Bitcoin’s mining network cooled off a bit. The seven-day average hashrate slipped to 915.8 EH/s, while miner-held BTC also edged lower. That does not mean the network is broken, but it does suggest some miners are under pressure and changing how they run the business.
- Hashrate eased: Seven-day average fell to 915.8 EH/s.
- Miner reserves dipped: Holdings fell by 1, 530 BTC over the week.
- Margins are being tested: Some miners are selling BTC or reshaping operations.
- The chain kept moving: Bitcoin continued producing blocks as designed.
On Sept. 26, Digital Asset put Bitcoin’s seven-day moving average hashrate at 915.8 EH/s, down about 34.86 million TH/s from a week earlier. CryptoQuant data cited in the same report showed miner reserves at 1, 192, 766 BTC, down 1, 530 BTC over the same period.
That is a real shift, but not a panic signal on its own. Hashrate data is noisy. Different trackers use different estimators, averaging windows, and inputs, which is why one dashboard can show one thing while another shows something else entirely. CoinWarz, for example, showed network hashrate around 954 EH/s on Sept. 25 after roughly 984 EH/s the day before, and readings above 1 ZH/s on Sept. 15 and Sept. 9.
In plain English: hashrate is the combined computing power securing Bitcoin. A seven-day average smooths out daily noise, but it still moves around for all kinds of reasons, maintenance, power prices, weather, equipment changes, and block timing variance. One week lower does not mean the network is falling apart. It means the network is doing what large, messy, decentralized systems do, it fluctuates.
Bitcoin’s block production kept going normally. That matters more than the hand-wringing. The protocol is built to absorb changes in mining power through its difficulty adjustment, which recalibrates roughly every 2, 016 blocks so block times stay near target over time. Short-term softness in hashrate can affect miner economics, but it does not automatically threaten block production.
The miner reserve data needs a careful read too. CryptoQuant’s own framing is useful here: when tracked miner balances fall, it means coins have moved out of those wallets, but the destination and purpose can vary. That could reflect spot sales. It could also mean treasury reshuffling, custody transfers, lending, collateral moves, or wallets that are simply not tagged the same way.
So no, lower miner reserves do not automatically mean miners are “dumping.” Sometimes they are covering operating costs. Sometimes they are moving coins around to manage the balance sheet. Sometimes they are selling because electricity and financing do not wait around for a romantic HODL thesis.
There is at least some evidence that miners are feeling the squeeze on revenue, even if the picture is not dramatic. Digital Asset put Bitcoin’s Puell Multiple at 1.13 on Sept. 26, up 0.24 on the week. HalvingLens showed a reading closer to 1.01 and labeled it normal. The difference likely comes down to methodology, timing, or price inputs, but the takeaway is similar: miner issuance revenue was around typical levels, not at some extreme blowoff or breakdown point.
The Puell Multiple compares the daily USD value of newly issued Bitcoin with its 365-day moving average. A reading above 1 means issuance revenue is above its one-year average. That is useful context, but it is not a profit metric. It does not account for electricity, debt, cooling, staff, or hardware costs. A miner can look “fine” on a chart and still be getting squeezed in real life. Charts are neat. Power bills are rude.
Raphael Zagury, CEO of Twenty One Capital, called the period Bitcoin’s first “hashrate bear market.” That is a sharp phrase, but it should be treated as commentary, not doctrine. It captures the idea of a sustained pullback from a prior peak, not some formal network failure. Bitcoin has been through miner stress before, and it will again.
Some of that stress is showing up in how miners allocate capital. Hyperscale Data stopped Bitcoin mining at its Michigan facility on Sept. 1 and has shifted toward AI computing. That example is concrete, and it fits a broader pattern: some operators are repurposing power, land, and data-center capacity for workloads that can produce steadier returns than Bitcoin mining in certain locations.
That is not a betrayal of Bitcoin. It is business. If a site can earn more on AI inference or training than on hash production, of course management will compare the options. The real question is not whether miners should be loyal to a logo. It is how much infrastructure stays committed to Bitcoin if AI keeps offering better economics.
CleanSpark shows another side of the same market reality. In August, the company mined 593 BTC but sold 821 BTC, ending the month with 13, 703 BTC and an average operating hashrate of 38.3 EH/s. That is not a contradiction. It is treasury management. Mining companies are not just stacking coins for internet points; they are running capital-intensive businesses that need cash flow, debt discipline, and room to keep buying hardware.
Power constraints can also bite hard at the local level. Ethiopian Electric Power cut electricity supplied to Bitcoin miners to 23% of contracted levels after reservoir inflows fell 20%. In the previous fiscal year, Bitcoin mining reportedly consumed nearly one-third of Ethiopia’s electricity and generated 35% of the utility’s revenue. That is a stark reminder that mining is, first and foremost, an electricity business.
The Ethiopia example is regional, not a global apocalypse button. Still, it shows why mining economics can shift fast when hydrology, regulation, or grid stability changes. Nature does not care about bullish narratives, and utilities care even less.
The bigger picture is straightforward. A short-term dip in hashrate, paired with a modest decline in miner reserves, looks more like pressure and adaptation than a network problem. Some miners are selling more. Some are changing strategy. Some are moving infrastructure toward AI. Bitcoin is still producing blocks.
That is the core takeaway: the mining industry is under margin pressure, but the protocol is doing exactly what it was designed to do. Bitcoin does not need every miner to be comfortable. It needs enough economic incentive, enough hardware, and enough cheap power to keep the system secure. So far, that machine is still running.
Key questions and takeaways
Does a lower hashrate mean Bitcoin is weakening?
Not necessarily. A seven-day average can move for technical and operational reasons without signaling a structural problem. The network can still be secure and keep producing blocks.
Are miners definitely selling BTC?
No. Falling miner reserves mean coins moved out of tracked miner wallets, but that can reflect sales, custody transfers, treasury moves, lending, or collateral activity.
What does a Puell Multiple near 1 mean?
It suggests daily miner issuance revenue is around its historical average. That does not tell you whether miners are actually profitable after electricity and other costs.
Why are some miners pivoting to AI?
Some sites can earn better returns from AI workloads than from Bitcoin mining. Miners are infrastructure operators, so they will chase the higher-value use for their power and data-center assets.
Will Bitcoin struggle if hashrate keeps slipping?
The protocol is built to handle changing hashrate through difficulty adjustment. Prolonged weakness could pressure smaller miners, but it does not automatically stop the network from functioning.
Further reading
A few extra resources for anyone tracking miner economics, network fundamentals, and the bigger debate around Bitcoin’s path forward.
- Bitcoin hashrate falls to 3-week low as miners cut BTC
- Comprehensive guide to Bitcoin RPC API methods
- Van Domelen Community Health Center now open in Orangetree
- Renewable & alternative fuels data browser
- The Path to a Bitcoin Standard
- Bitcoin's 4-Year Cycle Never Disappeared, It Just Evolved
- Bitcoin miner revenue squeezed as Puell Multiple drops
- Bitcoin mining difficulty drops 3% this Friday: impact on miners and network
- China slaps $14.5M fine on Xinjiang firm for powering Bitcoin mining operations
