Flipside Crypto Grabs $3.4M to Push Algorithms for Crypto Investing
(Page 2 of 2)
Although market cap makes sense for valuing companies in the stock market because it’s based on a firm’s publicly reported financial earnings and stock price, Balter argues it’s a “flawed” metric for evaluating cryptocurrencies and other digital assets. For one, a cryptocurrency’s market cap can be moved by issuing more digital tokens, he says. Or the value can climb because of buzz, not because of any concrete data, he says. (That can happen in the stock market, too—look at companies such as Long Island Iced Tea, whose stock price jumped after it changed its name to Long Blockchain.)
“I believe market cap means zero” for cryptocurrencies, Balter says. “The number of companies in this space with ‘billion’ behind [their] value is staggering, and it’s complete B.S. ‘X company’ is worth $300 billion. How?”
If the algorithmic approach to cryptocurrency investments eventually wins out, then the question becomes: who has the best lines of code?
Flipside’s algorithms analyze three metrics, Balter says. The first is a “speculation” analysis adapted from hedge fund trading algorithms, he says. Basically, it runs simulations of about 40 different trading strategies to see which ones might perform the best, based on historical and real-time cryptocurrency pricing data.
The second algorithm examines the activity of software developers working on cryptocurrencies. “The philosophy is follow the engineers—where people are building, good things will come,” Balter says. Conversely, when the level of developer activity fades early on, it might signal a doomed digital token—or even a scam, he adds.
The third algorithm tries to gauge the “utility” of the cryptocurrency by tracking transactions executed by the network of computers running the blockchain system underpinning the digital currency. Part of the idea is to identify when a small number of users are executing a large number of transactions, which could indicate a “pump and dump” scheme, where there are “a few people trading between each other in order to make it look like there’s movement,” Balter says.
Some of these data points are starting to prove themselves, but “there is still quite a bit of work to do to create longer-term, sophisticated models,” Balter admits. Still, he says, there’s a “huge opportunity if you can get all of the algorithmic data working together.”