Bitcoin's network just performed its routine self-adjustment — and the number is a story in itself. At block height 959,616 on Saturday, July 26, 2026, mining difficulty retargeted down 0.74%, the network's 15th adjustment of the year, per Bitcoin.com News. After a year of dramatic double-digit cuts, this one barely moved the needle — and that is precisely why it matters.
We flagged this retarget as a committed follow-up in Sunday's difficulty-adjustment guide, where trackers were projecting roughly a 1% downward move. The actual print of −0.74% landed almost exactly on that estimate — and far below some of the double-digit forecasts that had circulated earlier in the epoch. For readers new to the mechanism: every 2,016 blocks (about two weeks), Bitcoin measures how fast blocks have been arriving and rewrites its own difficulty so the next stretch lands back on the 10-minute-per-block target. No committee, no vote — just arithmetic.
The sequence: a network finding its floor
Read in isolation, a 0.74% cut is a non-event. Read as the latest entry in 2026's sequence, it is the clearest sign yet that the mining industry's bleeding has slowed. The year's downward adjustments have been shrinking in magnitude:
| Retarget | Change | Difficulty after | Context |
|---|---|---|---|
| February 2026 | −11.16% | — | US winter-storm curtailment; largest cut of 2026 |
| June 14, 2026 | −10.09% | 124.93 T | ~15% June price slide compressed margins |
| July 11, 2026 | −5.00% | 127.17 T | 14th adjustment; hashrate down ~7.9% in ten days |
| July 26, 2026 | −0.74% | ~126.2 T | 15th adjustment; the gentlest trim of the run |
Sources: The Block (June 14); news.bitcoin.com (July 11 and July 26); CoinWarz difficulty chart (current). As of July 27, 2026.
The shape — −11.16%, −10.09%, −5.00%, −0.74% — traces a network that has largely finished shedding its unprofitable hashrate. Each of the big cuts earlier in the year did its job: every downward adjustment raises the amount of bitcoin earned per unit of surviving hashrate (June's cut alone lifted it about 11%, per The Block), mechanically restoring margins for the miners still running until an equilibrium is reached. A near-flat retarget says that equilibrium is close.
Hashrate and hashprice: the stress gauges
The backdrop remains one of pressure, not health. Network hashrate has been grinding lower for months — the seven-day average sat near 917 EH/s around this retarget, per CoinWarz, down meaningfully from roughly 1,065 EH/s at the start of the year and about 21% below the October 2025 peak near 1,154 EH/s. Hashprice — the daily revenue a miner earns per unit of hashrate, and the single best one-number read on mining P&L — recovered to about $31 per PH/s per day after the July 11 cut but remains roughly 37% below its October 2025 high.
That gap is the entire 2026 mining story in one figure: bitcoin's price is down sharply from last October's top while energy costs have risen, and difficulty has spent the year cutting to close the spread. When hashprice climbs because difficulty fell rather than because price rose, the recovery is real for survivors — but it confirms that weaker operators were carried out along the way.
The AI wrinkle: some hashrate is not dying, it is defecting
There is a structural force in 2026 that a difficulty chart cannot see on its own. A growing share of the hashrate leaving the network is not going bankrupt — it is being redeployed to artificial-intelligence and high-performance computing. Miners including IREN, Core Scientific and TeraWulf have signed AI and HPC hosting contracts collectively worth tens of billions of dollars, chasing the stable, dollar-denominated, high-margin revenue that AI hosting offers over volatile mining. We examined the collision between AI power demand and Bitcoin mining in our AMD–Anthropic power-race analysis.
This reframes what a difficulty cut even means. Historically, falling difficulty signaled miner capitulation — operators forced offline by thin margins. In 2026, part of it signals opportunity cost: a megawatt of power is worth more pointed at an AI cluster than a mining rig, so rational operators redirect it. Difficulty has quietly become a partial proxy for how a scarce grid prices compute — a very different signal from the death-spiral fears of previous cycles.
What to watch next
The next retarget is estimated for around August 11, 2026, and trackers currently project a much larger downward move — on the order of double digits — based on recent block times, though such estimates drift and should be treated as provisional until the retarget block is mined. If that larger cut materializes, it would suggest the near-flat July 26 print was a pause rather than a floor; if the next adjustment also comes in small or turns positive, the case that miners have found their equilibrium strengthens considerably.
For now, the takeaway is measured: after a brutal first half, Bitcoin's mining economics are stabilizing, and the network's thermostat is making smaller and smaller corrections. That is what the bottom of a miner shakeout tends to look like — not a dramatic event, but the gradual disappearance of drama. For the market backdrop into this week's Fed decision, see our Fed-week news file.
Investment disclaimer: This article is for informational and educational purposes only and does not constitute financial, investment, or trading advice. Cryptocurrency markets are highly volatile and you can lose money. Always do your own research and consult a qualified financial advisor before making investment decisions.