What Is Hashprice? Bitcoin Mining’s Key Metric in 2026

Ask a Bitcoin miner how business is going and you will rarely get an answer about the price of Bitcoin. You will get a single figure: hashprice. It is the revenue a machine earns for each unit of computing power it contributes to the network every day, and it is the closest thing mining has to a wage. In late October 2026 that wage was brutal. Hashprice sat in the mid-$30s per petahash per day, within touching distance of the lowest levels ever recorded, even as the network locked in a fresh difficulty record of 148.36 trillion on October 29.

This guide explains what hashprice is, how it is calculated, and why it has become the one number that decides which miners keep their machines switched on. It also covers why that number keeps falling even as more computing power floods the network than ever before, and what miners are now doing to hedge it. If staking yield is the per-unit return a validator earns for securing a proof-of-stake chain, hashprice is its proof-of-work twin: the realized, after-the-fact return on raw work.

What hashprice actually measures

Hashprice measures expected mining revenue per unit of hashrate per unit of time. The industry-standard unit is United States dollars per petahash per second per day, usually shortened to dollars per PH/s per day. One petahash is one quadrillion (10 to the power of 15) hashes per second, roughly the output of four or five top-end machines running flat out. The metric is deliberately machine-agnostic: it describes the network’s payout to work, so a miner can hold a brand-new rig and a four-year-old one up to the same scale.

If a miner is told hashprice is $35 per PH/s per day, it means a machine contributing one petahash of sustained hashrate should earn about $35 of gross revenue over 24 hours, before electricity and everything else. There are two flavors. USD hashprice is denominated in dollars; BTC hashprice, sometimes called hashvalue, is denominated in bitcoin. The dollar version tells a miner whether the lights stay on this month; the bitcoin version strips out price and isolates how the protocol itself is rewarding work, as Luxor’s index documentation lays out.

One point trips up newcomers constantly: hashprice is gross revenue, not profit. It says nothing about a miner’s power contract, hosting fees, debt, or hardware depreciation. Two miners facing the exact same hashprice can be worlds apart on margin, because one pays four cents a kilowatt-hour and the other pays eight.

Scale helps the unit make sense. The whole Bitcoin network now does on the order of a thousand exahashes per second, and one exahash is a thousand petahashes, so a single PH/s is a tiny sliver of global hashrate. Hashprice takes that sliver and tells you what it is worth per day, which is why it travels so easily between a hobbyist with one machine and a listed company running hundreds of thousands.

The word, and why Luxor invented it

Hashprice is not a protocol term. It does not appear anywhere in Bitcoin’s code. It is a market convention coined by the mining services firm Luxor in 2019 to give miners, investors, and lenders a common yardstick. Before the term existed, operators compared revenue with a messy mix of coins per terahash, dollars per megawatt, and back-of-envelope breakeven prices that were almost impossible to line up across different fleets.

Luxor’s Bitcoin Hashprice Index turned that into a single published benchmark: the expected value of one PH/s of hashrate per day. Because it is a benchmark rather than a protocol value, the exact figure depends on methodology. Luxor’s index uses a 144-block lagging simple moving average (roughly one day) to smooth transaction fees, and an average of spot bitcoin prices across three United States exchanges for the dollar conversion, settling once per UTC day from the average of 15-second snapshots.

Lenders care about this as much as miners do. Because hashprice sets the revenue that services mining loans and equipment leases, a standardized index made it possible to underwrite the debt that financed much of the industry’s build-out. Jaran Mellerud, founder of the mining advisory firm Hashlabs, argues that if a miner tracks only one number, it should be hashprice, because it folds price, difficulty, subsidy, and fees into a single revenue figure. That is why it appears on nearly every miner dashboard and in nearly every mining company’s investor deck.

It helps to see where the number has been. In the 2021 bull market, with bitcoin near its highs, far less hardware online, and a 6.25 BTC subsidy, hashprice ran many times higher than it does today. Every bear market and every halving since has stepped it lower, and that long arc is the clearest picture of how much more competitive mining has become.

The formula: four inputs, one number

Under the hood, hashprice is simple division: daily network issuance in dollars, divided by the amount of work the whole network is doing. More precisely, bitcoin hashprice depends on three inputs, and the dollar version adds a fourth.

