Hashprice in 2026: Record Security, Record-Low Pay

The 48 hours that split Bitcoin in two

On 29 October 2026, Bitcoin’s mining difficulty retargeted higher by 5.54 percent to 148.36 trillion, its highest reading since the all-time high set in late 2025 and the second jump of nearly 6 percent in two weeks, after a 5.92 percent move on 16 October (CryptoSlate). In the same window the number miners actually watch, hashprice, fell to about $33 per petahash per day, its weakest level since August. The day before, the Federal Reserve held its policy rate at 3.75 to 4 percent (CNBC), Bitcoin slipped below $80,000 (CoinGecko), and the one lifeline miners had been counting on, a rate cut, came off the table.

Put those two readings side by side and you have the defining tension of Bitcoin mining in late 2026. The network has rarely been more secure: difficulty sits within roughly 5 percent of its record, and total computing power has climbed back above one zettahash per second. Yet the people paying for that security have rarely earned less per unit of work. Hashprice is the single figure that captures both facts at once, which is why, on most mornings, it matters more to a mining operation than the Bitcoin price itself.

This piece explains what hashprice is, how it is assembled from four moving parts, and why network security and miner pay pulled in opposite directions this autumn. The short version: security is a dollar total that rises with price and hashrate, while pay is a rate per unit of hashrate that falls as hashrate rises. In October, both forces fired at once.

What hashprice actually is

Hashprice is expected miner revenue per unit of hashrate, per day. It is quoted in United States dollars per petahash per second, per day (written USD/PH/day). If hashprice is $33, then a machine or a fleet contributing one petahash per second of computing power earns about $33 in gross revenue over 24 hours, before it pays for a single kilowatt-hour of electricity. The term was coined by Luxor’s Hashrate Index in 2019 (Luxor documentation), and it has become the mining industry’s version of a wholesale spot price: the going rate for selling raw computation to the Bitcoin network.

A miner does not really sell bitcoin; a miner sells hashes. Two things set the payout: how much bitcoin a unit of hashrate wins in a day, and what that bitcoin is worth in dollars. Hashprice folds both into one number, so it answers the only question an operator cares about on a given day: at today’s price, today’s difficulty, and today’s fees, what does my hardware earn? That question is more useful than where the Bitcoin price sits, because a miner with expensive power can go broke in a bull market while a miner with cheap power can thrive in a bear market.

The formula: a numerator and a denominator at war

Strip hashprice to its mechanics and it is a fraction. Expected daily revenue per unit of hashrate equals the network’s daily payout divided by the network’s total hashrate: roughly 144 blocks a day, times the reward per block (the 3.125 BTC subsidy plus average transaction fees), times the Bitcoin price, all divided by total network hashrate. Luxor’s index expresses the same relationship through difficulty, which is tied to hashrate, but the intuition is cleanest as a numerator over a denominator.

The numerator is what the network pays out: the block count, the block reward, and the dollar value of each coin. It rises when Bitcoin’s price rises or when a fee spike fattens each block. The denominator is how many machines share that payout, the sum of every miner’s computing power worldwide. It rises every time new hardware plugs in.

Here is the part that trips up newcomers. The reward pool in the numerator is fixed in bitcoin terms. The protocol issues the same 450 BTC a day (144 blocks times 3.125 BTC) whether one machine is running or ten million. So when hashrate climbs, the same pot is split into thinner slices, and hashprice falls even if the Bitcoin price has not moved a cent. That is precisely what late October delivered: a bigger denominator, an unchanged pot, a smaller slice for each terahash.

From terahash to take-home: a worked example

Numbers make it concrete. Take a single Antminer S21 Pro, a mainstream 2024-era machine rated at 234 terahash per second (0.234 PH/s) and 15 joules per terahash, so it draws about 3.5 kilowatts. At a $33 hashprice it earns 0.234 times $33, roughly $7.72 a day in gross revenue. Running flat out it burns about 84 kilowatt-hours a day. At a cheap industrial rate of 5 cents per kilowatt-hour, that is $4.21 of electricity, leaving around $3.50 of gross margin per machine per day, before staff, rent, pool fees, and the depreciation of the hardware itself. At about 9 cents per kilowatt-hour, the same machine earns nothing: revenue and power cost cancel exactly.

That break-even power price is the number that decides which machines stay switched on. It is a single division: break-even dollars per kilowatt-hour equals hashprice divided by 24 times the machine’s efficiency in joules per terahash. Feed in the current hashprice and you get a survival map by hardware generation.

