Hashrate is one of the key technical indicators of the bitcoin network. It directly affects the profitability of mining. Miners do not have the ability to hedge the risks of fluctuations in the hashrate due to the lack of reliable methods for calculating the indicator, according to CoinMetrics.
The company's specialists decided to fill in the gaps and proposed new metrics for determining the hashrate, as well as futures contracts based on it.
They described their innovative tools and products in the article.
There is a lot of speculation about the impact of the upcoming halving on the bitcoin hashrate. If for most of the community this is more of a thought experiment, then for miners the result will determine not only profitability, but also the probability of survival.
This is largely due to the lack of a reliable, generally accepted methodology in the market for hedging industry uncertainties.
CoinMetrics has offered two new tools for effectively hedging hashrate-related risks using derivatives:
1. CMBI Bitcoin Hash Rate Index (bitcoin hashrate index).
2.Work Observed (observed work).
The value Hasrat
Mining is one of the main innovations of Satoshi Nakamoto. Without mining, there would be no bitcoin.
network security, prevent abuse, and prevent hackers from interfering with the public registry;
mining new bitcoins;
ordering and broadcasting all transactions;
checking and adding information about new transactions to the registry.
Usually, miners estimate their costs based on three factors:
hardware (capital expenditures);
electricity (operating expenses);
hashrate of the network to determine the possibility of making a profit.
The hashrate of bitcoin is not constant and predictable. The indicator changes significantly over time and does not always follow the price of the cryptocurrency.
It is critical for miners to hedge the hashrate and improve their ability to maintain profitability.
For example, let's take a large mining company that decides whether to enter the market. Its budget is sufficient to purchase equipment that will provide 1% of the network's hashrate. At the same time, they can count on 18 BTC per day. At a price of $8000 per coin, if operating expenses exceed $144,000 per day, the company will not make any profit. If the operating cost is $100,000 per day, the company will make a profit and should consider entering the market.
Three months later, ASIC miners arrive at the site — the price of bitcoin has risen to $10,000 per coin, but the network's hashrate has doubled. Now the company has only 0.5% of the total indicator. This means an award with an average of 9 BTC per day.
At a price of $10,000, the revenue will be $90,000. Taking into account the operating costs of $100,000, the company receives $10,000 daily loss.
If it had the ability to hedge risk by trading derivatives based on a hashrate, the firm would minimize the impact of a change in the indicator.
Note: the example uses illustrative shapes, and for simplicity, ignores the impact of the upcoming halving.
Development of tools necessary for a hashrate-based financial product
It is almost impossible to get a reliable hashrate value in real time.
Currently, when determining the indicator, the speed of creating blocks and the current level of complexity are taken into account. This is the same as if traders trading oil futures used prices at gas stations to calculate oil production around the world.
In fact, the hashrate indicator is derived from historical data. CoinMetrics has identified three key issues to overcome when creating a hashrate derivative:
1. Short-term predictability. Since the hashrate calculation depends on past data, most of it is already known. For example, if the indicator is determined based on a 48-hour observation window, then we already have 47 points to determine the hashrate for the last hour. The short-term hashrate is relatively predictable.
2. The implied hashrate fluctuates around the equilibrium point (graph below). This creates two types of settlement risks for contracts. First, the calculation will take place at the upper or lower point of the oscillation range, which to some extent determines the randomness. This can significantly affect the result of a trade transaction. Second, fluctuations can be manipulated by large miners who control a significant share of the hashrate.
3. A fixed-duration hashrate contract does not take into account what happens between opening and closing. Imagine that a 3-month contract opens and closes at the same level. If the miner wanted to hedge his position by opening a long position, at the time of calculation, his profit would be $0. But if the average hashrate for the period was 20% higher than the opening/closing rate, it would also have received less revenue than expected.
Another approach to the development of derivatives for hedging Hasrat is to use of complexity. Although this indicator has some advantages over the hashrate-based index, it is also not without disadvantages:
1. The difficulty is recalculated every 2,016 blocks (~2 weeks). This means that it is difficult to evaluate it at the early stages of the contract, since the estimation of future complexity is essentially impossible and subject to significant fluctuations.
2. Long-term contracts for complexity do not take into account its levels over the entire duration of the contract. The complexity may remain unchanged when opening and closing a contract. If it was higher during the period, it does not provide effective hedging (unless the contract has liquidity that can be managed in real time).
3. Complexity is subject to manipulation. A few days before the contract closes, miners can significantly affect the outcome of the recalculation by deliberately disabling the equipment.
To reduce the impact of many of the above problems, CoinMetrics has developed the following tools that can become the basis for an effective hashrate-based derivative.
1. CMBI Bitcoin Hash Rate Index (bitcoin hashrate index)
The hashrate can be calculated based on the time it took for miners to generate new blocks over a recent period. Bitcoin was designed so that the average block mining time is 10 minutes. Every two weeks, the complexity is adjusted to maintain this interval.
Since finding a solution for bitcoin blocks is a random process that follows the Poisson distribution law, the time between blocks can vary greatly.
This can lead to volatility when determining the hashrate. Today, the industry standard is to calculate the indicator based on the previous 24-hour period. However, for a structured financial product, CoinMetrics considers this too unpredictable and volatile. For this reason, we have introduced a 48-hour period for THE cmbi Bitcoin Hash Rate Index.
