Lesson 3 | Strategy Diversification
Hello Bubble Riders!
I ended the last lesson with a problem: identifying statistically uncorrelated assets (like stocks or cryptos) turns on an assumption, namely that past history will repeat in the future.
But why should it?
People often call AI (or better: machine learning) algorithms “black box” algorithms precisely because they can’t answer that question–that’s how they’re programmed. You have no idea what’s going on in there.
The primary reason I use “white box” algorithms (and Ray Dalio also insists on this) turns on the fact that I can have an answer to that question. I know what the rules are that make the algo run. It’s a “white box” algorithm.
If your goal is to diversify away risk, you need to know why uncorrelated returns have been uncorrelated historically. Otherwise, you’re just guessing–and that’s not how you make money long term.
In this lesson, I’ll show you how that works and we’ll stumble into one of the strangest paradoxes of portfolio management theory: it often turns out that adding lower returning strategies (or asset classes) boosts overall returns in the long run.
This is why my subscribers get access to so many strategies.
As a reminder that we at AOTB are giving away 1 year’s worth of the Crypto Rider subscription for mostly doing what you’re already doing.
Just click this link. You get additional points for following on FB and Twitter (and referring a friend).
Additionally, we’re going to make the product offerings on Patreon even better.
The DIY-er Plan will now feature 2 separate reports. One is the Yield Farming report, which would be a top hedge fund if it were compared to stocks. The other is the Moonshot report, which will assess promising new moonshots.
We’re even going to interview the founders of these projects and make those videos accessible on Patreon.
Notably, the DIY-er plan is a Do-It-Yourself plan because it doesn’t have access to algorithms that signal when to scale in and out of the market.
The Crypto Rider plan uses the 3-Stage Algorithm to do just that. Also, because I’m launching a fund, I’ve adjusted it slightly so that it reports a top 15 coins. The idea is to scale in and out of those coins according to the algorithm. Those top coins are selected because they are both promising and present a relatively diversified portfolio.
Finally, the Bubble Rider plan features 4 separate and logically diversified algorithmic strategies for selecting stocks.
Let’s get into what that logical diversification means.
Two Uncorrelated Strategies
We’re going to do this analysis with stocks because we have much better data for a much longer period of analysis.
First up, then, is the Triple Momentum Strategy I developed a while back to get something uncorrelated with my macro-value portfolio. I describe the strategy in more detail here (on SeekingAlpha).
The blue line at the bottom of the graph is the S&P 500. The red line above that is the performance of the Triple Momentum Strategy itself. It does bounce around a lot, but you’ll see that even with the recent dip in growth stocks it’s still done 19.4% annually.
Also, despite the fact that it’s a pure momentum strategy, it has less of a draw down than the S&P 500 itself. So, it performs better and it’s safer.
Next is the Tech Trader (also included in the Bubble Rider plan). It’s got the dumbest idea (and I love dumb ideas). Basically, it uses the Basic Economic Cycle algorithm to assess whether the market is at least what my subscribers recognize as a Yellow 3 rating (on a 5 point scale). If so, then it holds the QQQ (top 100 stocks from the tech heavy Nasdaq). If not, then it holds gold.
Even with the recent decline in the QQQ, you’ll notice that this has achieved a 20.25% compound annual growth rate (=5% better annually than Warren Buffett). And compared to the benchmark (the blue line) which just is the QQQ itself, it has about half the historical drawdown.
And you’d expect that because the Tech Trader performs better, your portfolio would be better off just owning that one. But you would be wrong. Remember the lesson about the portfolio Holy Grail (Lesson 2): adding uncorrelated assets should boost performance and reduce risk.
Here’s a correlation matrix of these two strategies (in my software, I named the Tech Trader the QQQ Trader – GLD so I could follow my own code more easily).
You’ll see that they have a .53 correlation. So, they are fairly strongly correlated, but different enough that they should do better together.
And if you run them together, this is what you get.
Yes, even though the Triple Momentum Strategy yields about 1% lower returns on an annual basis than the Tech Trader when you combine them the annualized return goes up about .5%.
Even better, the maximum drawdown for both strategies drops. For the quants, that gives better than a 1 Sharpe Ratio, which is outstanding.
That is the magic of portfolio diversification. When you combine two relatively uncorrelated strategies, their returns go up and the risks go down. That happens even by adding a lower-performing strategy.
Now what I’ve shown about these two strategies is that they have statistically uncorrelated returns (at about .53). But they are also logically uncorrelated. The Triple Momentum strategy runs on signals that only trade price action and are completely unaware of any value component to the stocks traded.
The Tech Trader responds to the health of the macroeconomy and just shuffles in and out of the standard tech index based on that reading.
In short, the logic of one is price momentum while that of the other is macroeconomic stability. They are completely different in their rationale, so I can be confident that the lack of correlation between the two will hold up longer term.
If you extend this idea out, you’ll end up hunting for lots of strategies, like I have, and build a portfolio group around that. Even if you look at what I’ve been doing in cryptos, my approach has followed this path.
- Yield Farming = Lending LP, Decentralized Exchange LP, Staking LP
- Moonshots = VC styled selection of projects
- Crypto Bubble Trading = 3 Stage Strategy, which responds to 3 different signals
That’s actually 5 different strategies. And I’m looking to add 2 more such strategies.
This week I wrote a number of pieces that are related to this post, so you might want to have a look if you’ve missed them.
That’s it for this week. Remember to join us on Discord if you haven’t already.
General financial disclaimer: This post is provided for entertainment purposes only. I am not giving you financial advice and I am not a financial advisor. You should expect no financial returns one way or another based on my statements. These points hold equally for any statements that could be attributed to The Art of The Bubble or any related business entities. If you decide to buy or invest in anything, then your returns and potential losses are your own. No statements about taxation are taxable advice and you are encouraged to consult your own tax professional. You are also encouraged to do your own due diligence before investing in anything