Read the table as a waterline. A new, efficient rig still clears a profit at power prices up to 10 to 14 cents, which is why the industry keeps buying the latest Antminer and WhatsMiner units. An older S19j Pro needs electricity under about 5 cents just to break even. So when a falling hashprice meets a rising denominator, as it did this autumn, the oldest and least efficient fleets are the first pushed off the grid. Which side of that waterline a given rig lands on depends almost entirely on its electricity contract, the one variable hashprice leaves out; HOGE Wire’s companion primer on the metric maps that energy geography in detail.

Scale that single machine up and the stakes become clearer. An industrial site running one exahash per second, roughly 4,000 of those S21 Pro units, earns about $33,000 a day in gross revenue at a $33 hashprice, or close to $1 million a month. Trim the hashprice to $28, the level it touched in June, and the same site loses around $5,000 a day of revenue without changing a single thing about its own operation. That is why operators watch hashprice the way an airline watches jet fuel: it is the input they cannot control yet cannot ignore.

Why difficulty just rebuilt to a near-record

Bitcoin’s difficulty is a thermostat. Every 2,016 blocks, about two weeks, the protocol measures how quickly blocks were found and resets difficulty so the next stretch averages one block every ten minutes. More hashrate means faster blocks and a higher difficulty next time; less hashrate means slower blocks and a lower setting. There is no vote and no committee, only arithmetic.

Through the first half of 2026 that thermostat ran in reverse. Bitcoin spent much of the spring and early summer trading below many miners’ production cost, so high-cost rigs switched off, hashrate fell, and difficulty dropped, bottoming near 124.93 trillion in mid-June. JPMorgan’s Nikolaos Panigirtzoglou described the loop precisely: “When bitcoin trades below its production cost, higher-cost miners power down, the hashrate declines, and difficulty adjusts lower” (The Block). The bank put Bitcoin’s production cost near $78,000 and noted that the sensitivity, or beta, of difficulty to price had risen to 0.62, meaning a growing share of miners now sit close to break-even and flip machines on or off as the price moves.

Run that loop forward and you get the autumn of 2026. Bitcoin’s third-quarter recovery pulled idled machines back online just as new, efficient hardware shipped. Network hashrate crossed back above one zettahash per second in mid-October for the first time since March, coasting near 1,025 exahash per second (news.bitcoin.com). The thermostat answered with two of the year’s largest hikes back to back: 5.92 percent on 16 October to 140.58 trillion, then 5.54 percent on 29 October to 148.36 trillion. Jamie Coutts, chief crypto analyst at Real Vision, called the surge “the largest absolute amount of computing power added during a comparable hashrate recovery since 2018” (CryptoSlate). JPMorgan had flagged exactly this behaviour, expecting “larger and more frequent mining difficulty adjustments” for as long as Bitcoin trades below its production cost.

The full 2026 ladder, compiled from CoinWarz and Hashrate Index tracking, shows how far and how fast the network climbed off its June floor.

Every one of those adjustments landed on hashprice before a single miner saw an extra coin, because difficulty is the denominator. The 29 October hike alone shaved several percent off revenue per terahash the moment it took effect.

Record security, record-low pay

Here is the paradox of late 2026 in a single line. Bitcoin’s security budget, the total paid to miners to defend the chain, is near a record, while the pay rate per unit of that security is near a multi-year low.

The security budget is just hashprice multiplied by hashrate, annualised. At about $33 per PH/day across roughly 1,080 exahash per second, miners collectively earn on the order of $35 million a day, close to $13 billion a year. You reach the same figure from the issuance side: 450 BTC a day, times a Bitcoin price near $78,000, times 365, is about $12.8 billion (CoinGecko). That pile is roughly what an attacker would need to out-spend to rewrite recent history, so a bigger pile means a chain that is harder to attack. By that measure the network has seldom been stronger.

Security and pay diverge because they are measured against different things. The budget is a dollar total, and it climbs with both the Bitcoin price and hashrate. Hashprice is a rate per unit of hashrate, and it falls as hashrate climbs. This autumn, the price recovery and the hashrate surge together pushed the dollar total up, while the hashrate surge alone pushed the per-unit rate down. September was miners’ best revenue month since January, about $1.12 billion (MillionMiner); five weeks later the per-unit hashprice was at a three-month low. Both statements are true at once, and hashprice is the only number that holds them together.

Transaction fees: the numerator’s missing half

Fees sit in the numerator beside the subsidy, but in practice they barely register. Through most of 2026, transaction fees have run well under 1 percent of the total block reward, which means more than 99 percent of what miners earn is freshly issued subsidy rather than fees paid by users. That is comfortable while the subsidy is large. It becomes the central question of Bitcoin’s security the moment the subsidy shrinks.

