This week I reached a fork in the road. I am putting aside the idea of actively trading and instead will be investigating a longer-term investment in cryptocurrencies.
Welcome to the crypto-investor’s diary, week 6.
Why you no trade no more?
Story time: as a youngster, I decided that I wanted to work out the pattern in prime numbers. To do this, my first step (whether I realised it or not) was to assume I would succeed where a lot of very smart people had failed.
Long story short, I was unable to discover a pattern in prime numbers.
To believe that, through trading, I can reliably return anything even approaching, say, 100% year on year is similar to believing I can find a pattern in prime numbers. Warren Buffet, one of the richest old white males, has only managed 20.8% pa over his lifetime. The amazing Medallion fund averages a pretty lame 70%.
So I have slowly but surely come to the conclusion that literally everyone else had already come to: I’m going to chuck some money at crypto based on the properties of individual cryptoassets.
I’m not sure how I’m going to approach this, but I’m thinking I’ll have two personal funds. I will call them the Turtle Fund and the Hare Fund.
For now, I’m focusing on the boring, long-term stuff; the Turtle Fund.
For this, I’m not going to go after the tiny altcoins in order to jump on the initial 10,000x hype train. I’m going to start by simply selecting a handful of coins that I think are likely to see a solid, consistent increase in value over the next few years.
More specifically, each month I’ll buy $2,000 in one cryptoasset. First up bitcoin, then next month Ethereum, then Ripple, and so on. I will aim to hold each asset for as long as possible.
The tax benefit of holding vs switching is surprisingly big. I’ve looked at the numbers for three scenarios, each imagining I had $10,000 and two cryptoassets I was interested in. The scenarios:
- I put $10,000 into one asset but switch back and forth often (let’s say every 6 months I choose whichever is performing better)
- I put $10,000 into the better asset and switch it every 12 months (for lower tax)
- I just put $5,000 into each asset and leave it in, no switching
For this example, I’m imagining that despite my best efforts, the returns were the same in all cases (200% pa).
In the first example, I’m paying maximum tax on my profits at the end of each year. So of course less money rolls over and I get less compounding. Let’s get sloppy and call the result half a million bucks.
If I hold each position for at least 12 months, I will pay half the tax (this is Australian tax rules), but I will still be taking money out each year to hand over to the tax man. I’ll call the end result one million bucks.
If I don’t sell/switch assets, then the tax man is none the wiser, until right at the end where I owe a truck load in tax. When the time comes I will bitch and moan like the world was ending, but it will have been the right move. Final result: two million bucks.
Those are massive differences! And the only difference is switching between assets vs not switching.
In reality, I might think I can get better returns by switching. In the below, the end result is roughly $500,000. But the returns (second column) are now different for the three scenarios.
I would need a return of 216% to get to $500,000 if I switched assets frequently, or only 127% if I sat tight and didn’t switch at all.
All this tells me that switching assets frequently (e.g. to continually track the top 10) must result in significantly higher returns to be worth it. And that tax should be thought of as a key consideration in my plans, not merely a pesky eventuality.
I imagine getting this balance right (risk, returns and tax) will be wonderfully complex. I see a book in my future: Modern Portfolio Theory for Risky Speculations.
OK so that’s my general thoughts so far on a strategy. As with everything, I’ll look back in two weeks and think that was dumb. Now…
How do altcoin returns compare to bitcoin?
A question that’s been lingering for me is: if I threw $10,000 at bitcoin or $1,000 at 10 altcoins, which would cross my palm with more silver?
As it turns out, masterthecrypto.com already did this in their Bitcoin vs Alt Coins Returns article. Respectfully, this is a dodgy article that appears to compare the increase of bitcoin from June 2016 to November 2017, with the top 10 altcoins (by market cap in October 2016) from October 2016 to November 2017.
Inconsistent date ranges aside, they also happened to be writing that article at the height of a bubble (through no fault of their own, I assume). Since the altcoins seem to generally be a multiplier of bitcoin, any difference taken at the top of a bubble is probably not indicative of the longer term difference in returns (since the altcoins are also like multipliers of bitcoin in a downtrend, but there was no downtrend in the date range used in the article).
As luck would have it, we’re currently in a slump after a bubble, meaning the past 12 months have a better balance of uptrend and downtrend, and are perhaps more indicative of what can be expected in the long term.
Ah, the blue line of hindsight.
These three charts are on different y-axis scales. The Ripple peak is four times the height of the Bitcoin peak.
So it’s not a completely crazy point in time to be measuring a one-year return (although I’m aware that in the grand scheme of things, calculating a one-year return for an altcoin is a crazy thing to do).
(I’m also aware that investing in different assets with such a high correlation is not good portfolio management. I want to show off that I just learned the term ‘positive covariance’, but can’t work it into a sentence.)
Less talk, more chart!
Here’s the one-year returns for the top 10 cryptoassets (by market cap in May 2017).
I quite like this chart. The differences are modest, and some of these coins are getting quite old now (Ripple and Litecoin are six years old, Monero is four and Ethereum is almost three. They grow up so fast!)
And even the decade-old bitcoin is in the top five for growth (in the context of the top 10 coins by market cap, so maybe that’s meaningless).