Let's explain by an example when the production of three blocks in 24 hours took more than 50 minutes in September. As a result, the daily hashrate fell by more than 30%. However, this was most likely the result of an accidental, unlikely, but explicable event. The indicator for the 48-hour period was also affected, but its drop was less noticeable - < 20%.
In the image above, you can see that the 48-hour segment repeats much less volatile movements than its 24-hour counterpart.
2. Observed Work
As discussed above, it is not enough to develop a hashrate index taking into account predictability, undesirable randomness (due to fluctuations), and the problems of traditional models. For this reason we have created the Observed Work.
The blockchain assigns a fixed number of hashes to a block based on the current complexity (i.e., between difficulty adjustments, regardless of whether the block search takes one second or one hour, the block is assigned the same value).
To better calculate the work of miners and the number of hashes, the observed work from CoinMetrics is calculated as follows:
Observed Work = (implied hashrate level for 48 hours) X (time spent searching for the last block)
This indicator better reflects the realities of mining than the standard model for calculating completed work.
Coinmetrics ' proposed futures contract for Observed Work
The miner knows the hashrate of their hardware, but does not know what computing power other miners generate and what they will be able to generate in the future. Based on the implied hashrate, the miner understands its current share in the aggregate and, therefore, the expected revenue/share of block rewards.
Observed Work was developed for financial service providers to create financial products:
for market participants to speculate on hashrate;
so that miners can effectively manage the risks associated with hashrate, having the ability to hedge based on the observed number of hashes over a certain period of time.
The complexity allows you to better understand the hashrate expectations over a period of 2,016 blocks. For this reason, a derivative based on Observed Work will allow you to effectively speculate on the expectations of the number of hashes in the absence of knowledge about the movements of the hashrate on fixed timeframes.
The following is a theoretical example of a 50-minute (3000 seconds) contract for Observed Work, provided that:
The market expects that at time t = 0 the hashrate will be 100 EH/s.
The block search is expected to take 600 seconds, as defined in the bitcoin white paper.
At the time of opening the contract, a reasonable expectation of the estimated price would be 300,000 EH/s (3000 seconds X 100 EH/s). As shown below, despite the same hashrate value when opening and closing a position, the contract will adjust and close higher than expected at 308,050 EH/s. Let's see what happened, block by block:
Block 1 took the expected 600 seconds. The estimated hashrate has not changed, and the observed work in 60,000 EH/s is equal to the expected work.
Block 2 was detected after 10 seconds, which increases the implied hashrate. At time 0, the expected operation in 10 seconds was 1000 EH/s (10 seconds X 100 EH/s). However, as the hashrate increased, the observed work increased to 1050 EH/s.
Block 3 was also fast, so the estimated hashrate increased, and the observed performance was higher than expected at time 0.
Block 4 was the first slow block to take more than 600 seconds. The hashrate dropped to 105 EH/s. This is still higher than the 100 EH/s we expected at time 0, which leads to this block also having higher observed performance than expected.
Block 5 became another slow block, resulting in an implied hashing speed of 100 EH/s, which is the initial figure. Given this, the observed work for this block is equal to the expected work, producing 119,000 EH/s.
This scenario is not uncommon for bitcoin. A similar result was demonstrated in mid-January (pictured below).
As you can see, the implied hashrate did not change at the closing point. Despite this, the complexity increased by 5%, as the hashrate was higher than expected for most of the period.
If the miner had opened a long position on the hashrate, it would not have reached the predicted yield from its activities at the beginning of the period, since the average hashrate was higher than expected. In addition, it would have made a very small profit on the long position on the hashrate, because it closed near the opening level of the contract.
The observed performance futures have different results for the same period:
the observed work exceeded the expected one;
a miner with a fixed hashrate contract would have received a lower return over the period than expected;
if the miner opened a long contract he would have received the profit from the increase observed in the work during the entire period.
Below is an example of a three-month contract for observed work since the first complexity recalculation in 2020. You can visually observe how the expectations of such a contract change over time as more information becomes available.
From the examples above, we can see that such a derivative could solve many of the problems that prevented the emergence of successful hashrate-based products:
Predictability. The work performed during the contract is always increased, making it less susceptible to hashrate fluctuations. The observed performance is highly dependent on the random block mining time combined with hashrate fluctuations. In this sense, the discrepancy between expected and observed performance is less predictable than short-term changes in hashing speed.
Measure performance over the life of the contract. Hashrate contracts can be closed at the upper or lower points of fluctuations, which creates an undesirable risk for traders. In addition, the levels Hasrat do not reflect the behavior of the metrics during the term of the contract. The observed work reflects the entire history of the " work” and is not so affected by the fluctuating hashrate.
Manipulability. The observed work can improve the resistance to manipulation of hashrate-based products by adding time-weighted measurements that obey the law of random Poisson distribution.
Mining is one of the main innovations of bitcoin, which allowed everyone to use a decentralized, distributed and sovereign currency. Hashrate is an important on-chain metric that provides market participants with information about network security.
To date, miners do not have effective tools for hedging risks, they are completely dependent on the price of bitcoin. However, the market continues to develop with the support of venture capital and representatives of traditional markets. These companies will look for mechanisms to hedge their risks and operations in the same way as with traditional assets.
CoinMetrics expects that the cmbi Bitcoin Hash Rate and Observed Work index will become the basis of financial products that will eventually be able to provide the markets with the necessary tools for effective trading and / or hedging the hashrate