Fees are not always an afterthought. When the Runes protocol launched in April 2024, transaction fees briefly made up about 75 percent of miners’ per-block revenue, their highest share ever, and the average daily fee jumped more than sevenfold in a single day (CoinDesk). The spike did not last: within a year, Runes-related fees had collapsed from around 90 percent of Bitcoin fees to under 2 percent (BlockEden). Whether users through fees, or holders through issuance, ultimately foot Bitcoin’s security bill is the question this cluster keeps circling back to, and HOGE Wire has argued the case at length.

The 2028 cliff

Every four years the block subsidy halves. In spring 2028 it drops from 3.125 BTC to 1.5625 BTC. Hold the Bitcoin price, fees, and hashrate steady and hashprice halves with it, because the subsidy is the dominant term in the numerator. The security budget halves too. Bitcoin has absorbed three prior halvings by rising in price fast enough to offset the cut, but each halving raises the bar for how much price growth the math requires.

Fred Thiel, chief executive of MARA Holdings, frames the problem bluntly. Mining, he argues, “is a zero-sum game. As more people add capacity, it gets harder for everybody else” (CoinGeek). On the long-promised shift from subsidy to fees, he says it simply “hasn’t happened,” and warns that “if Bitcoin[‘s fiat value] doesn’t grow at 50% or more annually, the math gets very tough after 2028.” His forecast for who survives is stark: “By 2028, you’ll either be a power generator, be owned by one, or be partnered with one.” At a hashprice near $33 and a production cost near $78,000, that 2028 clock is already shaping which miners invest and which quietly wind down.

When the Fed becomes the other lever

If difficulty is the denominator miners cannot control, the Bitcoin price is a numerator they cannot control either, and in late 2026 that price has taken its cues from the Federal Reserve. On 28 October the Fed held its target rate at 3.75 to 4 percent in a 10 to 2 vote (CNBC). Both dissenters, Governor Lisa Cook and Cleveland Fed President Beth Hammack, wanted a 25 basis point hike, not a cut, with inflation still described as elevated. Chair Kevin Warsh said politics played no role ahead of the midterm elections. Markets read the meeting as hawkish: the odds of a December hike climbed above 80 percent, and Bitcoin slipped below $80,000.

For miners that is close to the worst combination. A rate cut would have loosened financial conditions and, on the recent pattern, lifted Bitcoin and the numerator with it. Instead the price fell just as difficulty, the denominator, posted its second near-6 percent hike in two weeks. Both levers moved against hashprice inside the same 48 hours, which is why the late-October print felt less like weather and more like a vice closing.

Trading the number, and who regulates it

Because hashprice behaves like a commodity spot price, a market has grown up around hedging it. Luxor offers hashprice forwards out to twelve months, and Bitnomial lists CFTC-regulated hashrate futures, ticker HUP, in one-petahash monthly contracts. Through the autumn the forward curve has sat in backwardation, with six-month contracts priced below spot, the market’s way of saying it expects the price-driven part of any recovery to be clawed back by difficulty. A miner who sells forward locks in revenue and gives up the upside; one who stays on the spot rate keeps the upside and wears the volatility. In a year when the spot number swung from a five-year low near $28 in June to about $41 in early October and back toward $33, that choice has mattered a great deal.

A worked example shows why a miner might give up that upside. Picture an operator running five exahash per second who sells half of it forward at $34 for six months. If spot hashprice then sags to $30, the forward leg pays the difference on the hedged 2.5 exahash, cushioning the shortfall; if spot instead rips to $45, the operator has capped the gain on that half but still rides the full upside on the unhedged portion. Hedging does not make a miner richer on average; it narrows the range of outcomes, which is what lets a capital-intensive business borrow, budget, and survive a bad quarter.

The United States regulatory line is now reasonably clear. In March 2025 the SEC’s Division of Corporation Finance stated that proof-of-work mining, including pooled mining, is not the offer or sale of securities, because miners rely on their own effort rather than the efforts of others (SEC). Hashrate futures, by contrast, are commodity derivatives overseen by the CFTC. Spot mining is not a security; a bet on where hashprice is heading is a regulated commodity contract. That division has held even as the wider enforcement posture keeps shifting, a story HOGE Wire has tracked through 2026.

Miners versus AI

The fiercest competitor for a miner’s megawatt is no longer another miner; it is an artificial intelligence data center. The same power contracts, land, and grid connections that host ASICs can host graphics processors, and the economics are not close. John Todaro, an analyst at Needham & Co., put it plainly: “The revenue per megawatt and EBITDA margins are far higher for HPC and AI colocation than for mining” (CoinGeek). Through 2026, several listed miners have signed multi-billion-dollar compute deals, and a few have begun stepping back from bitcoin production altogether to lease capacity to AI tenants instead.

