• Francisco Hernandez

Bitcoin 2020 Halving: Price, hashrate, difficulty, and z-scores

Introduction

There are many perspectives on how the Halving will affect price, miners, hashrate, and difficulty. For example, according to a research report by Blockware Solutions (2020), the authors hypothesize price support is established by miner capitulation, reductions in hashrate, and difficulty adjustments. BitMEX Research (2020) suggests that hashrate will not be impacted as severely due to the relationship between the difficulty adjustment and the mining industry structure. Julie Ros (2020), of Profit & Loss, interviewed a handful of digital currency market participants which offered a variety of perspectives ranging from the halving "is priced in," Bitcoin "is not a flight-to-safety investment," and "price could do anything." In this article, I analyze the Bitcoin price, hashrate, and difficulty using z-scores, a statistical method used to standardize and compare different samples of data. I compare the 2012 Halving environment to the current 2020 Halving environment and identify an oscillating pattern between price, hashrate and difficulty.


Z-Scores

First, I will briefly explain why I am using z-scores. If you already know about z-scores, you can skip to the graphs section below. Then, I will go into the analysis.


I'll start by listing the data samples I am using: price, hashrate, and difficulty. These are all measured in different units. For example, the Bitcoin price is measured in USD and the current value is in the ten-thousands (i.e., $XX,XXX.XX) place. Hashrate is measured in hashes per second (h/s) and the value is in the one-hundred millions (i.e., XXX,XXX,XXX.XX h/s) place. If I plotted price and hashrate on the same axis it would look like this:

Price and hashrate, no z-score, same axis.


The hashrate units are so large, it makes the price look like a flat like at zero. Not very useful. Here is how the graph looks when hashrate is on a different axis:

Price and hashrate, no z-score, different axis.


This is better to look at, however price and hashrate are two completely different things, measured in completely different units. While this graph is better to look at, it still does not allow for a true comparison given price and hashrate are in different units. This is where z-scores come in. Without going into too much detail, a z-score is a standardized unit used to compare different samples of data, such as price and hashrate. For more details on z-scores see this wiki. The important thing to know about z-scores is that it allows for a true comparison between two or more samples of data that originally were measured in different units. Here is the full history of price, hashrate, and difficulty in z-scores (notice how hashrate is adjusted below compared to the graph above):

Full history of price, hashrate, and difficulty in z-scores.


Graphs

For the remainder of the article, I will be using z-score charts. Specifically, I will be using charts that plot the z-score's 14-day moving average across price, hashrate, and difficulty. I use the 14-day moving average because that is roughly the length of time it takes for miners to mine 2,016 blocks (Fun Fact!: Every 2,016 blocks there is a difficulty adjustment.) I also use the 14-day moving average because the graph looks smoother. Look at both the full history graph above (no moving average) and the full history graph below (yes moving average).

Full history of price, hashrate, and difficulty in z-scores, 14-day moving average.


Analysis

Hashrate and difficulty

First, I want to discuss the relationship between hashrate and difficulty. Then, I will break down the charts and examine them in portions between 2012-2017 and 2017-2020.


Below is the hashrate and difficulty from 2017-2020. Notice the lines follow each other closely. They are essentially one line. This has to do with the relationship that hashrate and difficulty have. When hashrate (miners securing the network) drops, difficulty will adjust accordingly and reduce in difficulty once 2,016 blocks are mined. This is true vice versa too. If hashrate increases, difficulty will increase after the next 2,016 blocks. Difficulty does this in response to hashrate to ensure that one block is mined on average every ten mins (i.e., 2,016 blocks every two weeks). Difficulty lags behind the hashrate, constantly trying to balance out with hashrate. In other words, difficulty will do what hashrate does. This direct relationship can also be seen in a correlation.

Hashrate and difficulty, 2017-2020, 14-day moving average.


A correlation between hashrate and difficulty suggests the two have a very strong, significant, and positive relationship, r(4134) = .993, p < .01. In other words, if hashrate goes up, difficulty will also go up, and vice versa. A correlation coefficient of r = .993 suggests that hashrate accounts for about 98.61% (i.e., X^2; coefficient of determination) of the changes in difficulty. Hashrate and difficulty are two peas-in -a-pod, they are best buddies, and cannot be apart from each other longer than 2,016 blocks. Given the strong relationship between hashrate and difficulty, I treat them as if they were a single line at times. Other times, I will examine them separately.

