You would think that the most active blockchain networks might also have the most in-demand tokens. But is that really true? Just how closely do token prices track with activity on their networks?
To find out, LongHash analyzed daily price data and address data from Coin Metrics for 18 top tokens (excluding stablecoins). Specifically, we looked at the Coin Metrics daily price for each token, which tracks the closing price (i.e., the price at midnight UTC) for each asset in USD, and Coin Metrics’s Active Addresses Count. This metric represents the sum of unique addresses that participated in at least one transaction on the network that day, either as a sender or a recipient (but only counts each individual address once per day).
For each asset, we then calculated the Pearson correlation coefficient for these two variables, price and active addresses, over the entirety of Coin Metrics’s available data. Because different tokens began to be tracked at different times, some tokens have more data than others, but the correlation coefficients of all but one token (BSV) are based on at least a full year of data, and most are based on significantly more.
Pearson correlation coefficients are represented as numbers between -1 and 1, with -1 representing a perfect negative correlation (i.e., when one value goes up, the other goes down), and 1 representing a perfect positive correlation (i.e. when one value goes up, so does the other). Values of above 0.3 or below -0.3 are typically considered to indicate some level of correlation, with values above 0.7 or below -0.7 indicating strong correlation.
As we can see from the chart above, some tokens seem to show a strong correlation between price movements and active addresses. Bitcoin, Litecoin, Link, and NEO all fit this bill, suggesting that when more users are transacting on these networks, the price generally goes up.
While no tokens show a significant negative correlation, quite a few tokens show no measurable level of correlation between price and active addresses. BSV, XLM, TRON, TEZOS, ETC, MAKER, XEM, and BAT all fall into this category. More unique user activity on these networks doesn’t seem to correlate with price rises, at least according to Coin Metrics’ data.
It’s worth noting that ETH very nearly falls into this category as well, showing at best a weak correlation between unique active users and price. This may be due to the token’s position as the “oil” that fuels some dApps on its network, which could mean a lower proportion of its users are regularly engaged in the types of trading transactions that most directly fuel price movements. For example, sending ETH to a dApp would cause an account to be included in that day’s ETH unique active user count by Coin Metrics, but it would not affect the ETH price on exchanges in the same way a TUSD for BTC trade would affect the BTC price.
While unique active users probably wouldn’t be an effective metric to predict prices for day trading, it looks like over a longer term, some tokens like LTC really have shown a strong correlative connection between their token price and the number of active users transacting on the network in a given day.
More broadly, though, the data suggests that network activity and token price aren’t always closely related in the short-term. That means that a spike in user activity isn’t necessarily going to lead to a spike in price, particularly if it’s a spike in users of a token like Maker or Tron that has historically shown little correlation between active users and token price.