Bitcoin hashrate and difficulty: what they say about the network

Bitcoin hashrate is the combined speed at which miners worldwide guess block hashes, measured in exahashes per second (EH/s, 10¹⁸ hashes). Mining difficulty is the network rule that resets the bar every 2,016 blocks so that, whatever the hashrate, a block is found about every 10 minutes. Hashrate tells you how much computing power secures the network and how much pressure miners are under; difficulty is how the protocol responds. As of Oct 14, 2026, hashrate was estimated at roughly 1,005 EH/s, difficulty stood at 132.72T, and the next adjustment was expected around Oct 16 at about +5.8% (mempool.space).

What is hashrate, and why can't anyone measure it directly?

A miner builds a candidate block and hashes its header with SHA-256 over and over, changing a field called the nonce each time. The block is valid only if the resulting hash is below a target number. There is no shortcut to finding such a hash, so the odds of winning a block scale with the number of guesses per second. That rate is hashrate.

The catch: nobody can see every mining machine on the planet. Network hashrate is an estimate derived from two observable facts — the current difficulty and how often blocks actually arrive:

hashrate ≈ difficulty × 2³² ÷ average block time (seconds)

Two consequences follow. Different dashboards show different hashrate figures because they average over 3, 7 or 30 days. And short spikes are often just luck — a run of fast blocks rather than new machines coming online. That is why analysts watch moving averages, not single-day readings.

How the bitcoin difficulty adjustment works

The idea goes back to the Bitcoin whitepaper, which describes difficulty as a moving average aimed at a steady number of blocks per hour. In Bitcoin Core it is implemented in chainparams.cpp and pow.cpp:

1. Target block time is 10 minutes; a target epoch is two weeks (2,016 blocks × 10 minutes).
2. After every 2,016 blocks, nodes measure how long the epoch actually took.
3. New difficulty = old difficulty × (two weeks ÷ actual time). Faster blocks push difficulty up; slower blocks pull it down.
4. A single adjustment can't move difficulty by more than a factor of 4 either way — the code clamps the measured timespan between a quarter and four times the target.
5. Difficulty stays fixed inside an epoch. If a large share of miners drops off mid-epoch, blocks simply slow down until the next retarget. Bitcoin has no emergency adjustment.

Every full node enforces this rule independently, so miners can't collude on an easier target.

Hashrate and difficulty as of Oct 14, 2026

These figures come from the "Bitcoin network" panel in our market numbers, sourced from mempool.space, snapshot at 4:00 AM ET on Oct 14:

Run the formula: 132.72T × 2³² ÷ 600 s ≈ 950 EH/s — the hashrate that would produce exactly one block every 10 minutes at today's difficulty. Blocks have actually been arriving about every 567 seconds, which implies more power: 132.72T × 2³² ÷ 567 ≈ 1,005 EH/s. That 5⁠–⁠6% gap is exactly why the network is on track to raise difficulty by roughly 5.8%.

Zooming out, mempool.space's adjustment history shows difficulty about 15% below its October 2025 peak, while estimated hashrate is about 23% below the daily high near 1,306 EH/s on Oct 25, 2025. 2026 has been choppy for miners: difficulty fell 11.16% on Feb 7, jumped 14.73% on Feb 19, dropped 7.76% on Mar 20 and 10.09% on Jun 14.

What hashrate says about miners: hashprice

Miners care less about difficulty itself than about hashprice — the dollars one PH/s of computing power earns per day. You can approximate it yourself: the network produces about 144 blocks a day at 3.125 BTC each, or roughly 450 BTC (see our halving explainer for how the subsidy shrinks).

Example for Oct 14, 2026: 450 BTC × $83,582 (BTC price at 4:00 AM ET, from our dashboard) ÷ 1,005,225 PH/s ≈ $37 per PH/s per day, before fees. Fees barely matter right now: in the week to Sept 28 they made up just 0.70% of block rewards, and Hashrate Index put spot hashprice at $39.87/PH/s/day. Our back-of-the-envelope figure is rougher, but in the same range.

Why it matters: hashprice falls when BTC gets cheaper, when difficulty rises, and at every halving. Once it drops below the power cost of older machines, those machines get switched off. Hashrate falls, difficulty follows, and the remaining miners earn more per unit of work. In August 2026, ForkLog reported hashrate about a third below its October 2025 peak, with some public miners redirecting capacity to AI and high-performance computing.

Does hashrate predict bitcoin's price?

"Price follows hashrate" is half true at best. The causality usually runs the other way: a higher price makes mining more profitable, and new machines get plugged in. Buying and deploying hardware takes months, so hashrate lags price and doesn't forecast it on its own.

The best-known indicator built on this data is Hash Ribbons by Charles Edwards of Capriole. It compares the 30-day and 60-day moving averages of hashrate: a 30-day average below the 60-day marks miner capitulation, and the cross back up marks its end (LookIntoBitcoin chart and notes). As of Oct 14, 2026, our calculation on mempool.space daily data puts the 30-day average (~968 EH/s) above the 60-day (~945 EH/s) — no capitulation signal. For other cycle gauges, see our guide to MVRV and the Mayer multiple.

Why a high hashrate means a safer network

To rewrite recent blocks, an attacker needs more than half of the network's hashrate. At around 1,000 EH/s that means millions of modern ASIC miners and gigawatts of power, which makes an attack prohibitively expensive. Over the long run, rising hashrate is read as thicker armor for the network; a sharp, sustained drop is a reason to look at what's happening to miners.

History also shows how resilient the mechanism is. The deepest difficulty cut in mempool.space's records is −27.94% on Jul 3, 2021, after China's mining ban. Blocks slowed for a while, the network kept running, and hashrate eventually recovered.

Common mistakes

• Comparing hashrate across sites. They're estimates with different averaging windows; a 5⁠–⁠10% gap is normal.
• Reading one day of data. Daily hashrate is noisy — use 7- and 30-day averages.
• Treating a difficulty increase as bullish for price. Difficulty just catches up with hashrate, up to two weeks late, and says nothing about future price.
• Confusing hashprice with profit. Hashprice is revenue before power, pool fees and hardware.
• Expecting an instant response. If hashrate drops mid-epoch, difficulty only falls at the next retarget.

FAQ

How often does bitcoin mining difficulty change? Every 2,016 blocks, roughly every two weeks. The exact date can't be known in advance because it depends on how fast blocks arrive.

Why do hashrate numbers differ between websites? Hashrate can't be measured directly. It's estimated from difficulty and block times, and each service uses its own averaging window.

Does hashrate affect bitcoin's price? Not directly. Price usually drives hashrate, since a pricier BTC draws in more miners. Hashrate is a gauge of miner health, not a forecast.

Where can I track hashrate and difficulty? In the "Bitcoin network" panel on our dashboard and on mempool.space. The next halving date is on our halving indicator page, and the coin profile is on the Bitcoin page.

This is not investment advice.