We designed our public dual momentum model, GEM, to be simple and easy to use for do-it-yourself investors. GEM exists to help protect smaller investors from significant drawdowns while allowing them to earn returns that exceed the market over the long run.
But momentum works best when it incorporates multiple trend determinants. There is a synergistic effect when doing that.
Most investors do not give enough importance to price trends. Greyserman & Kaminsky show that simple trend following has outperformed buy-and-hold and reduced downside excursions back to the beginnings of many markets. No other investment factor has shown that.
Our models are the culmination of a lifetime of investment research and experience. Three of our four proprietary models use a channel breakout approach validated on 100 years of data as seen in our award-winning research study. We also insist on real world superior performance.
Richard Dennis taught something similar to this to his “turtle traders.” Jack Dreyfus became a billionaire using channel breakouts of stocks making new highs.
Most of our models incorporate mean reversion from overbought and oversold conditions as a complement to trend following. Our proprietary models use daily data and are highly adaptive to market conditions. They are based on rigorous, academic-quality research with validation on at least 100 years of data. This is the only way to be confident you haven’t overfit your data.
We spend considerable time on due diligence to find the best ETFs for our models and on portfolio construction. Thoughtful portfolio structuring is an essential part of our investing process. Doing so gives consistency, stability, and rebalancing profits. The construction of barbell-type portfolios as per Nassim Taleb is an important contributor to our success.
Our models work in many different markets. We chose ones where their combinations create optimal balanced portfolios that are responsive to different market conditions. Multiple models with low to moderate correlations are the best way to reduce model estimation error and uncertainty. Optimal combinations of models can enhance expected returns and reduce downside exposure.
We license our proprietary model signals to substantial private and institutional investors, as well as to select investment advisors who understand and appreciate what we do.
Most trading models exit to a safe harbor asset when not in their risk-on positions. Our models are unique in that they switch to other assets or models with positive trends before seeking a safe harbor. This layering captures additional profits and helps reduce whipsaw losses. Our models also all use the principle of confirmation of model signals from other closely aligned assets. Here are our current proprietary models.
SMURF is a stock market oriented model that combines mean reversion with trend and short-term calendar anomalies. It buys an S&P 500 ETF and may take modest positions in other stock market or non-stock ETFs (commodities, managed futures, long-term bonds) when their trends are positive. SMURF can also buy U.S. and non-U.S. stock market ETFs for calendar anomaly and short-term mean reversion trades. SMURF exits to safer assets when trends are no longer positive or on abnormal strength.
BADASS applies our trend model to ETFs representing blockchain technology stocks. It also allocates a moderate amount to a spot Ethereum ETF when its trend is positive. BADASS has been our most profitable model.
GLTR applies trend following to gold ETFs. Gold is often mean-reverting and challenging to trade, but our trend strategy handles it well. GLTR also incorporates short-term mean reversion and calendar anomalies. Even without trend following and mean reversion, gold has outperformed the S&P 500 over the past 25 years. GLTR has a low correlation to our other models and is usually a good portfolio diversifier.
FORT is our latest proprietary model. It is unique in that it trades the equity curve of managed futures ETFs by buying the dips and selling the rallies. FORT is an ideal compliment to our more traditional trend-following models and an important component of our barbell-based portfolios.
SALT holds a short-term Treasury bill ETF. It occasionaly holds short-term positions in long duration Treasury bonds or high yield corporate bonds to exploit mean reversion and calendar anomalies. SALT has the lowest volatility and the lowest correlation of all our models. It excels at giving our portfolios stability and rebalancing profits.
Here are the results of our SMURF, GLTR, and SALT models since 2008.
GLTR spends around half its time in gold stock ETFs and the rest of its time in the SMURF and SALT positions. SMURF spends around 63% of its time in the stock market and 37% in the SALT positions. SALT holds mostly a short-term Treasury bill ETF. 60/20/20 is a balanced allocation of 60% SMURF, 20% GLTR, and 20% SALT.
