Hello Bubble Riders!
This week, we’re continuing with the theme of explaining trading strategies that work during a “crypto winter.” What I wanted to do with this one is how you, step by step, how to execute on the crash cost averaging strategy I explained before.
That strategy outperform Hodl-ing and dollar cost averaging (DCA-ing) by a wide margin in simulations. But how would it have actually performed recently? Would it still make sense to use it right now?
Improving ETH Returns Results
The results of this real-world back-test of those strategies are pretty shocking. DCA-ing did result in 3.5x better returns than HODL-ing while crash cost averaging (CCA-ing) outperformed DCA-ing by better than 6.4x
Paying subscribers will get even better returns than the CCA strategy, but this is really good on its own. And I’m just going to give this away. So, let’s start.
We need to make sure that the comparison that follows is a fair one. To do that, we need to check for the right time scale. I think that trading Ethereum since January 1, 2018, proves sensible. Here’s a breakdown of ETH’s weekly returns since Jan 1 of 2018.
Incidentally, this report was constructed in conjunction with the research team from 1.2 Capital. You’ll notice that for most of that time, the weeks have been negative.
And that makes sense because we’re looking at 2 bear markets (crypto-winters) and just 1 actual bull market (2020-2021). You’ll also notice that there are more weeks with better than 30% increases than weeks with 30% losses. The greatest mass of weeks, however, moves between -3% and 7%. Losing weeks, moreover, tend not to exceed 10%
All those observations imply that we’ve got the right testing environment for a crash cost averaging strategy. This timeframe has lots of declines, but lots of increases too. It’s at least plausible that a dollar-cost averaging strategy would outperform competitors, given all that bumpiness involved (remember, DCA is supposed to be rational because you don’t know the direction of the market).
Next, let’s make sure we’re clear on the conditions for the success or failure of this test. We want to know:
- How much would each approach return?
- Would the returns of CCA-ing beat reasonable benchmarks–the two obvious benchmarks being “hodling” and dollar cost averaging?
To test this out, let’s assume you have a $10,000 portfolio set aside for investing in cryptos. It’s 2018, January 1st and you’re excited to jump on the crypto hype train (also, you’re completely ignorant of the fact that you’re about to buy into the top of a bull run just before a massive decline).
For the Hodl benchmark, we’ll just assume you put all $10k in on January 1st 2018 and forgot about it until August 3rd, 2022.
For the dollar-cost averager, we’ll assume that you bought $20 each day until your money ran out. Since $20 is a small sum, we’ll also assume there are no transaction costs for any of this–prejudicing the results in the favor of the DCA strategy. That should build in some margin of safety for our analysis.
The Crash Cost Averaging Plan
The idea here is to improve over dollar cost averaging through buying at fixed intervals of decline. But we’ll also have to buy more at the end of a decline, should we get a chance.
This will require us to make a plan for our buys. If everything bounces back before we deploy all our invested capital, that’s fine as we’ll have “gotten a bargain” on those earlier buys. At that point, we’ll just trade back in as normal.
To make things super easy, let’s divide our investment strategy in the following way.
- At a 60% decline – Buy 10%
- At a 66% decline – Buy 10%
- At a 72% decline – Buy 20%
- At a 78% decline – Buy 20%
- At an 84% decline – Buy 20%
- At a 90% decline – Buy 20%
This will put most of the buying into those maximal decline points. It’ll get us the best bargain. And, notably, it doesn’t buy past a 90% decline, since that might be a sign that the coin we’re trading just isn’t going to make it.
We can control for that “total collapse” scenario using only more established coins. That’s why for this approach analysis, we only used ETH.
We’ve still got a problem though. When do you sell?
Traders are going to tell you to pick your exit points before you sell, but we’re trading super volatile assets. More importantly, they’re bubbly, meaning that the top is irrational. And to recall what is literally Lesson 1 from AOTB, if you get out too early in a bubble, you lose most of your gains.
So, let’s use a simple momentum strategy to help us with this. There are literally more than 300 peer reviewed articles supporting the out-performance of momentum trading, and these strategies work even better with volatile assets. That’s perfect for us, so let’s explore that point.
A Momentum Strategy
I should note that Crypto Riders & Bubble Riders, who pay for our AOTB premium plans, this part is taken care of by our proprietary algorithms. They perform significantly better than the strategy that I’m going to describe, but it’ll be helpful to go through this one to know what’s basically at play.
To be clear, I’m giving the following strategy (an algorithm) away. It’s good enough for smaller investors, and it’ll explain our principal points (though remember, no expected returns implied by anything, DYOR).
Here’s the big counter-intuitive point. The algorithm should look at industry momentum rather than individual coin momentum. Here’s why. Have a look at this fake ETH rally in the middle of the 2018 crash.
You would have invested at exactly the wrong time if you were following a basic momentum strategy on ETH. Now, part of a momentum strategy is that sometimes signals are wrong (that’s part of the process), but no one likes losing unnecessary money.
That’s why the strategy doesn’t trade momentum on individual coins. It looks at lead indicators instead. For the crypto world, that leading indicator is Bitcoin (or the Total on TradingView, if it’s 2021 or later). Here’s that same rally for BTC.