• Block subsidy: newly minted bitcoin per block, currently 3.125 BTC after the April 2024 halving.
• Transaction fees: the extra bitcoin users pay to get their transactions into a block.
• Network difficulty: how hard it is to find a block, which rises and falls with total hashrate.
• Bitcoin price: the dollar value of the bitcoin earned (dollar hashprice only).

The first two sit in the numerator and push hashprice up: more reward per block means more revenue per unit of work. Difficulty sits in the denominator and pushes hashprice down: when more machines compete, each one’s slice of a fixed daily issuance shrinks. Price simply scales the whole thing into dollars.

That table is the whole reason late 2026 is so punishing. The only input working in miners’ favor, price, is flat to soft, while difficulty, the input that hurts them, keeps setting records.

From terahash to take-home: a worked example

Numbers make this concrete. Modern machines are quoted in terahash, so start by converting: one PH/s equals 1,000 TH/s. At a hashprice of $35 per PH/s per day, each terahash earns 3.5 cents of gross revenue per day ($35 divided by 1,000).

Walk a flagship rig through it. A current-generation air-cooled miner does roughly 200 TH/s, or 0.2 PH/s. At $35 per PH/s per day it grosses about $7.00 a day before power.

These figures are illustrative arithmetic drawn from the hashprice, not live quotes, and they are gross. Now the other side of the ledger. A 200 TH/s machine often draws around 3.5 kilowatts; running 24 hours is 84 kilowatt-hours a day. At five cents per kilowatt-hour that is $4.20 of power against $7.00 of revenue, leaving $2.80 before hosting, overhead, and depreciation. At eight cents, power alone costs $6.72 and the machine barely clears its electricity bill. That gap is the entire ballgame, and it is why the next two ideas, the difficulty ratchet and energy-adjusted hashprice, matter so much.

Why hashprice sits near record lows in late October 2026

Put the inputs together and the late-2026 picture is stark. Difficulty rose 5.92% on October 16 to 140.58 trillion, then climbed another 5.54% on October 29 to a record 148.36 trillion. Each jump is a direct cut to hashprice, because the same daily issuance is now split across more work.

Meanwhile bitcoin has drifted lower, trading around $80,000 to $82,000 in late October after a softer month. With the subsidy fixed at 3.125 BTC and fees thin, there is nothing in the numerator to offset the climbing denominator. The result is hashprice in the mid-$30s per PH/s per day, with one tracker putting it at $36.55 on October 22.

For context, that is not far from the all-time low. Luxor’s index printed a record $35.06 per PH/s per day, or 0.00041 BTC, on November 22, 2025, when bitcoin had fallen 32.2% from its October 2025 peak of $124,485 to about $84,348 and difficulty had roughly doubled since the previous hashprice bottom. Late October 2026 has miners back in that same neighborhood. The squeeze shows up directly in miner share prices and in the quiet shelving of expansion plans; when hashprice sits this close to breakeven, the marginal operator’s question is not how fast to grow but whether to keep the oldest machines plugged in at all.

Context matters for how severe this really is. The 2022 bear-market bottom, considered brutal for miners at the time, sat near $55 per PH/s per day; late 2026 hashprice is well below that even though bitcoin trades far higher than it did in 2022. That is the difficulty story in one sentence: the network has absorbed far more hardware than price alone would support, and hashprice is where the imbalance shows up.

The difficulty ratchet: a denominator that only climbs

The cruel part of hashprice math is that the denominator almost never falls for long. Difficulty adjusts every 2,016 blocks, about every two weeks, to keep blocks arriving every 10 minutes on average. When more hashrate joins, blocks come faster, and the next adjustment makes mining harder to pull the pace back to target. The adjustment is purely mechanical and backward-looking, so miners feel the pain of a difficulty increase whether or not the Bitcoin price cooperated over the same stretch.

Through 2026 that ratchet has been relentless. Network hashrate fell hard in late 2025 and early 2026, dropping about 27% at one point, then came roaring back; the seven-day average pushed back above 1,000 EH/s in mid-October 2026, its highest since early March. For perspective, hashrate sat far below today’s levels in the stretch right after the 2024 halving. More machines, same issuance, lower hashprice.

