Bitcoin Chain Analysis (Data Science) Pt. 1: Increasing # of Users Are Gravitating Toward ‘Actual Use’ of Bitcoin vs. Trading


This article is a bit different than some of the other ones that you may be used to from the brand because we’re going to do something called, ‘chain analysis’ for Bitcoin.

The purpose of this article will be to reveal some of the trends in the on-chain activity for Bitcoin by extracting data from, importing it into Tableau (data visualizer software), then making inferences from it based on the resulting visualization.

Methodology for Bitcoin 

As stipulated before, had to be consulted first.

Specifically, this subdirectory of the domain was consulted in order to retrieve the data:



As you can see, the different metrics that the data provided tracks is:

  • TX Count
  • On-chain transaction volume
  • Value of created coins
  • Price
  • Market Cap
  • Exchange Volume

For an explanation of the columns and the way that the data is recorded, make sure to check out this link here:

On data and certainty – Coin Metrics
_Being certain is a lovely thing. Despite what many would allege about the poor finality of proof of work, the relative…

We won’t delve too deeply into the documentation there since you all can obviously follow the link and read.

Extracting the Data for Bitcoin 

Once the spreadsheet for Bitcoin was downloaded, the data was then extracted from said spreadsheet into a program called Tableau.

If you’re not familiar with them, they allow individuals to create visualizations out of mountains of data.

Specifically, for the purposes of this article, we’re going to take a look at some of the information that we can glean from the report that we cited earlier.

Moving Into Tableau Data Visualizations for Bitcoin 

In the chart above, we’ve extracted three variables, which are:

  1. TX Count
  2. Average Difficulty
  3. Average Fees

Below, we’ll take a look at each one of these categories and see what we can glean from them.

TX Count for Bitcoin 

This category is basic. It’s TX Count. Every single point on the data visualization shown above represents a day’s worth of transactions.

As seen below, Bitcoin was handling about 490k transactions at its max on December 14th, 2017:

When you calculate this per second (there are 86,400 seconds in a day), this amounts to roughly 5.67 transactions/second. It’s worth noting that this data point is extended significantly further (visually) than the rest, so this could even be considered an outlier beyond the normal capacity of the protocol in terms of throughput (up to this point; we have no live examples of Bitcoin reaching comparable throughputs).

What’s most interesting however, is the trend of data following that point.

Obviously, after the massive sell-off in December, there was a significant downturn in the amount of interest for Bitcoin, in general (exemplified by the charts below):


Up to this point, the rate of TX on the protocol mirrored the price action itself.

However, we can see a stark divergence in the total transactions being fulfilled on the protocol after April.

In the chart above, the jump in transactions from March 31st to October 15th, 2018, represents an 80% increase.

However, over that same period of time, the price has decreased by 1.5% to $6.4k at the time of writing.

It’s worth noting that there have been numerous fluctuations in the price since then, however, with volatility lending price jumps as large as 50%+. This was seen in April, specifically.

Analyzing Bitcoin Trading Volume Throughout 2018 for Bitcoin 

One of the most notable narratives in blockchain currently, is the reduction in volume of trading of Bitcoin.

This has largely been used by many pundits in the space to argue that there is waning interest in Bitcoin.

However, this stands in blatant contradiction to the clear uptick in average transactions per day from April to present time that can be seen in our data spreadsheet.

But, for reference sake, let’s take a look at the declining volume in Bitcoin:

  1. First, CoinMarketCap was consulted for their charts.
  2. Then, data was exported from those charts into an Excel document.

Once this data was extracted, it was then imported directly into Tableau in order to create the visualization that you see below:

Specifically, let’s look at the trend in the price data from the top of May down to the end of October.

In the picture above, we detect a localized peak in the exchange traded volume of Bitcoin per day (in USD) at a value of approximately $11 billion.

Per the most recent data offered by CoinMarketCap (at the time of writing), the total exchange traded volume for Bitcoin on October 21st, 2018, (in USD) was approximately $3.2 billion.

Thus, between the two dates (April 25th to October 21st), the volume declined by approximately 70%.

In the picture above, we can see that the average volume throughout 2018 is resting at approximately $6.2 billion.

Thus, that means that the volume is currently trading at approximately half of what the average is.

How Can We Reconcile These Two Seemingly Discordant Facts?

It appears that there are a number of “bullish” conclusions that observers of the protocol can come to:

  1. The growing number of transactions on the protocol indicate that Bitcoin is far from ‘dying’. In fact, if anything, the growing number of transactions on the protocol could be indicative of the fact that individuals are becoming more predisposed toward using Bitcoin for ‘legitimate’ purposes rather than simply speculating on its USD value.
  2. However, the divergence between the trend of exchange traded volume and the overall number of transactions on the protocol could be indicative of extraordinary manipulation suppressing the price of Bitcoin. This can be surmised because, it only stands to reason that a greater number of transactions would lead to a higher demand for Bitcoin. However, this hypothesis can only be confirmed by looking at the total number of wallets on the protocol first.
  3. Overall, 300k transactions appears to be a comfortable threshold for Bitcoin at the time of writing. Anything beyond 300k transactions appears to place a stress on the protocol’s throughput that may result in prohibitive costs for certain actors interacting with the protocol (i.e., heightened fees, slower transactions, etc.), that may cause individuals to pursue less ‘expensive’ (in terms of costs, whether monetary or otherwise) alternatives on the market. The assertion in this point is something that is not yet corroborated with the data displayed within this segment of the Bitcoin chain analysis, but we will be sure to defend this proposition with a wealth of evidence.


Given the breadth of data and information that must be covered on the Bitcoin protocol in order to do a substantive and thorough chain analysis, this will serve as a sufficient ‘part 1’ to our series of examining Bitcoin’s metrics.

In the next part, we will begin delving into mining difficulty on the blockchain.

In that piece we will attempt to:

  1. See if we can curate information about the ‘slope’ (rate of increase) of the mining difficulty.
  2. Run a regression analysis to see if there is a statistically significant correlation between mining difficulty and any other notable variables on the protocol.



Leave a Reply

Your email address will not be published. Required fields are marked *

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