Bitcoin at $100,000? For any prediction, ask “what’s your methodology?”

Chart: Coindesk

CNBC reported that, according to Saxo Bank analyst Kay Van-Peterson, the price of bitcoin “could hit $100,000” in 2018. This is a good time to ask, “Where do numbers like this come from?”

Broadly speaking, analysts use detailed analysis and models to identify what is happening and predict what will happen. But what is the source of the analysis? What are the inputs to the model? And how does the model make its predictions? These are the questions you should ask of any analyst prediction that interests you.

While analyst models are typically proprietary, analysts will generally share the methodology behind the model, so you can at least get an idea of how they work.

For example, when I was an analyst, I predicted the spread of DVRs, the fortunes of cable and satellite, and shifts in the advertising industry. At Forrester, we started from consumer data about preferences and behavior. We examined historical trends. Then we made five-year predictions.

You might be surprised to learn that most basic analyst models do not include math more complex than multiplication, division, and percentages. (Economists’ models tend to be far more complex.)

The most important question is what the assumptions are. In a well-formed model, there are no more than a few assumptions, and everything runs off of those. When you have too many assumptions, you’re basically guessing.

Another important question is how those assumptions drive the results. In an effective model, the analyst will be able to tell you something like “We assume shifts in advertising prices based on historical data. We also assume share of different ad formats based on historical trends and shifts we’ve learned of from discussions with vendors and advertisers. We then model the number of avails of different types, and then combine those data to assess where the advertising market will go.”

How did Van-Peterson model bitcoin’s price?

As far as I can tell, no one knows. But there are clues in these quotes from the CNBC article:

“First off, you could argue we have had a proper correction in bitcoin, it has had a 50 percent pull back at one point, which is healthy. But we have still not seen the full effect of the futures contracts,” Van-Petersen said.

The CME and Cboe both launched bitcoin futures trading contracts last year. The move was seen as a way to get more institutional investors involved in the cryptocurrency market and legitimize it. But trading got off to a light start. Van-Petersen said that more institutions will get on board over time, but it won’t happen quickly. . . .

Previously, Van-Petersen said that $100,000 could take ten years to hit.

Also, bitcoin has been trading sideways since the start of the year. But the Saxo Bank analyst said that bitcoin tends to trade around a certain level then “re-rates”.

“I wouldn’t be surprised if it’s something we are seeing. It’s kind of building a foundation, then will re-rate a bit higher.” . . .

Van-Petersen, who owns different cryptocurrencies and is a noted bull in the space, joins a chorus of other figures also predicting big price rises for bitcoin.

Stock analysts evaluate price moves based on where the market is going overall, a company’s profit outlook, and discontinuous events like mergers and acquisitions. Because the stock market has been going on for so long, there is ample data to determine how these factors will change the value of a company’s stock. As in any prediction, there is uncertainty, but if you trust the analyst’s valuation calculations, the uncertainty comes directly from those predictions of the profit outlook or chances for a merger. This is because stocks have an intrinsic value that comes from their assets and profit-making business activities.

Bond analysts similarly evaluate macroeconomic trends, likely moves by central banks like the Fed, the level of risk involved in the assets that underpin the bonds, and so on. Again, there is an intrinsic value in these assets.

Commodities analysts make predictions based on supply and demand shifts for commodities like gold or oil.

Examining Van-Peterson’s quotes, his comments refer to short-term shifts based on the availability of bitcoin futures contracts and technical analysis of previous ups and downs. Short-term price predictions are risky and uncertain, because they respond to market events that no one can foresee (for example, a break in the security of blockchain, or the possibility of nuclear war with North Korea).

Statements that identify Van-Peterson as “a noted bull in the space” also worry me. A true analyst predicts not based on what he wants to happen, but on what is most likely to happen.

But the most troubling part of this prediction is that there is no discussion whatsoever of the intrinsic value of bitcoin. For an asset to have a long-term value, there must be some idea of where that value comes from. There is no coherent argument for the intrinsic value of bitcoin. As former Fed Chairman Alan Greenspan said, “The question is I do not understand where the backing of bitcoin is coming from. . . . You have to really stretch your imagination to infer what the intrinsic value of bitcoin is. I haven’t been able to do it.”

Never trust a prediction until you know the methodology

Van-Peterson’s prediction is worthless. His only credential is that he got the price right last year. If he is right, there will be no way to know why he is right. If he is wrong, we will not have learned anything about his method. This is not analysis, it is fortune-telling.

Never trust a prediction until you know the assumptions and how the model works. Even then, assess these predictions for the way they enlighten you, not because they accurately foretell what will happen.

And don’t buy bitcoin unless you have money that you can afford to lose.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.