Fred Wilson made reference to an interesting topic a while ago: how to predict what will happen to crypto prices based on their fundamentals. Stock markets have plenty of mechanisms to do this: price-to-earnings (P/E) ratio, dividend yields, etc. Crypto markets, on the other hand, are very much in their infancy and so are the analysis tools for them. People are thinking about it though and attempting to create systems that can predict, to a useful degree, how prices will trend.
The network value to transactions (NVT) ratio is espoused by Chris Burniske, Willy Woo and others and is similar to the P/E ratio; it’s basically total value of currency divided by daily number of transactions. It’s a simple metric that can capture some of the price changes.
It has a catch though: there’s a lag between what it captures and the actual price increases and decreases. Cryptolab Capital’s Dmitry Kalichkin described this lag and an improvement to the metric that uses a moving average of daily transactions. This appears to mitigate the lag and offer better accuracy.
Others have looked at the velocity of transactions i.e. how often transactions occur day to day. More specifically, the rate of change in transaction velocity over a period of time. This is generally known as the velocity thesis, though there are a number of different interpretations and formulae that are under that umbrella. And there are more variants out there.
Are They Accurate?
As I read all these well-reasoned articles and theses, all I think is “what’s the point?” Crypto currencies like Bitcoin and Ether have been around and used for a very short time. There are simply not enough data points to extrapolate any form of pattern. It could be several years, decades even, before there are enough data points to create a robust formula that, approximately, predicts price trends at least for the short term.
This is one of the biggest traps when doing statistical analysis and you see people fall into it regularly, even when they should know better. Some of these formulae may indeed become good indicators of future changes in crypto prices. Or they could be complete flim-flam, not worth even the paper/disk space they’ve been written on.
I can see where these people are coming from. Crypto is cutting edge right now and it’s open to anybody. Imagine creating a formula that could be used for years to come? You could have your name on it. You and your family could be set for life from consulting and speaking about it. Or you might love the pioneering aspect of creating something new for an entirely new system. None of these are bad motivations.
But I also think of something that likely doesn’t spring to mind for most analysts: “this is dangerous”. Dangerous seems excessive, no? A bit over-dramatic? But, I wonder, how many of these analysts would stake actual, significant money on the accuracy of their formula? Because that’s what other people will do, because they too will fall into the same statistical analysis trap. They will see the nice graphs with the lines matching up roughly on the Forbes article and think “it’s a sure thing”.
Now, the standard response is “people know what they’re getting into and they know they are speculating on something volatile”. Indeed, investors and early adopters for crypto do know this and accept the risks. Do other ordinary people know this though? I’ve heard anecdotes of so and so’s mum and dad talking about investing in crypto.
Crypto investment has very few barriers and those barriers are getting easier to overcome as the various blockchains mature. Right now, all you need is an internet connection, an internet browser and some money to invest. Easily doable for most people on the planet. The major barriers are the user unfriendliness of the websites and how they don’t hide blockchain complexity.
Once those are overcome, and they will be, your mum, your dad, your brother or your sister will be able to invest. And when they see those nice graphs, will they truly understand the risks?