The Great Bitcoin Power Curve, part 2.
Back in March 2018 I wrote an article about Bitcoin’s price history (https://medium.com/@phil.monk_69275/the-great-bitcoin-power-curve-31e584e00a4c). The gist of the article was that Bitcoin’s price has tended to follow a power curve for almost all of its history as a trade-able commodity from 2011 to 2018. The TL;DR picture of the article is this:
In summary I suggested that Bitcoin was good buying when the price was below the curve (shown in cyan). The data points that I chose for the curve were:
Point A: 8 June 2011 0740 UTC $31.9099 (this was a Mt Gox price)
Point B: 8 January 2017 0755 UTC $881.3
Point A represented a blow-off top of an early bull market. Point B was chosen arbitrarily as a point defining a line where the price had bounced around in the days preceding, and had interfered with the price consistently over the years.
Seeing as it’s been 9 months since that article I thought it would be worth revisiting and seeing how it’s stood up, and whether it merits re-evaluation.
Original prediction
I predicted then when the price was about $7,000 that Bitcoin’s price would decline until it at least touched the power curve. As of time of writing, that has not happened yet. The closest approach so far was on 15 December when the price dipped to $3,122.28, and the power curve price was about $2,778.
It did occur to me after writing the article that Point B was arbitrarily chosen. There could be other power curves that run fairly close in gradient that could be equally argued to represent a ‘turning point’ in Bitcoin’s price action. I drew a second line in magenta, which is visible on the chart above. This line still starts at the same Point A back in June 2011. I chose its Point B as the low on 27 December 2018 when the price briefly dipped to $3566.
The new line over its entire history looks like this:
Interestingly the magenta line has intersected the price basically every time the cyan line has. In fact it has had some clean bounces as well, on 26 January 2015 (top of first recovery from 2014 bear market), 4 November 2015 (brief rally blow-off top) and a less clean one on 16–20 June 2016 (Bitfinex hack), as well as other periods when the price has hovered near it:
In the middle of 2013 when the market was consolidating after an epic rise it experienced in early 2013, there was a clear (to me) 5-sub-wave 4th Elliott wave, before a final 5th wave which carried the price past $1,000 in late 2013. Each sub-wave crossed both the cyan and magenta lines, like this:
Updated prediction
In the short term, if you are an optimist you might say that the magenta line is a better indicator of the market turning from Bear to Bull. But it’s possible we still need to cross both lines before the market turns. Either way, I’m reasonably optimistic of a change in the market soon.
In the longer term, I said in the previous article that Bitcoin needs some sort of a 4th wave before it can make a new all-time high, and nothing has happened yet to change my opinion. For much of 2018 I thought the meme-triangle we saw from February to October could have been it, but it broke down, not up.
I’m now thinking that the market has only just completed or will soon complete the first sub-wave of a 5-sub-wave 4th Elliott wave consolidation pattern. In the previous article I thought that might take 24–30 months to complete, in part out of the fact that the next halving is due in mid-2020 and also in part because Bitcoin was in a continuous Bull market for 3 years from 2015–2017 and the subsequent Bear market would need to be of similar duration.
That prediction hasn’t been invalidated yet, but it was surprising to me that sub-wave 1 (if that’s what it is) took almost a full 12 months. If the 4th wave does take 30 months the later sub-waves will need to be a bit more compact, like this:
Therefore it would not surprise me if Bitcoin has a mini-revival in the first part of 2019, but I would be cautious about expecting a new all-time high before the next halving. Especially if there’s four more consolidation sub-waves to go.
Trivia
For the magenta line, I used Excel to calculate the daily rate of increase, using the following data points:
Point A: 8 June 2011 $31.9099
Point B: 27 December 2018 $3566
Number of days between A and B: 2759
Ratio of price between A and B: 111.75215
Daily rate of increase = EXP(LN(111.75215)/2759) = 1.00171087960839, or about 0.171% per day.
To get the annual rate of increase, raise the daily rate to the power of 365.25, which gives 86.7% per annum, compared to the cyan line’s 81.1% per annum.