S&P500 | SMURF | GLTR | SALT | 60/20/20 | |
CAGR | 10.6 | 19.9 | 20.5 | 4.8 | 17.1 |
STD DEV | 16.6 | 10.4 | 13.3 | 3.1 | 7.5 |
SHARPE | 0.62 | 1.69 | 1.37 | 1.09 | 1.97 |
UPI | 1.02 | 9.95 | 7.11 | 10.60 | 13.45 |
MAX DD | -52.9 | -9.3 | -9.6 | -2.4 | -6.1 |
Results do not guarantee future success nor represent returns that any investor attained. All trading involves risks that may not be foreseen. CAGR is the compound annual growth rate. Drawdowns are on a month-end basis. UPI is the Ulcer Performance Index, which divides return by the Ulcer Index. The Ulcer Index measures the depth and duration of drawdowns from earlier highs.
Because of our models’ risk controls, we can use assets more aggressively than buy-and-hold investors. We also reduce portfolio volatility by combining models with modest correlations and by including low-risk Treasury bills with our SALT model. Combining low and high volatility with modest correlations creates the desirable “barbell effect.” This can lead to additional rebalancing profits and greater portfolio stability.
Here are the BADASS results with allocations to SMURF, BADASS, SALT, and GLTR in that order. SALT is the also the final backstop for all the models when their trends are negative. S&P 500 and BLOK ETFs are shown as benchmarks. Real-time performance since January 2018, when we went public with GLTR and BADASS, has been similar to these portfolio numbers.
SMURF | BADASS | SALT | GLTR | S&P500 | BLOK | 20/30/30/20 | 15/35/35/15 | |
CAGR | 24.5 | 83.6 | 5.9 | 23.7 | 13.4 | 16.9 | 35.0 | 36.7 |
STDDEV | 11.5 | 36.9 | 2.7 | 12.7 | 17.3 | 41.0 | 13.0 | 14.3 |
SHARPE | 1.77 | 1.78 | 1.27 | 1.56 | 0.68 | 0.52 | 2.21 | 2.12 |
UPI | 15.37 | 23.94 | 17.85 | 14.13 | 1.15 | 0.86 | 29.49 | 29.31 |
MAX DD | -7.4 | -12.9 | -2.4 | -5.9 | -23.8 | -68.8 | -4.2 | -4.6 |
AVG DD | -1.0 | -2.2 | -0.1 | -0.9 | -4.4 | -22.3 | -0.5 | -0.6 |
W%MOS | 77 | 76 | 86 | 71 | 64 | 60 | 79 | 79 |
FORT and BADASS have the highest expected returns among our models, while SALT is the most stable. FORT, BADASS, and SALT therefore create near-ideal barbell-type portfolios. SMURF and GLTR are also included significantly in the portfolios through model layering. Here is a range of barbell allocations for more aggressive, moderately aggressive, and less aggressive portfolios that combine BADASS, FORT, and SALT in that order. These have the best reward-to-risk characteristics among all our models’ allocations and are the ones we use. The S&P 500 is shown as a benchmark.
S&P500 | BADASS | FORT | SALT | 30/30/40 | 25/25/50 | 20/20/60 | |
CAGR | 16.3 | 107.3 | 33.0 | 5.9 | 35.3 | 29.3 | 19.2 |
STD DEV | 17.7 | 39.4 | 11.1 | 2.7 | 11.0 | 8.8 | 6.8 |
SHARPE | 0.68 | 1.94 | 2.44 | 1.27 | 2.52 | 2.59 | 2.67 |
SORTINO | 1.15 | 9.31 | 6.97 | 3.64 | 13.37 | 14.12 | 15.07 |
UPI | 2.12 | 28.49 | 19.72 | 17.85 | 46.96 | 49.81 | 52.37 |
MAX DD | -23.8 | -12.9 | -8.9 | -2.4 | -2.9 | -2.3 | -1.9 |
AVG DD | -4.5 | -2.2 | -0.7 | -0.1 | -0.3 | -0.2 | -0.2 |
W% MOS | 64 | 76 | 79 | 86 | 81 | 81 | 83 |
Results do not guarantee future success nor represent returns that any investor attained. All trading involves risks that may not be foreseen. CAGR is the compound annual growth rate. Drawdowns are on a month-end basis. UPI is the Ulcer Performance Index, which divides return by the Ulcer Index. The Ulcer Index measures the depth and duration of drawdowns from earlier highs.
