I’m undertaking a 1000-day reinvention project, blogging here daily to track my progress. In Monday Money, I write about money management.
Today: revisiting my implementation of Meb Faber’s global tactical asset allocation model. In the process, identifying my investing principles.
I am reconsidering two aspects:
- The allocation I’m using especially the use of momentum funds for U.S. large and small caps (SPMO and XSMO, respectively) vs broad market funds (SPY or SPYM and IWM).
- Whether I want to keep doing weekly tranche-based rotations or not, with rebalancing vs switching to monthly-only rotations without rebalancing.
I was going to look at the papers I’m basing this approach on:
- A Quantitative Approach to Tactical Asset Allocation (2007) by Faber
- Global Tactical Asset Allocation: Updated Results and Real-Market Implementation Using Python and IBKR (2025) by Gabriel, Pagani, and Zarattini
And compare the various backtesting statistics.
But as a data scientist, I know that a backtest can’t tell you what will work in the future, even if it gives you some guidance for what to think about.
These past few years I’ve learned so much as an investor, even if my results were nothing stunning for the most part. Maybe I’ll start with the principles I want to base my investing on going forward:
- No system works at all times. To succeed with a particular portfolio allocation approach you must stick with it through the bad times to get to the good times. For example, a portfolio that includes diversifying sleeves such as commodities and international funds has lagged many times in the past fifteen years relative to the S&P 500 index. But there have been times of outperformance from these sleeves. If you exit a system because certain aspects of it haven’t done well recently, you may exit right before those aspects helped you. Caveat: this of course assumes you are following a reasonable approach for example, a diversified set of ETFs that cover a broad range of asset classes (U.S. stocks, foreign stocks, government and corporate bonds, commodities, cash).
- You must pick an approach that works with your temperament as an investor. If you are given to panicking during drawdowns like me, ideally you come up with an approach to managing this. My developing approach is called Structured Panicking, which allows me to get out during the beginnings of a potential drawdown but also tells me when to get back in (just as important!).
- Also your approach must be manageable with the time and attention you have available. Buy and hold with quarterly rebalancing can work fine, if you are diligent about sticking with the quarterly rebalancing. What I’m finding out is that weekly tranche-based rotation is too frequent for me. I don’t want to mess around with my portfolio that often.
- Be reasonable about what kind of returns you can get. Yes, I know if you put all your money into an S&P 500 index fund ten years ago your return would have been almost 14% annualized. That is incredible! It’s also not something you can count on. From 2000 to 2004, the S&P 500 returned almost a negative 20% (considering price only) or about -11% (with dividends reinvested). Of course there was inflation during that time (there always is) so your real returns were worse. Instead of comparing my retirement returns to the S&P 500 alone, I compare them to a moderately aggressive model portfolio that Schwab provides in its Portfolio Performance tab. This has returned an average of around 7.6% over long periods of time, and is a more meaningful benchmark to use for my diversified approach to retirement investing.
- You can’t predict the future. Let me say that again! You cannot predict the future. You don’t know whether the S&P 500 is going to return an annualized 14% for the next five years. Maybe it will! You don’t know that inflation is going to surge like in the seventies, or moderate back to where the Fed would like it. You don’t know where oil or ag commodity prices are going. You don’t know that European stocks are going to languish because Europe sucks as a business environment. I know there are macro traders who make a living based on predicting the future. But it’s never worked for me! So this is one of my big principles.
- But charts and indicators can give you a little bit of an edge. A stock or ETF below its 200-day moving average is less likely to do well going forward than one above it. A ticker that has seen a march to progressively higher lows and higher highs is a better bet than one that is seeing lower lows and lower highs. You can fold this into your portfolio allocation approach if you want to. I want to! It’s part of my structured panicking.
Based on those principles, here’s what I want to do:
- Move to a monthly rotation in my Structured Panicking portfolio.
- Switch from SPMO to SPY (or SPYM, smaller share price) and XSMO to IWM or other Russell 2000 small cap fund. These just feel more emotionally safe to me!
- Rebuild my current Portfolio Steward app mode to enact these changes.
A lot of work to do, but it’s good work.