Coin Metrics: Can miners influence the price of BTC?

By Karim Helmy

Miners are often blamed for Bitcoin’s price declines. These accusations are often unfounded — and worse, they are sometimes based on faulty metrics that conflate mining pool payments with miner payouts, ultimately misleading users.

Accurately assessing the extent and impact of miner selling is critical to understanding the market. In Following Flows: A Look at Miners' On-Chain Payments, Coin Metrics unveiled a new method for estimating miner activity. This method takes into account the way pool wallets are structured, allowing users to distinguish between pool and miner activity.

In this post, we will refine our estimates of miner activity with data on the relationship between miners and exchanges, allowing us to determine when and where miners are selling coins. All metrics used in this post will be available in our upcoming 4.9 release of Network Data Pro.

Following traditional experience, we found that miners tend to choose Huobi and Binance over other exchanges. We also found that the flow from mining addresses accounts for a small proportion of the total inflow of exchanges, about 5.5% at the time of writing, and is not a major source of market volatility .

Mining Market Metrics

Coin Metrics' miner-exchange flow is based on our existing estimates of miner and exchange activity . Existing inflow, outflow, and supply metrics provide useful context about market sentiment, exchange health, and network decentralization. Due to the different problem structures, exchange flow and miner flow are built on two different clustering techniques, each with their own shortcomings.

Exchange flows are estimated using common inputs as well as wallet owners, assuming that addresses that are inputs to the same transaction share a single owner. This technique is precise but requires at least one seed address per exchange, limiting coverage to a predetermined universe of exchanges.

Trading on centralized exchanges usually requires users to deposit coins into the exchange, which is then held in trust by the exchange operator. Mining generally works in a similar way: miners share resources with each other to increase their chances of finding blocks, coordinating with centralized operators through mining pools. Mining pool operators typically receive newly mined coins to addresses they control, and then distribute the coins to miners through payment transactions.

Coin Metrics’ Miner Stream illustrates this architecture by clustering addresses based on their hop distance from the coinbase transaction. Addresses that have received coinbase rewards, or 0-hop addresses, are assumed to belong to the mining pool. 1-hop addresses that have received payments from 0-hop addresses are marked as belonging to miners. This analysis is less precise than the common input-ownership analysis, but roughly reflects the structure of mining pool wallets and provides better coverage.

Neither clustering can detect fresh address usage, and the results of each clustering should be considered rough estimates. Since exchanges are more precisely labeled, we prefer this analysis when resolving conflicts: addresses labeled as belonging to both miners and exchanges are treated exclusively as exchanges.

By combining these two techniques, we can assess where miners store their coins, which roughly corresponds to where they sell their coins.

The flow of funds from miners to exchanges is generally similar to the overall exchange inflows, but there are a few key differences. Binance and Huobi are currently the best performing exchanges, and Huobi's share is significantly higher than its share of overall inflows.

These results seem intuitive, as Huobi and Binance are the only two exchanges in our coverage area that also operate mining pools. These two exchanges also have close relationships with miners and have a strong presence in Asia, where most miners are located.

The outflow of funds from exchanges to miners is also mainly from Binance and Huobi. This shows that miners are also buying on these exchanges.

As with inflows, Huobi has a significantly higher share of outflows compared to exchanges that do not operate mining pools. Binance has a roughly equal share of both distributions.

Going forward, miner-exchange traffic could serve as a proxy for the geographic distribution of mining power. This would rely on the assumption that miners use exchanges in their region, and more complete coverage of exchanges would be needed in order to paint a representative picture.

Measuring Miner-Exchange Flow

Our approach can be used to assess the composition of miner-exchange flows and their overall size. In general, mining pools are net depositors on exchanges, although the amount of funds transferred each day is small.

As with their respective overall flows, 1-hop miner-exchange flows are an order of magnitude larger than 0-hop flows. While these flows have declined in BTC terms since their 2016 peak, they have gradually increased since 2018. In contrast, USD-denominated flows have surged past their previous highs, reflecting the rise in Bitcoin’s price.

Counterintuitively, based on our assumptions, miners appear to be net buyers. This is most likely a methodological issue caused by a number of factors.

One factor could be that most miner selling is done off-site, meaning funds are not immediately sent to exchanges. Accumulation by early miners could also be a contributing factor, which could be addressed by filtering out outflows from addresses that have not recently received funds from 0-hop addresses. Finally, payments from exchange-affiliated mining pools could be conflated with exchange withdrawals.

Despite this result, the aggregate flows appear to be in the right direction, and combined with the flows on specific exchanges, they are granular enough to be useful. However, since the difference between these inflows and outflows is always unexpected, the net flows calculated from these time series are likely to be of limited value.

Traffic content

Miner-Exchange Flow is best understood in the context of its constituent flows. To understand the scale of Miner-Exchange Flow, it is helpful to compare it to Total Miner Flow and Total Exchange Flow.

Transfers to exchanges typically make up a high single-digit percentage of miner flows, though this percentage has historically varied. Deposits to exchanges also make up a not insignificant percentage of mining pool flows.

Withdrawals from exchanges account for a higher percentage of miner inflows, likely due to exchange-affiliated mining pools and over-the-counter sales. Withdrawals from zero-jump addresses generally account for a much smaller percentage of total flows, although they have recently begun to increase.

In theory, miner deposits to exchanges should correlate with selling, but miner deposits have typically accounted for only a single-digit percentage of exchange inflows over the past few years. At the time of writing, miner deposits account for about 5.5% of exchange inflows. While most miner selling occurs over-the-counter, this number is likely skewed upwards due to the wide net cast by the 1 analysis, meaning the true on-exchange selling pressure provided by miners may be lower.

While miner-exchange traffic is inherently volatile, its fluctuations are small and insignificant in the context of overall exchange traffic, and traffic between 0-hop addresses and exchange addresses is negligible. Therefore, we see little reason to attribute the downward movement in Bitcoin prices to miner selling.

Miner withdrawals account for a larger portion of total exchange outflows, generally staying above 10% last year. The discrepancy between this figure and the proportion of inflows coming from miner deposits may be related to the role of exchange affiliates, which may have close ties to their parent companies and may make payments from overlapping addresses.

The lack of relationship between price and miner activity is evidenced by the low correlation between price and deposits. This lack of correlation also holds true when training only on days when prices fall.

The lack of relationship between price and miner spending can also be visually confirmed by comparing the changes in price and miner deposits. These values ​​rarely move in sync, indicating low correlation. While the distributions of daily percentage changes are skewed compared to the distribution of the underlying variable, they are suitable for informal visual analysis.

in conclusion

The impact of Bitcoin miners on the broader market remains understudied. However, by developing more sophisticated methods to assess the scale of miner activity, we can begin to better understand the role miners play. Given the small size of miner deposits compared to total transaction flow, and the lack of correlation between miner transaction flow and price, we find little reason to believe that miners are the cause of Bitcoin's price decline.

There is still a lot of room for improvement in these metrics, especially in terms of increasing exchange coverage and filtering out addresses that have not recently engaged in mining activity. Nevertheless, exchange-level flows are consistent with conventional wisdom that Binance and Huobi are the exchanges of choice for miners, informally providing a check on the validity of these metrics.