Correlation matrix between price, hashrate and difficulty.


As seen in the correlation matrix above, hashrate and difficulty are also significantly correlated with price; respectively, r(4134) = .766, p < .01 and r(4134) = .758, p < .01. This relationship is positive and moderate in strength. However, each hashrate and difficulty only explain about 50% (i.e., X^2; coefficient of determination) of their relationship to price. This suggests there are other variables that are affecting price besides hashrate and difficulty. In other words, price may not always follow what hashrate and/or difficulty do, but it can.


Now, I would like to move on to the next section where I analyze two periods of time, 2012-2017 and 2017-2012, and how price, hashrate, and difficulty have behaved in the past and how it may behave in the current 2020 Halving environment.


2012-2017

The graph below starts on 7/18/2010, which is the date for the first recorded Bitcoin price on coinmetrics.io. Price, hashrate, and difficulty all were essentially moving horizontally from 2010 to early 2013, even through the first Halving in 2012. It was not until a year later that price peaked in late 2013.


Before that, price remained below both hashrate and difficulty. Then, price broke through, peaked, and started the 2014 bear market. Price eventually broke down past both the difficulty and hashrate line but managed to stay hovering below the hashrate line. Notice that up to this point, hashrate and difficulty are flat and parallel. This suggests that price changed not because of changes in hashrate or difficulty, but in response to some other phenomenon. Perhaps it was lower supply and higher demand, a higher stock-to-flow ratio, FOMO, and/or the launch of Bitcoin magazine earlier that year? I do not know, but it does seem that hashrate and difficulty had nothing to do with change in price at that time. That is, until the end of the bear market just months before the next halving.


In the months preceding the 2016 Halving, price, hashrate, and difficulty began to be responsive to each other. As price started to break past the hashrate line in November 2015, both hashrate and difficulty started to rise. Overall, the farther price pushed up, the higher hashrate and difficulty rose. This continued into the remainder of 2017, which we will look at next.


Anyway, let us back track a little. Look at the 2016 Halving. Price was rallying prior to the Halving and was above both the hashrate and difficulty line. Perhaps the increase in price was due to positive sentiment regarding the Halving, given no significant changes occurred in hashrate or difficulty (although the trend is slightly upward). Price then corrected, broke down past the difficulty line, and continued upward into the next bull run.


Before I move on, I want to point out a pattern I see: Price is oscillating between both hashrate and difficulty. At first, price started below both hashrate and difficulty. Then price broke above them. Then price broke below them. Then back above. Price tends to wonder back towards the hashrate and difficulty lines. Keep this pattern in mind as I will mention it again.

Price, hashrate, and difficulty, 2012-2017, 14-day moving average.


2017-2020

As price continued to increase, so did hashrate and difficulty. In fact, hashrate and difficulty continued to rise well into the bear market of 2018. This is where that pattern comes back into play. Price soared above the hashrate and difficulty, so the pattern suggests that price must return towards the hashrate and difficulty lines. This is exactly what occurred in September of 2018. In the following months, the price was dumped and deviated below the hashrate and difficulty lines. According to the pattern, price, hashrate and difficulty should oscillate back toward each other in the upward direction, which occurred in May of 2019. Price remained above both the hashrate and difficulty lines for a few months before oscillating back to the downside. After that, we arrive to the present, a few days before the 2020 Halving. Currently, price is significantly below both the hashrate and difficulty line. The pattern suggests that they should all converge once again and oscillate towards the upside. The question is when will they converge?


According to BitMEX Research (2020), the authors predict that hashrate (and therefore difficulty too, given the strong correlation) could fall up to -35%. Look at what happened to hashrate and difficulty around 3/12/2020 (one day after the WHO declared COVID-19 a world pandemic). That was a -31% fall in hashrate, which led to a similar size drop in hashrate. A -35% drop in hashrate from the current position, would place the convergence point near the 2.8 z-score, which roughly translates to between the $12,000 and $13,000 area. That gives me a target, but still does not answer the "when" question. However, look at the hashrate and difficulty from March 2020 to today. Hashrate and difficulty both hit an all-time high and have formed a lower high. A lower low will likely follow, given a -35% drop in hashrate. This would set a bearish tone to the hashrate and difficulty. This would give me a better idea of when the next convergence, or oscillation, between price, hashrate and difficulty will happen.

