An Open Letter to CoinDesk: Please Stop Publishing Price Predictions

4 min readApr 15, 2019

Technical analysis of price charts is not news, and hardly qualifies as analysis. That basic point, understood by most reputable forms of financial media, seems lost on CoinDesk. The popular blockchain and crypto news site loves publishing articles like this, chocked full of sentences such as:

  • A break above last week’s high of $5,347 would invalidate the weekly chart doji candle, although an immediate rally to $6,000 looks unlikely with the daily RSI still flashing overbought conditions.

I hereby respectfully call on them to stop publishing this pseudo-science and all market predictions other than the rare few that can be considered newsworthy because they were made by someone prominent, or managed to move the market.

The first problem with publishing such articles is that they are not credible. As a general rule of thumb, people who know how to trade short term moves successfully (those rare unicorns, made even rarer by a hard winter) don’t tell you their secrets for nothing (or whatever CoinDesk pays its writers). They raise big funds and charge investors hefty fees. Given how hard of a time the prominent crypto hedge funds are having generating alpha, what are the odds some web pundit will get it right?

The second problem is that they are bound to be wrong more often than not. Short term trading itself is a negative sum game (it starts out as zero sum, requiring a loser for every winner, but turns negative since exchanges and the Tax Man always get their cut). If the actual trading is often wrong, how good can the writing about the trading be? Unless of course…

Given how hard it is to predict actual moves, pundits tend to hedge every call by also predicting the opposite. That’s why you get a lot of this “it looks like it’s going to go up, unless of course it goes down” analysis. The article linked to above tells you if Bitcoin falls below $4900 it’s going to $4500, but also says if it goes to $5300 it’s going to $6000. And those are just the opening bullets! Imagine if weather forecasts had the same conviction…”if it rains you’ll need an umbrella, but if it doesn’t you should have brought your sunglasses.”

I don’t mean to pick on anyone, but this wishy washiness is a staple of the creepy market prediction industry (just tune into CNBC around lunchtime or after the close to see what I mean) and infects almost every CoinDesk article predicting price. Being a successful trader is about being right, while being a successful pundit is about being able to pretend like you were right after the fact. That’s why you never see a CoinDesk article that says “Bitcoin is definitely going up” and instead get essays on the doji on Dogecoin being bullish while the RSI on Ripple is bearish. It’s also why no CoinDesk pundit ever says “Here is a link to my synthetic portfolio where I execute every trade so you can measure my ability in real-time.”

But the main reason why CoinDesk should stop with these articles is that they confirm the worst stereotype of the crypto industry being nothing but a bunch of amatuer punters. That point was made today by the sites very own resident expert Michael Casey:

There’s something fundamentally wrong with reducing the measure of bitcoin’s worldwide importance to a price metric that’s denominated in a fiat currency that its advocates hope to replace. It pushes the debate into an inane all-or-nothing binary set of predictions: bitcoin is either going to zero or “to the moon.”

The price of cryptocoins is important for a host of reasons, including security, adoption and attention. I spend a lot of my time as a consultant countering the “blockchain, but not Bitcoin” narrative that is so prevalent in legacy industries, as it entirely misses the point of the important connection between the two. But writing news articles about what the market has done is very different than publishing forecasts on what it will do. One is the stuff of respectful reporting, while the other is best left for aggressively marketed trading newsletters that don’t pretend to be journalism.

So I hereby call on CoinDesk to stop publishing these ridiculous articles. Doing so diminishes their otherwise respectable brand and quality journalism and, as them being an important symbol for the overall industry to outsiders, makes us all look bad. At a certain point, we all have to grow up.