What if I’d taken the top 10 coins a year earlier, in May 2016 (from this coinmarketcap.com snapshot) and held on for two years?
That’s the return per annum (not the increase over two full years). If you’re interested, I’ve taken the price increase over the two years and used
SQRT(1 + twoYearReturns) — 1 to get the single year return, taking into account compounding. I only share this level of detail because it took me ages to work it out.
So NEM averaged 1,545% pa over those two years, which means it’s price actually increased by 27,000% (not a typo).
And I’d never heard of it until this week.
My takeaway from this is that it’s surprisingly hard to go wrong picking from the top 10. The worst you could have done is only a return of 165%, and that would be your own damn fault for picking a coin with a stupid name like MaidSafeCoin.
Now, this is the top 10. But why 10? Maybe playing with the top 20 is better. Perhaps it’s best to stick to the top 5. There will be more charts in future weeks, I’m sure of it.
Although I say it’s ‘hard to go wrong’, when it comes to picking something that’ll serve me well for the next five years, some level of intelligent comparison of cryptoassets is probably a worthwhile endeavour.
So, how does one assess the potential of a cryptoassset to increase in value?
And more to the point, how can I make high quality decisions now with my low quality knowledge?
And also another thing, I think assessing the potential for long-term growth is a different beast to short-term growth. Everything below is about long-term returns for my Turtle Fund.
A unified set of metrics (hmmmm)
I started down the track of coming up with some unified criteria by which to assess altcoins (GitHub activity, social following, etc). This is done quite thoroughly in the article A VC’s take on evaluating cryptocurrencies.
On the surface, there’s something appealing about this idea. I guess because if it works it would make it easy to pick good earners. In other words, easy money, which is traditionally quite and appealing thing.
But it leaves me with questions that I need to investigate:
- Does it work? Do more GitHub commits predict future growth? Does more noise on Reddit convert to real world gains? If these results are “best used in conjunction with other factors” how much precision do they add to those other factors?
- Does it outperform something as simple as just picking the top 10 assets by market cap? By how much?
- Does it accurately rule out frauds? Ponzi schemes tend to have lots of positive press, because bots are cheap. Would that, combined with an active GitHub page get an A+ rating?
- Does it work better for shorter time frames or long?
- And the big one, does it even makes sense to seek out metrics that can be applied across all cryptoassets?
In the world of stocks/shares, all listed companies share the same goal, and therefore have the same measure of success: make money. So you can quantify this and come up with some sort of framework for assessing all stocks by the same criteria.
But you wouldn’t try and apply the same quantitative analysis to wheat or silver or the Eueo/Venezuelan bolívar fuerte. Right?
So are cryptoassets like commodities? Or stocks? Or fiat currency pairs? I think they’re a bit different to all these things, and to each other.
I think they’re more like handymen, or handywomen.
If I want someone to clean my gutters, or prune my hedges, or renovate my kitchen, each of these jobs would have a different measure of success. A different set of criteria that must be met before I say “here, take my money”.
Perhaps for Ripple, being used by banks to transfer money is the primary sign of success. For Ethereum, it might be to have a certain number of DApps. For bitcoin, it might be for hedge funds to include it in their portfolios.
(I don’t think there’s any reason for a measure of success to align with the initial vision of a project. I don’t know though, I’m making all this up as I go.)
This handyperson analogy is pretty crap, but it suits my needs because what I really want to do is steer clear of assets that haven’t demonstrated success, based on the assumption that these are the ones with the greatest potential to fail quickly.
For example, I don’t know anything about EOS, but I get the impression it’s like a bathroom renovator with a good reputation showing me some awesome 3D renderings of the plans they have for my new bathroom. They look great — I’ve always wanted one of those porous sinks where the water just disappears — and people tell me that the bathroom dude is a genius. And if I pay for the bathroom now, it’ll be cheap, but if I wait till later I might have to pay a lot more.
That’s all great, and I’m not doubting anyone, but I’ll wait to see how well the plans are transformed into reality before adding EOS to my Turtle Fund.
I might do a little heading here for something else I’ve been thinking about…
I’d love to do a survey of everyone who has lost money in cryptoassets and discover what proportion of those losses were the result of Fear Of Missing Out.
I’m guessing lots.
And I know I’m susceptible to FOMO because I was reading the Wikipedia page for ‘Ponzi scheme’ yesterday and found myself nodding, thinking “this sounds pretty good”.
So, the way I plan to fight this expensive enemy within is to think what it will be like to brag that I “got into Cryptocurrency X in it’s first year of operation”.
I’m seriously not trying to pick on EOS, but its capitalisation/utilisation ratio (
ERROR: cannot divide by zero) makes it a good example.
I’m not ashamed to say that I have some EOS FOMO, I can feel it burning within — I don’t want to miss this train. But I just need to imagine myself, ten years from now, bragging to passers by as I sit in the park feeding pigeons “I remember investing in EOS back in 2019 when it had only been operating for 12 months. That’s how I can afford all these pigeons”.
Sentiment analysis (hmmmm)
I’ve been thinking about sentiment analysis and whether this could be used to gauge the popularity of an altcoin. And the answer is yes, it can, but what predictive power does popularity have?
