% | −33.7% vs −33.7% | 0.53 vs 0.83 |
| Dual Momentum: Antonacci’s Global Equities Momentum (GEM) | Nov 2014 | 7.3% vs 13.1% | −33.7% vs −33.7% | 0.52 vs 0.80 |
| Vigilant Asset Allocation (VAA) | Aug 2017 | 5.0% vs 14.1% | −20.8% vs −33.7% | 0.48 vs 0.80 |
| Mama Bear Portfolio (Muscular Portfolios) | Feb 2018 | 5.4% vs 13.1% | −25.1% vs −33.7% | 0.48 vs 0.74 |
| Paired Switching (quarterly SPY and TLT rotation) | Sep 2011 | 6.6% vs 14.5% | −35.6% vs −33.7% | 0.48 vs 0.89 |
| Kipnis Defensive Adaptive Asset Allocation (KDA) | Feb 2019 | 4.2% vs 15.3% | −17.9% vs −33.7% | 0.48 vs 0.84 |
| Paul Novell’s Tactical Bond Strategy | Jan 2015 | 2.4% vs 13.1% | −10.9% vs −33.7% | 0.42 vs 0.79 |
| Bold Asset Allocation (BAA) | Aug 2022 | 3.5% vs 15.8% | −11.3% vs −18.8% | 0.41 vs 0.99 |
| Trending Fast and Slow (volatility-switched momentum) | Jan 2022 | 4.1% vs 10.5% | −27.6% vs −24.5% | 0.38 vs 0.66 |
| Quint Switching Filtered (Lewis Glenn’s QSF) | Mar 2018 | 1.1% vs 13.9% | −26.9% vs −33.7% | 0.16 vs 0.78 |
| Bold Asset Allocation G4 (BAA aggressive) | Aug 2022 | −0.2% vs 15.8% | −20.4% vs −18.8% | 0.05 vs 0.99 |
| Resilient Asset Allocation (Keller’s RAA) | Jan 2021 | −0.1% vs 13.8% | −25.7% vs −24.5% | 0.05 vs 0.86 |
Too new to judge yet, with under 3 years since publication: Gold Cross-Asset Momentum (gold and Treasury rule) (since Feb 2026), Defense First (Carlson’s defensive tactical model) (since Aug 2025), HAA 60/40 (NLX Finance’s Hybrid Asset Allocation variant) (since Apr 2024), Pragmatic Asset Allocation (Quantpedia’s PAA) (since Feb 2024), Inflation Compass (Varadi’s growth and inflation model) (since Aug 2026). Their own pages show their results so far.
Did the strategies do worse after publication?#
Yes, on return. 49 of these strategies also have at least 3 years of ETF prices before their publication date. Before publication, 28 of them beat the S&P 500, and the typical one was 0.5 points a year ahead of the S&P 500. After publication, 2 of them beat it, and the typical one was 6.8 points a year behind the S&P 500.
Two things can move a strategy after it is published. We cannot fully separate them.
- The market changes. Half of these strategies came out in 2016 or later. Over their years after publication, the S&P 500 earned a typical 14.1% a year. That was a strong run for US stocks. A strategy that holds bonds or cash part of the time to dodge crashes tends to trail when stocks rise fast. For most of them, the years before publication include the 2008 crash, when dodging paid off.
- Publication itself. Authors pick the rules that worked best on the past, and some of that is luck that does not repeat (why past results mislead). A study of 97 stock-market patterns found that their returns were 58% lower after publication (McLean and Pontiff, 2016). Another argues that so many ideas get tested that many published findings are likely flukes (Harvey, Liu and Zhu, 2016).
What this means if you use these strategies#
- Expect less than the paper shows. The backtest in a paper or book is close to a best case. Judge a strategy on what it did after its rules were fixed, and read how to read a backtest before you trust any result.
- Look at the worst drop, not only the return. 43 of 55 strategies fell less than the S&P 500 at their worst. Most of them were built to do that, and they pay for it with lower returns when stocks rise.
- Decide what you want first. If you want the most growth and can sit through a big fall, only 2 of 55 beat simply holding the S&P 500 after publication. If a smaller fall matters more to you, check how far each one fell.
- Test before you trust. Open any strategy in the table to see its rules, its full backtest and a live version you can copy and change in the builder.
This page does not tell you which strategy to use. It shows what happened.
How we measured#
Each strategy in our classics library is built from its author’s own paper, book or post, and its page links that source. For this page, the clock starts at the first month after that source came out. The author could not have tuned the rules to anything after that date, so it is a fair test. Researchers call it “out of sample”, and it guards against the traps in why past results mislead.
- Same dates. Each strategy is compared with the S&P 500 (the SPY fund) over exactly its own dates after publication.
- Real prices. Every figure uses daily prices of real ETFs, with dividends reinvested and a cost of 0.05% on every trade. No simulated history is used on this page.
- Enough time. A strategy counts once it has at least 3 years after publication. Newer ones are listed but not counted.
- Dated sources only. Rules with no single author or date, such as the 60/40 portfolio and the golden cross, are left out. So is any strategy measured against something other than the S&P 500.
- Updated daily. The figures come from the same results as each strategy’s own page and are recomputed after every US market close.
Our methodology lists every assumption our backtests make, and how to read a backtest explains each measure in the table.
How to cite this page#
Tactfolio Research Lab, “Do classic strategies still work after publication?”, Tactfolio, figures through November 18, 2026. https://tactfolio.com/strategies/classics/scorecard
You are welcome to quote these figures. Please link to this page, because the figures change after every market close.
Common questions#
Do tactical asset allocation strategies still work?#
It depends on what you want them to do. After publication, 43 of 55 classic strategies had a smaller worst drop than the S&P 500, and 2 of 55 earned more a year. So most of them helped you lose less in a crash, but most earned less than simply holding US stocks.
Which classic strategy did best after publication?#
On Sharpe ratio, HAA Simple: Hybrid Asset Allocation with one asset did best: 1.27 against 1.26 for the S&P 500 over 3.7 years. On return alone, Leverage Rotation Strategy (UPRO above the 200-day average) earned the most: 24.2% a year against 14.8%, with a worst drop of −51.2%.
Why can a strategy do worse after it is published?#
Two reasons. Authors choose the rules that worked best on past data, and part of that is luck. And the market after publication may simply be different from the years the author tested.
Is beating the S&P 500 the right test?#
Not on its own. Many of these strategies are tactical asset allocation rules that hold bonds or cash part of the time. They aim to lose less, not to earn more. That is why this page also shows the worst drop and the Sharpe ratio next to the return.
How often is this page updated?#
After every US market close. The figures come from the same daily results as each strategy’s own page.
Sources#
- R. David McLean and Jeffrey Pontiff, Does Academic Research Destroy Stock Return Predictability?, The Journal of Finance 71(1), 2016
- Campbell R. Harvey, Yan Liu and Heqing Zhu, ... and the Cross-Section of Expected Returns, The Review of Financial Studies 29(1), 2016
- William F. Sharpe, The Sharpe Ratio, The Journal of Portfolio Management 21(1), 1994
- Each strategy’s own paper, book or post, linked under Sources and caveats on its page