Why would anyone add machines when revenue per machine is falling? Because efficiency improves. Newer rigs produce more hashes per watt, so an operator with cheap power and new hardware can still profit at a hashprice that bankrupts someone running older machines on expensive power. Each efficiency wave pushes difficulty up and squeezes the laggards out. It is a treadmill, and hashprice is the speed setting. That dynamic has pushed mining toward large, well-capitalized operators and professional pools, with smaller participants routing hashrate through shared pools to smooth revenue, a shift underscored when Binance wound down its mining pool.

The ratchet also explains a counterintuitive survival tactic: curtailment. Large miners increasingly sign demand-response deals with grid operators, switching machines off when power prices spike and effectively selling that electricity back for more than hashing would have earned. In the thinnest stretches of 2026, those grid payments, plus the occasional fee spike, have been the difference between a positive and a negative month for some public operators, which is why the biggest fleets now sit beside flexible power markets rather than merely cheap ones.

The halving effect: hashprice’s four-year cliff

Every four years the block subsidy halves, and hashprice takes an instant, mechanical hit. The April 2024 halving cut the subsidy from 6.25 BTC to 3.125 BTC, and hashprice responded at once. On April 29, 2024 it printed what was then an all-time low of $49.44 per PH/s per day (0.00079 BTC), roughly half the level just before the halving and about 10% below the previous bear-market record of around $55 set in November 2022.

That cliff is the most predictable event in mining economics, and it reshapes the industry every cycle. Miners who cannot get below the new breakeven are forced off; those who can pick up market share, and often the cheap secondhand machines the losers leave behind.

The next halving, expected in 2028, will cut the subsidy again to 1.5625 BTC. Unless price or fees rise to compensate, the arithmetic points to another step down in hashprice, which is exactly why fees are the number miners watch most closely for the long run.

Transaction fees: the wild card in the numerator

Of the four inputs, three are either fixed (the subsidy) or set by the wider market in ways no single miner can move (price and difficulty). Fees are the only part of hashprice that can surprise to the upside. When blockspace is in demand, users bid up fees, and that revenue flows straight to whoever mines the block.

Fee spikes are episodic and usually tied to new on-chain activity. The 2024 launch of the Runes protocol around the halving briefly stuffed blocks with fee-paying transactions and lifted hashprice before the frenzy faded; the ongoing race for scarce short Runes names is a direct descendant of that moment. Ordinals inscriptions did something similar in 2023. The lesson for miners is that fees can briefly double block revenue, but they cannot be relied on, which is precisely why Luxor’s index smooths them with a 144-block average.

The long-term worry runs deeper. As the subsidy shrinks with every halving, fees must eventually become the dominant source of miner revenue if the network’s security budget is to hold up. In late 2026, with fees near multi-year lows, that transition still looks distant, and it is the central open question hanging over mining’s next decade.

There is a structural reason fees have stayed quiet in 2026. Much of the speculative blockspace demand that drove the 2023 and 2024 spikes has either moved to layer-two systems and sidechains or simply cooled, leaving base-layer blocks cheaper to get into. For a miner that is a double-edged outcome: lower fees mean a thinner numerator today, but they also mean the network has spare capacity to monetize the next time a new on-chain craze arrives.

Energy-adjusted hashprice: the number that decides who survives

Hashprice tells a miner what a machine earns. It does not tell them whether that machine makes money, because it ignores the single largest cost: electricity. To close the gap, analysts use energy-adjusted hashprice, which converts the benchmark into a breakeven power price.

The formula is straightforward: energy-adjusted hashprice equals hashprice divided by machine efficiency divided by 24 hours, which resolves to dollars per kilowatt-hour. In plain terms it answers the question every operator actually cares about: at today’s hashprice, what is the highest power price at which my fleet still breaks even?

Take a machine at about 17 joules per terahash, a reasonable current-generation efficiency, which is 17 watts per TH/s, or 17 kilowatts per PH/s. At a hashprice of $35 per PH/s per day, the breakeven electricity price lands near eight to nine cents per kilowatt-hour. Below that, the machine earns more than its power costs; above it, every hour it runs loses money. Cooling matters here too: immersion and hydro setups let operators push machines harder for the same electricity, nudging effective efficiency up and the breakeven power price with it. This is why miners obsess over power contracts and site themselves near stranded hydro, flared gas, or curtailed wind: a two-cent edge on power is the difference between profit and shutdown at today’s hashprice. It also sets up the single biggest structural story in mining right now, the exodus to artificial intelligence.