Obviously, it’s not nearly as large here, which is to be expected. “Alt-coins” always bounce more than the lead indicators, which is why looking at the lead indicator reduces “noise” (false signals) in the strategy.
With that in mind, here’s the free strategy, which is a basic algorithm. Let’s start with some definitions.
- Portfolio = ETH
- Lead Indicator = BTC if before 2021, the TOTAL starting Jan 1 of 2021.
- Lead Inicator is abbreviated as “LI”
- MA = moving average as calculated on TradingView. We’ll be using the 160 day time scale. This has a “smoothing” modifier and it is set to 5.
- LSMA = least squares moving average as calculated on TradingView. It has an offset, which is set to 0. The algo uses the 160 day time scale.
The algo’s Portfolio could have more than ETH, but then it would need a way to choose among those coins. Again, premium members get exactly that, but this strategy is a free algorithm.
- Using closing prices on the lead indicator (LI).
- If LI closes above the LSMA 160 for 3 days, then Portfolio should enter 50% in the market.
- If LI closes above the MA 160 for 3 days, then Portfolio should enter 50% in the market.
What this means is that the algo will buy ETH in two chunks, given BTC’s performance (because BTC is the lead indicator).
- Using closing prices on the lead indicator (LI)
- If LI closes below the LSMA 160 for 3 days, then Portfolio should exit 50% from the market.
- If LI closes below the MA 160 for 3 days, then the Portfolio should exit 50% from the market.
That’s the free algorithm. Basically, the trend is your friend, whether that trend is going up or down. This is a slight adjustment on the strategies discussed in Lessons 1 to 9 from The Art of The Bubble guide.
Combining with Crash Cost Averaging
There are a few complications though. We’re not just following that momentum strategy. We’re using it to know when to sell out of our Crash Cost Averaging strategy. And that means that we’re not exactly entering the market with a clean slate.
Also, if we meet with a false rally, meaning that the market does not make a new all-time high, we’ll need to revert back to our crash cost averaging strategy.
The entry conditions don’t really need much modification. If we already have the portfolio 100% invested, we’ll just keep hodling. That was the plan all along. But we do need to modify the exit conditions a bit, since we might buy ETH in what turns out to be a false rally. Let’s spell those out then.
Adjusted Exit Conditions:
- Mark portfolio holding % upon satisfying an entry condition (e.g. was already 40% invested or whatever).
- If false rally, then upon exit condition revert to marked portfolio holding %.
Basically, we’ll have the strategy exit to whatever the crast cost averaging strategy was holding before the false rally happened. And it’ll continue on its way.
I think the results speak for themselves. $10,000 invested at the beginning of 2018 would have resulted in the following returns for Crash Cost Averaging (CCA), Dollar Cost Averaging (DCA), and HODL-ing, respectively.
HODL – $21, 498
DCA – $72,117 (= 3.35x HODL)
CCA – $463,363 (= 6.42x DCA)
So, the actual results are much better than the simulation I wrote about previously. The reason is that we’ve gone through 2 crashes and this strategy includes a momentum algorithm not present in the first simulation.
To visualize the differences, let’s start with a logarithmically scaled chart, which highlights early performance returns. You’ll see, actually, that DCA (in orange) outperforms CCA (in blue) for a bit.
The reason is obvious, the DCA strategy holds a lot of money in reserve for a long time. CCA does not. As a result, DCA (by virtue of this testing period) is able to consistently pour money in for a longer period than CCA. As long as you have consistent money to pour in, while CCA does not, DCA will outperform.
Of course, that means that you’ll need more money than you had for your CCA strategy, making it an unfair comparison, but still. That’s what the graph shows.
If you want an apples-to-apples comparison, then you’ll want to look at the non-logarithmically scaled chart of returns (using the same colors as before).
In this chart you can see why the CCA strategy works so well. The momentum algorithm shuts off the losses during a decline, and then starts buying back in when those declines turn massive.
DCA, by contrast, is stuck riding up and down whatever direction the market goes. As a result, it prove both more volatile and less intelligent.
Alright, I know this was a long one. It also completely detailed a strategy that outperforms DCA-ing by better than 6.4x. Premium subscribers, again, will get better performing signals than even this account outlines, but there is real value here.
Incidentally, the next ETH by target is $770.
The AOTB team will be announcing buys and sells on Twitter (and Discord) so follow if you want really fast updates on this (@lspurcell). In the next one of this series, I’ll look at combining this strategy with crypto-maximalism.
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.
This newsletter is provided for educational and entertainment purposes only. Robin Technologies and Analytics LLC is the firm that distributes The Art of The Bubble products. The firm does not provide individually tailored investment advice and does not take a subscriber’s or anyone’s personal circumstances into consideration when discussing investments; nor is Robin Technologies and Analytics LLC registered as an investment adviser or broker-dealer in any jurisdiction.
You should expect no financial returns one way or another based on statements contained herein. These points hold equally for any statements that could be attributed to The Art of The Bubble or any related business entities or personnel operating in association with Robin Technologies and Analytics LLC. 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