Research: Bitcoin way overvalued, expect months of sideways prices
The bad news is that bitcoin is still in a bubble. The good news is that it's a fairly stable bubble.
Researchers from ETH Zurich have found a novel way of predicting bitcoin prices, in a paper published 16 March. It suggests that bitcoin's in for a rough few months of sideways price movement.
"Looking forward, our analysis identifies a substantial but not unprecedented overvaluation in the price of bitcoin, suggesting many months of volatile sideways bitcoin prices ahead," the researchers wrote.
The analysis concludes that bitcoin is currently overvalued about four-fold, and that it's probably not on track for any immediate sharp growth or sharp plummets in the coming months.
To decide whether or not you want to believe this, it's worth looking at how they reached this conclusion.
How to predict bursting bubbles
How much is a bitcoin worth? A lot of people have spent a lot of time trying to answer this question.
One of the keys to many analyses is Metcalfe's Law. This was devised in the 80s as a way of valuing networks, and proposes that the value of a network is based on the number of participants in it. In the case of bitcoin, that means the value of the coin is proportional to the number of miners working it.
Even though this law was devised decades before bitcoin was invented, analysts have found that it can predict about 94% of bitcoin price movements. As such, one popular method of valuing bitcoin and trying to predict the market is to start with Metcalfe's Law and then add other metrics to the mix. With these new metrics, the formula is re-tested against past price movements to see whether it's more or less accurate than Metcalfe's Law alone.
In this way, the researchers have found a compelling new formula that combines Metcalfe's Law with the "Log-Periodic Power Law Singularity" model (LPPLS). The LPPLS is an effort to objectively track market bubbles and their broad likelihood of bursting at a given time. It does this by tracking the rate of super-exponential price growth (where the growth itself is also growing), and measurable positive feedback phenomena like herding and imitation, or hype and FOMO.
This model says that bitcoin is still in a bubble and still way overvalued (by about four-fold), but also that the current bubble won't pop or kick into bullrun any time soon. The end conclusion is months of sideways prices, perhaps trending downwards.
The final straw.
The paper starts with the mathematical assumption that bitcoin is a bubble because Metcalfe's Law compellingly suggests that it's extremely overvalued. But it also points out that bubbles, where the price of something significantly exceeds its measurable value, are a normal fact of markets.
The idea is that bubbles naturally emerge and grow in markets, but become more fragile as they inflate. If it's a fragile super-bubble then any old piece of bad news might trigger a collapse and sell-off. If it's a small durable bubble then its prices can shrug off the bad news.
Bad news doesn't cause the crash, the researchers argue. It just happens to be the final straw that breaks the camel's back.
"Market participants often lament that crashes are unforeseeable due to the unpredictability of bad news," the researchers argue. "However, focusing on the news that may have triggered the crash is akin to waiting for 'the final straw', rather than monitoring the developing unsustainable load on the poor camel's back."
The good news, researchers say, is that you don't have to predict bad news. You just need to know how fragile the bubble is at any given time.
Is this model accurate?
This model assumes that Metcalfe's Law and the LPPLS are both accurate. Metcalfe's law is needed to determine how much bitcoin is worth, and therefore the extent of its deviation from fair prices for the purpose of applying the LPPLS.
Metcalfe's Law has tended to be a strong mirror of bitcoin prices over the years though, and is widely used as a baseline in various models.
Consider how the number of active bitcoin addresses has mirrored bitcoin's market cap over the years.
On one hand, it's eerily accurate for a very basic law that was devised long before the first bitcoin was ever created. On the other hand, it's based on the number of active addresses according to bitinfocharts.com, which as the researchers note, is an extremely flawed way of determining how many users there actually are.
The coloured points indicate the three most recent major crashes, which occurred at times when the LPPLS model showed a very fragile bubble.
In the case of the most recent February 2018 crash, the researchers suggest, it might have just gotten so fragile that it collapsed under its own weight, essentially breaking under the good news of hitting $20,000.
The prognosis is relatively pessimistic according to this model. The bad news is that bitcoin is still a bubble and might not be tracking for a bullrun any time soon. The good news is that the bubble doesn't seem set to abruptly burst and send prices free-falling.
If you want a more optimistic take on bitcoin's future then contrarian models like the Bitcoin Misery Index can also be convincing. But if you want the only guaranteed accurate prediction, it might be that bitcoin prices will continue to go up and continue to go down.
Disclosure: At the time of writing the author holds ETH, IOTA, ICX, VEN, XLM, BTC, NANO
- Ethereum price breaks all time high of US$1,448
- Ethereum price: Massive slide as market faces bearish pressure
- Ethereum 2.0: Roadmap, timeline and implications
- Bitcoin falls to US$34,000 as confidence in money markets improves with the Biden inauguration
- Bitcoin price lags while regulators raise fears and banks grapple