Trading hashprice: forwards, futures, and hashrate as collateral

Because hashprice is volatile and largely outside any single miner’s control, a market has grown up to let operators lock it in. Luxor, the same firm that coined the term, built the first hashrate forward market and later listed Bitcoin hashrate futures on the Bitnomial exchange. The idea is simple: a miner sells future hashrate at a fixed price today, receiving bitcoin upfront, and an investor on the other side gets exposure to mining revenue without buying a single machine.

Aaron Foster, Luxor’s director of business development, has described hashprice as the «hourly or daily revenue that miners get», one that «fluctuates a lot», and framed the derivatives as a way to «create a tool that allows investors to take a position on hashprice without effectively owning mining equipment». He calls the model «basically the collateralization of hashrate».

The forward curve has stayed low to match spot. In September 2026, Luxor’s forward market priced an average hashprice of about $36.65, or 0.00044 BTC per PH/s per day, over the October 2026 to March 2027 window. For miners, selling forward at those levels is less a bet on upside than an insurance policy: it converts a volatile, record-low revenue stream into something a lender will underwrite. The mechanics will feel familiar to anyone who has priced funding on a perpetual futures venue, where the cost of holding a position is set by a market rather than a protocol.

The buyers on the other side are mostly trading desks, funds, and energy firms that want Bitcoin mining revenue as an asset without the operational burden of warehouses and ASICs. For them a hashrate forward is a clean way to express a view on difficulty and fees; for the miner it is project finance. As the market deepens, the forward curve itself has become a signal, because a curve priced well below spot is the market saying it expects difficulty to keep grinding higher.

Miners versus AI: why hashrate keeps rising as hashprice falls

Here is the paradox of late 2026. Hashprice is near record lows, which should drive machines off the network, yet hashrate is near record highs. Part of the answer is efficiency and cheap-power expansion in places such as Ethiopia (up about 129% year over year), Paraguay (up 54%), and Oman (up 28%). But the bigger story is that the most valuable thing a miner owns in 2026 is often not its machines; it is its power contracts and data-center shells, and those are worth far more to artificial intelligence than to Bitcoin.

The clearest example came in August 2026, when Riot Platforms signed a 20-year lease worth roughly $9.1 billion (up to $16.1 billion with extensions) to supply 191 megawatts at its Rockdale, Texas campus to Anthropic for AI compute. Riot is a Bitcoin miner converting part of a mining site into AI capacity, and it is not alone: IREN has moved to retire roughly 23.2 EH/s of mining hardware by year end, while other operators redirect power entirely. For some listed miners, the market now values the AI and high-performance-computing option on their power pipelines more highly than the mining itself, which only deepens the incentive to convert megawatts away from hashing.

As public miners pull capacity for AI, private and international operators buy their retired rigs and plug them in where power is cheap, keeping hashrate elevated. Haipo Yang, chief executive of the mining pool ViaBTC, has argued that AI is effectively pushing Bitcoin mining out of its most expensive sites, while CryptoQuant chief executive Ki Young Ju has framed the hashrate recovery as an indirect bullish signal for the network’s health. The net effect on hashprice is downward: AI is rich enough to outbid Bitcoin for electricity, but the machines it displaces rarely vanish; they migrate to cheaper power and keep difficulty climbing.

What hashprice leaves out

A published hashprice is a network average, not the number any single miner actually banks. The first wedge between the two is the mining pool. Most miners point their machines at a pool and accept a payout scheme in exchange for steadier income. Full-pay-per-share pools, known as FPPS, pay a smooth blend of subsidy and fees and take a cut, usually a couple of percent; pay-per-last-N-shares pools, or PPLNS, pay closer to raw luck and reward loyalty. Either way the miner nets a little less than headline hashprice, and the gap widens when fees are volatile.

Luck and variance matter too. Over a day or a week a small operation can find more or fewer blocks than the math predicts; hashprice describes the expected value, not the realized one, which is exactly why pools exist. Uptime is another hidden variable: a machine that is curtailed during peak power prices, throttled by heat, or down for maintenance earns nothing while it is idle, no matter how high hashprice climbs.

Hardware is not static either. Operators overclock machines to squeeze out more terahash at the cost of efficiency, or underclock them to ride out a low-hashprice stretch on cheap but limited power. Each choice moves both the revenue and the power side of the equation, so two identical rigs can post very different real-world returns.

