Daily Crypto Thought #4: The Ethereum Mixer
From March 2017 to February 2018, most of Ethereum’s on-chain transaction volume was essentially fake. This should be bigger news.
While browsing the wonderful datasets at coinmetrics.io earlier today, I was shocked to come across a blog post that discussed how they had adjusted Ethereum transaction volume to eliminate transactions that were clearly due to a “mixer” (i.e. transaction washing bot).
The mixer was massive
- 17.7M ETH entered and exited (17% of circulating supply!)
- 393k input addresses, 2.38m output addresses
- 88% of total ETH transaction volume at its peak
- Stopped on February 28, 2018
First reported by cyber.fund last September, it was confirmed by bloxy.info and then Coinmetrics. I was impressed by the simple elegance of the Coinmetrics detection methodology:
First, we parsed the entire Ethereum chain and SQL-ized it (that’s the first step in virtually any blockchain analysis). Then we identified all “one-time addresses” — addresses for which the first ever and last ever transaction was less than one day. Then we built a graph where one-time addresses are vertices, and two vertices have an edge if there is a transaction between two addresses. We found that the largest connected component of the graph had 96.7% of all one-time addresses and 97% of all the volume flowing through one-time addresses.
We then assume that the connected component is the mixer and arrive at figures for the mixer which are virtually identical to the figures found by bloxy and cyber fund.
My Theory
We can have lots of guesses as to why the mixer operated. Perhaps it was used to obfuscate source of funds or wash coins before the US tax deadline.
However, given the start and end dates of the mixer, I have a different, and completely unsubstantiated, theory.
Starting in March 2017, Ethereum begun pumping, and many articles mentioned how it handled more transactions that other cryptocurrencies, implying real demand.
Since that Ethereum was the the blockchain of choice for ICOs, there was certainly some element of truth to this narrative. However, the hockey stick graphs and huge pie chart slices referenced in those articles, in hindsight, were certainly due to the mixer. Given the attention that investors pay to on-chain transaction volume, I believe that the mixer contributed significantly to price action in Ethereum during 2017. I think it was a massive pump-and-dump scheme.
To spend 17% of ETH’s circulating supply and pay the gas costs entailed in running tens of millions of transactions, whoever was behind the pump must have controlled or orchestrated a ton of ETH. It was a highly complex and expensive operation, but it certainly looks like it paid off.