That shift resolves an apparent contradiction: how can hashrate keep setting records while hashprice sinks? Part of the answer is new, efficient machines replacing old ones at the same power draw. Part of it is that the operators with the cheapest power and the strongest balance sheets keep expanding hashrate even as the marginal, high-cost miner powers down or pivots away. The broader scramble to make off-chain computation both cheap and verifiable runs well beyond mining, into the machinery that lets a network prove an AI model actually ran as promised.

Hashprice for miners, staking yield for stakers

Hashprice has a close cousin on the proof-of-stake side of this mining-and-staking cluster: staking yield. Both answer the same question, namely what the market pays to secure a chain per unit of the resource committed, energy and hardware for Bitcoin, capital for Ethereum. And both compressed to multi-year lows in 2026 for the same structural reason: the supply of security outgrew the budget that pays for it. Bitcoin’s hashrate hit records as hashprice fell; Ethereum’s staked share hit records as its staking yield drifted toward multi-year lows.

The difference is the control mechanism. Bitcoin’s difficulty thermostat resets automatically every two weeks, with no human in the loop, while changes to Ethereum’s issuance curve require contested governance. A validator key engineered so that no single operator can steal it is a very different safeguard from a difficulty retarget, but the economics rhyme: in both systems, the price of security falls when more of the securing resource shows up than the reward pool can pay.

How to read hashprice without fooling yourself

Hashprice is powerful precisely because it compresses so much into one figure, and that is also how it misleads. A few guardrails keep it honest.

• It is gross, not net. Hashprice says nothing about your power cost, the single input that decides whether a given rig makes money. A $33 hashprice is a fortune at 3-cent power and a slow bleed at 12-cent power.
• Watch the bitcoin-denominated version. Priced in BTC rather than dollars, hashprice barely moved through the autumn recovery, hovering near 0.00045 BTC per PH/day. That tells you the dollar swings were a price story, not a network story.
• It is a spot rate, not a forecast. The forward curve, not today’s print, is the market’s actual view of where revenue is heading.
• Mind the calendar. The next retarget, due in early November, was tracking higher again, and the Fed’s December meeting carries live odds of another hike. Two more tightenings could reach miners before year-end, and the 2028 halving is now less than 18 months away.

The lesson of late October 2026 is that hashprice is not one story but two held in tension: a network that has never been stronger, and a business that has rarely been thinner. The miners who survive the gap will be the ones who treat the number not as a paycheck to cash but as a spread to be hedged, powered cheaply, and defended against every rival for the same megawatt.

Frequently Asked Questions

What is a good hashprice for Bitcoin miners?

There is no single good number, because profitability depends on a miner’s power cost. A useful rule of thumb is the break-even power price: hashprice divided by 24 times the machine’s efficiency in joules per terahash. At the late-October 2026 hashprice near $33 per PH/day, an efficient modern rig breaks even around 10 cents per kilowatt-hour, while an older machine needs power under about 5 cents. Any hashprice that leaves a comfortable margin over a miner’s actual electricity rate is good for that miner.

Why did hashprice fall while the Bitcoin price recovered?

Hashprice is revenue per unit of hashrate, and it falls when total hashrate rises, because the fixed daily reward of 450 BTC is split among more machines. Through autumn 2026 the Bitcoin price recovery pulled idle machines back online and new hardware shipped, so hashrate climbed above one zettahash per second and difficulty jumped twice by nearly 6 percent. That bigger denominator outweighed the higher price, pushing hashprice to a three-month low near $33 even as the network set security records.

How is hashprice calculated?

Hashprice is the daily block reward paid to the whole network divided by the whole network’s hashrate. In practice that is about 144 blocks a day, times the block reward of 3.125 BTC in subsidy plus fees, times the Bitcoin price, divided by total network hashrate. The result is quoted in United States dollars per petahash per second, per day. The metric was coined by Luxor’s Hashrate Index in 2019.

What happens to hashprice at the 2028 halving?

In spring 2028 the block subsidy halves from 3.125 BTC to 1.5625 BTC. If the Bitcoin price, fees, and hashrate stay the same, hashprice halves too, because the subsidy is the dominant term in the numerator. Historically Bitcoin’s price has risen enough across a halving to offset the cut, but each halving raises the price growth required, which is why miners watch the 2028 date so closely.

Is Bitcoin mining still profitable in late 2026?

It depends almost entirely on power cost and machine efficiency. With hashprice near $33 per PH/day, miners with cheap electricity, roughly 5 cents per kilowatt-hour or less, and modern hardware remain profitable, while higher-cost operators running older rigs are at or below break-even and are powering down or pivoting to AI and high-performance computing. JPMorgan has estimated Bitcoin’s average production cost near $78,000, close to where the price has been trading.

Yuki Tanaka is a DeFi and mining correspondent for HOGE Wire, covering the economics of proof-of-work and proof-of-stake.