None of this makes hashprice less useful; it makes it a starting point. A disciplined operator takes the published figure, subtracts pool fees, applies a realistic uptime, and adjusts for its own machine efficiency to reach a true, site-specific revenue number. That adjusted figure, set against the power contract, is what actually decides whether a given row of machines earns its keep this month.

Hashprice for miners, yield for stakers: the same question

For readers who come to this from proof of stake, hashprice has a close cousin: staking yield. Both answer the same question, which is what one unit of committed resources earns for securing a network. A validator commits capital (staked ETH) and earns an annual percentage yield; a miner commits work (hashrate) and earns hashprice. Both are driven by issuance split across total participation, and both fall as more participants pile in.

The parallel is strikingly tight in late 2026. Just as hashprice sits near record lows because hashrate keeps climbing, Ethereum’s staking yield has drifted to cycle lows as the validator set has grown, with staked ETH paying only low single digits; our look at Ethereum solo staking in 2026 walks through why more validators mean a thinner slice for each one. The difference is cost structure. A staker’s main variable is opportunity cost, while a miner’s is a very real electricity bill, so mining is far more sensitive to its revenue number than staking is. That sensitivity is exactly why hashprice, not the Bitcoin price, is the figure that keeps miners awake at night.

How to read hashprice without fooling yourself

A few habits separate people who understand hashprice from people who quote it wrongly.

• Know which currency you are in. USD hashprice can rise purely because bitcoin’s price rose, even if the protocol is paying less per unit of work; BTC hashprice, or hashvalue, strips that out. Compare like with like.
• Remember it is gross. Hashprice says nothing about your power price, hosting, debt, or depreciation. Two miners at the same hashprice can have opposite fortunes.
• Watch the denominator, not just the price. A rising bitcoin price can be fully cancelled by a difficulty increase, and in 2026 difficulty has repeatedly done exactly that.
• Treat fee spikes as weather, not climate. A Runes or Ordinals surge can double revenue for a few days; no one should build a business plan on it.
• Mind the methodology. Published indexes smooth fees and average prices across exchanges, so a benchmark can differ from what your specific pool paid on a specific day.

Used carefully, hashprice is the most honest single number in Bitcoin mining. It will not tell you whether to buy the dip or where bitcoin is heading next. It will tell you, cleanly and daily, whether the machines in the warehouse are worth switching on.

Frequently Asked Questions

What is a good hashprice for Bitcoin mining?

There is no fixed threshold. A good hashprice is any level above your fleet’s energy-adjusted breakeven. An efficient miner on 4-cent power can profit in the mid-$30s per PH/s per day, the level seen in late October 2026, while an operator on 8-cent power running older machines may lose money at the same figure. The relevant comparison is always hashprice against your own cost per kilowatt-hour, not an absolute number.

How is hashprice calculated?

Hashprice divides daily network mining revenue by total network hashrate, expressed as dollars per PH/s per day. The bitcoin-denominated version depends on the block subsidy (3.125 BTC), transaction fees, and difficulty; the dollar version adds the bitcoin price. Published benchmarks such as Luxor’s index smooth fees with a roughly one-day moving average and average spot prices across several United States exchanges.

Why is hashprice so low in 2026?

Because the denominator has grown far faster than the numerator. Network difficulty hit a record 148.36 trillion on October 29, 2026 as hashrate climbed back above 1,000 EH/s, while the block subsidy is fixed at 3.125 BTC, fees sit near multi-year lows, and bitcoin has traded softly around $80,000 to $82,000. More work chasing the same issuance means less revenue per unit of hashrate.

What is the difference between hashprice and hashvalue?

Hashprice usually refers to the dollar figure, in United States dollars per PH/s per day, while hashvalue refers to the same metric denominated in bitcoin, in BTC per PH/s per day. Hashvalue isolates how the protocol rewards work regardless of price, while hashprice tells a miner whether the operation covers dollar costs such as electricity.

Can you hedge or trade hashprice?

Yes. Firms such as Luxor run hashrate forward markets and list Bitcoin hashrate futures on the Bitnomial exchange, letting miners sell future hashrate at a fixed price for bitcoin upfront and letting investors gain mining-revenue exposure without owning hardware. In September 2026 the forward market priced roughly $36.65 per PH/s per day for the following six months.

By the HOGE Wire mining desk, covering Bitcoin mining economics and proof-of-work markets.