Price, hashrate, and difficulty, 2017-2020, 14-day moving average.


Conclusion

According to both Blockware Solutions (2020) and BitMEX Research (2020), they each expect that there will be a selloff in price and a fall in hashrate (and therefore difficulty). Blockware Solutions suggests that price will find support after miner capitulation. That is, when the Halving occurs, miners will be rewarded less Bitcoin. The theory is that, due to the loss of income in Bitcoin, some miners (i.e., inefficient miners) will have to close their mining operations, hence reducing the hashrate. Based on the correlation, difficulty will follow the direction that hashrate goes. This means that the difficulty of mining blocks will reduce, making it easier for the remainder of the miners to mine. Remember, though, that difficulty lags behind hashrate and is dependent on miners mining 2,016 blocks. If there are less miners mining after the halving, they may take longer to mine 2,016 blocks, pushing back the difficulty adjustment, which could potentially congest the network. Miners will have to work through a few difficulty adjustments before reaching equilibrium between price, hashrate, and difficulty. As a result of this, it is estimated that hashrate could fall -34% (BitMEX Research, 2020). Such a drop makes the atmosphere bearish regarding hashrate but might actually be considered favorable in regard to difficulty (Blockware Solutions, 2020). Lower difficulty is considered favorable when the more efficient miners that are left over doing all the work earn the block rewards the inefficient miners (i.e., those who dropped off after the Halving) gave up. If it is easier to mine, then they can mine more blocks. The more blocks they mine, the more Bitcoin they get.


In the current environment, price is sitting below the hashrate and difficulty. This is like the 2012 Halving environment where price was below both. If price were to mimic what it did after the 2012 Halving, price could break above past both hashrate and difficulty and be looking at new all-time highs. Of course, I would then expect a correction after that. Alternatively, price may remain below the hashrate and difficulty lines for the next few months if it is not able to break past them (i.e., hashrate and difficulty act as resistance to price).


Overall, there has been a lot of positive sentiment leading up to the 2020 Halving. This is evident given the rising price. I hypothesize that Bitcoin will likely continue to rise in price given the Halving hype but will not be related to changes in hashrate or difficulty. I am hypothesizing this because: miners will not be affected until after the Halving (so they will continue to mine as much as they can with the higher pay out before taking a cut); and Bitcoin just had a difficulty adjustment on May 5th, 2020. The next difficulty adjustment is estimated to occur on May 18th, 2020, 6 days after the Halving. I do not expect the gains from the Halving hype to be sustained. I will have to see what happens to the hashrate in the days after the Halving. It is totally plausible that hashrate begins to drop off as soon as the Halving hits. It is also plausible that hashrate drops off in the week leading up to the next difficulty adjustment. In either case, I do expect miners and hashrate to drop off steadily in the weeks following the Halving. As for the price, I expect it to follow the pattern I discussed above. Price oscillates between hashrate and difficulty. Therefore, price should move towards the hashrate and difficulty lines. I predict this convergence may occur around the $12,000 to $13,000 area. Regarding when, the convergence may happen this month or it may happen in a few months, perhaps around October 2020.


Works cited

  1. BitMEX Research (2020). Mining incentives (part 4): The impact of the Halvening. Bitmex blog, April 2, 2020. https://blog.bitmex.com/mining-incentives-part-4-the-impact-of-the-halvening/

  2. Blockware Solutions (2020). Understanding Bitcoin market participants: Vulnerabilities in the price of Bitcoin driven miners. Research Report, March 17, 2020. https://www.blockwaresolutions.com/research-and-publications/2020-halving-analysis

  3. Ros, J. (2020). Halving in the time of Coronavirus. Profit and Loss, April 21, 2020. https://www.profit-loss.com/halving-in-the-time-of-coronavirus

BTSE regularly hosts the technical analysis of experienced BTSE traders. Please note that we neither endorse any analyses published nor give you financial advice. Please make any trading decisions based on your own research.

Our aim is to create a platform that offers you the most enjoyable trading experience. If you have questions or suggestions, please don’t hesitate to reach out to us at feedback@btse.com or DM us on Twitter: @BTSEcom.

Categories
Company
Community

© 2019-2020 btse.com All rights reserved