I've read online and on this forum and elsewhere comments from people who talk about the future of bitcoin whilst at the same time drawing their conclusions from the past. For example, some people have commented along the lines of "Bitcoin was $20k at the end of 2017 so it should go back there soon/ it's currently undervalued/ etc." There are others who talk about technical analysis providing signals to buy or sell (or to "go long" or "go short" in trader parlance (more on this later). And then we have the "fundamentalists" who talk about the economy, decentralisation, the value of the blockchain and so on. I wanted to see if any sense can be made from all this and whether we can identify any trends and the drivers behind them.
I started with looking at some statistics from this forum. Here are the stats from 1st January 2016 to 31st December 2017, a full two year period.
We can see an overall steady increase which then starts to increase exponentially during early 2017 and then drops off in the final month of 2017. This pattern applies fairly uniformly for new members, online, new posts and in fact all measures of activity. It will be easier to see if I plot some graphs, which I have done. Here is the first one.
We can see that aroundabout Dec 2016 new posts being created on bitcointalk started to increase month-on-month. By June/ July 2017 there were around 6,000 new posts being created compared to the 3,000 per month in Nov/ Dec 2016. The activity had doubled. And it continued to increase. This particular measure is a good indicator of
activity within the existing forum community. If we compare this to new members joining the forum we see something interesting, but not altogether surprising. Here is a graph of new memberships:
We can see straight away that the rise we saw with new posts came much later with an increase in new members. It means that existing forum members were already discussing bitcoin more, rather than the rise in posts being attributed to new members. What this boils down to is that new members were joining the party
after the bandwaggon was already some way down the avenue. In fact this increase in memberships only really took off around May 2017 - that's a full six months after the start of the increase in activity of existing members.
To get a better perspective I've taken the trends in bitcoin interest as measured by Google search. Here is a graph for the same two period looked at above:
Notice here that the increase in interest, measured by Google search activity, only occurred around Nov 2017 - that's a further six months down the line still. It means that interest, and by corollary, investment in bitcoin is taking place first within circles of people who are
already engaged with bitcoin with the people at large catching on much later.
The question is, where do price movements kick in? Here is a graph showing bitcoin's price (in USD) for 1st Jan 2016 to 31st Dec 2017:
It bears an uncanny resemblance to the level of interest as measured by Google search. If we look at these two graphs, above, we can see a clear similarity. The more people become "interested" in bitcoin, the more the price rises
and by proportionally the same relative amount. Number of searches for "bitcoin" goes up by 100% on Google then so does the price of bitcoin.
Of course, anyone might argue that all this is just pure coincidence. So we need to show whether or not it is. To do so we need to see whether there is any direct correlation between volatility of bitcoin compared to other assets. The closest comparisons would be other commodities but we should also include stocks, ideally something like the S&P 500 to get a good enough barometer of comparison. Here is a graph plotting the volatility over one year rolling periods since 2011:
It's clear to see that the three do not move in tandem. One final test, just to be sure, is to calculate actual correlations between bitcoin and a large enough sample of other assets. Here is a graph showing just this, correlation with 1,400 ETFs which cover pretty much the investment universe:
Around 99% of correlation occurs within -0.1 and +0.1 which is quite categoric in that bitcoin is uncorrelated with anything else outside of the crypto space. If you're now wondering whether bitcoin is correlated with other cryptos here is a table showing the correlation over the past three months:
It can be seen that bitcoin is indeed highly correlated to other cryptos ( the last four items in the table represent a comparison with other assets as follows: ^SPX = S&P 500 Index, ^VIX = CBOE Volatility Index, ^GLD = SPDR Gold Shares, ^TNX = CBOE 10-year treasury-note yield.)
To sum up, interest in bitcoin, which in turn affects its price, is the key driver. The increase in 2017 was extremely significant as it saw bitcoin's biggest rise since its inception in absolute terms. This was fed by a media frenzy which brought the existence of bitcoin to the awareness of people around the world who had never even heard of bitcoin. For fear of missing out many jumped in, driving the price up. This links in pretty much exactly with the Google search data above. It's worthy to note that the drop in Google searches for "bitcoin" since the third week of December 2017 have coincided with a corresponding and proportional drop in the price of bitcoin.