Pokemon Go was popular. Fidget spinners were popular. I used to be popular. But these things fade.
Oh, do they fade.
By their very nature, Ponzi schemes will have pretty good market sentiment from all the people that have already bought in.
So I wonder, what’s more important, 1,000 excited tweets extolling the game-changing virtues of #ripple, or one big bank releasing a product that uses Ripple?
This brought me to think about…
As I’m sure you already know, the thing that set Google apart from its competitors in the early days was the way that they ranked their search results: by looking at which other sites linked to a page.
Maybe what I want to measure is who else is showing interest in a given asset (I’m quite sure I’m reinventing the wheel here, but that’s how I like to learn).
Goldman Sachs is opening a ‘trading operation’ (whatever that is) for bitcoin.
Microsoft picked Ethereum for its blockchain as a service a while back, and I’m guessing they had some intelligent, level-headed folk assess the long term viability of the technology before jumping in.
Google has shown signs of interest in Ripple. Both in Ripple Labs the company via some early investment, and in the technology. Specifically, the Interledger Protocol, which landed in Chromium for Android in this commit back in 2017.
Again, I would expect that Google had some smart people assess the situation before jumping in.
Side story: I was trying to work out exactly what the phrase ‘Google-backed’ meant when used before the word ‘Ripple’. It turns out it means that Ripple Labs is one of the 323 companies Google Ventures have invested in.
Side story fun fact: Google Ventures have also invested in Ripple Foods, which makes milk-free milk products, including a Greek Yogurt Alternative. (In my opinion, the best Greek yogurt alternative is not eating Greek yogurt.)
I have a lot of work to do here, tying together “what does success look like” for a cryptoasset and then assessing the significance of the organisations who are involved in reaching that success.
Social networks as a source of information
I’m torn when it comes to Reddit and Twitter as places to go in order to increase my level of knowledge. On one hand, they’re communities of passionate people sharing interesting articles and diverse thoughts on all things related to cryptocurrencies.
On the other hand, they’re Reddit and Twitter.
With my beginner’s level of knowledge, I feel ill-equipped to distinguish between a hype-machine with a bot behind the wheel, honest repetition of incorrect information, and genuine insight from a knowledgeable and unbiased observer.
If I have no way to decipher what’s real from what’s fake, this does not bode well for making intelligent investment decisions.
So I wonder (and I haven’t decided yet), am I actually better off if I decide not to delve into the Reddit comments and Telegram groups and Slack channels?
It’s a tough one, I don’t want to miss out on info, but I want to have a certain reliability-threshold for info.
Perhaps when I start working on the Hare Fund I’ll get down and wet with these fire hose sources of information.
The press as a source of information
I’ve been searching for various assets all week and have seen plenty of articles pop up in the ‘news’ section of Googles results.
In a search for ‘Ethereum’, the following was the very first news result:
I saw this and thought three things:
- The alignment and typographic hierarchy needs some work
- Only 6%? Surely that’s a pretty tame day for Ethereum
- Saying “Ethereum falls 6 percent after event X” is a bit like saying “The Dow Jones crumbled after I put my pants on this morning” — it might be technically true, but it implies cause-and-effect when there is no evidence to suggest the two are linked
I went to look up the price chart to see this ‘6 percent fall’ with my own two eyes:
Firstly, that is not really worthy of a news article. Secondly, you might expect that to be a drop of 6%.
[I’ve just deleted a big long rant about this, because we live in a world where pointing out fake news isn’t even interesting.]
Long story short, nothing is true.
The next big question I want to answer is, how do I not go on this particular ride:
That’s a market cap of 2.7 BILLION FRIKKEN DOLLARS — and it just vanished.
Meanwhile, half of Africa is like, SMH.
I think everything I’ve said above, if done right, protects me from this sort of crash. Right?
There are already people doing ratings of cryptocurrencies, for example the Weiss Cryptocurrency Ratings.
Could I just follow them blindly? They undoubtedly know more than me, but at the same time have no track record (with cryptoassets) and don’t necessarily have the same risk profile as me.
I’m not sure I’m ready to hand over my $468 for a one year membership to access their 74 coin evaluations, but I’m tempted…
Also, there are index funds that track the top x cryptoassets. If I don’t think I’m better at predicting the future value than a fund that just picks the top 10 every month, then I might as well go with them (depending on fees and tax implications).
So, what next?
I’m going to make a list of each of the top altcoins, and for each one, write down what I think success looks like. Then for each, rank how well they’re doing. Maybe there’s a few coins in the top 20 that are all hype, and a few coins in the top 500 that already have some years under their belt and runs on the board and perhaps even feathers in their caps.
Maybe for each asset I’ll think about its ‘potential to fail quickly’. For example, Ethereum might fall away over several years if other cryptoassets show up that do the same things better, but that’s fine — I would expect that to be a slow decline and I’ll have time to move my money out.
But if EOS fails before it goes live, I imagine it has the potential to go down like a sack of potatoes. (I promise, that’s the last time I’ll pick on EOS in this post).
End of post.
Week 7 is here!