Le hashrate n’est pas ce que vous croyez | Bitcoin 2026

Hash rate is the most cited number in Bitcoin, but strictly speaking, it's not a number that we can measure. There's no global gauge that tells us what the current hash rate of Bitcoin is. Rather, it's something that we estimate backwards. We infer it from blocks against difficulty over a rolling window. So, a figure on a dashboard is an inference about work that must have been performed. In order to produce the chain that we observe in a steady state that distinction doesn't matter but under stress it becomes the entire problem. Hash rate measures effort. Security is what happens when that effort breaks. So the standard model says that more hash rate is more security and that has a one in a one period attack model that's uh directionally correct. You have a larger honest network so you have a higher work threshold for an attack but that holds one variable constant time. It assumes stable power costs, stable machine fleet, that miners remain solvent, and that the capacity on paper is capacity that can respond. It assumes that operations that or operators that look independent fail independently. In May of 2021, half of the global hash rate was went offline within weeks. The protocol survived. The difficulty adjustment did exactly what it was designed to do. But the recovery of capital, the reallocation of fleets, the reconfiguration of power markets and supply chains took 6 months to two years, depending how you measure. The hash rate is the stock, but the security is the flow. And the standard model only shows the stock. the standard model or rather the uh the hash rate is a snapshot but security is a trajectory. Those are different mathematical objects. You can have two networks with identical hash rate and completely different trajectories under stress. One can have uh miners that are lightly leveraged run flexible power contracts hold idled machines as reserve capacity and operate firmware that they control. And the second can have the same firm rate, the same hash rate concentrated in dense sites with short duration ASIC backed debt locked into taker pay power with similar firmware and exposed to the same power grid. So those look the same on the dashboard, but they have very different security trajectories. The question uh the about of uh for the attacker is not what's the how much hash power is there. It's how much credible hash power remains available over time. Machines that are not uh machines that can't get power are not uh are not credible security. Sites are not credible security if the transformers have an 18month replacement window. operations uh are not um credible security if the next margin shock breaches a covenant. What matters is not h, it's h of t. It's the hash rate over time and more specifically under stress. The power grid, the policy, the firmware and the credit environments uh are all variable mechanisms. um that affect the security uh aside from the hash rate. So let's say that we had a 30% hash rate shock overnight. The immediate effect is that that 10-minute block time gets divided by 7. You have uh 14.3 minute blocks. That means that the time for a difficulty adjustment goes from 2016 blocks being two weeks to about 20 days. That's a 43% increase in real time right off the bat. So the first uh change that I want to make explicit is that the protocol uh adjustment clock slow itself slows. So the non-equilibrium state lasts longer. Now consider the minor economics during that gap. The intuition is that with 30% gone, the remaining terraashes earn more power per ter earn more revenue per terahash. That's partially true, but before adjustment, the original difficulty remains and the revenue per terash doesn't change uh in a on a per wall clock time basis. So per unit of wall clock time, the revenue per terahash doesn't change. The difficulty adjustment is where the repricing actually occurs. But that's 20 days out. 20 days is a really long time in capital markets. That's time for operators with thin margins uh operating on fixed power obligations with ASIC backed debt to breach their debt service coverage ratios and trigger forced curtailment to push collateral underwater on a marktomarket basis and miners do not adjust to this uniformly. The ecosystem resorts. Some miners survive the repricing and they benefit when the difficulty drops, but the others are gone. So the 30% shock uh is not just a number, it's a sorting mechanism and it separates the capacity that's installed from the capacity that's resilient. This is where security is determined. After the shock, three things happen. One, there's lower work to compete. An attacker uh doesn't need as needs 30% less work to compete with the network. Fine. Two, the network has yesterday's view of the world. The difficulty hasn't adjusted. And third, and this is the one that's often not taken into account, nobody knows the true state of the system in real time. As we said earlier, the rate is inferred. So under real shock uh well rather nobody knows when that just by looking at the block times whether there's a real shock or whether it's just block time variance. Block arrival is on a poisonson process. So operations dashboards always smooth over the difficulty over a rolling window. They do this because the raw signal is too noisy. Nobody can uh infer um clean information from that stochcastic signal. But this induces decision lag. Exchanges, lenders, market makers, miners, insurers, power count counterparties, they're all trying to determine were the block were the miners unlucky? Is the network impaired? Is it entering a cascade? These entities update at different speeds and they update based on different data. This is the vulnerability window. It's not a cryptographic window. It's not just a cryptographic window. It's an information window. It's a coordination window. And it's a liquidity window. And so the attacker's problem changes. The required capital to compete uh is down. That's the cost effect. But there's also a time effect that uh 60 that a 60-minute confirmation time say for settlement on a on an exchange becomes an 8 becomes 85 minutes 43% up. This makes coordination between adversaries easier. It makes a defender response harder. Risk policies adjust unevenly against uh across counterparties and the defender's reaction surface fragments. The resilient parties close quickly. The there's flexible load. There's they have liquid operators. They have uh capacity that can return without new transformers, interconnects, uh new credit contracts, new firmware deployments. But the fragile miners leave that window open, not because the protocol failed, but because the physical and financial layers couldn't respond fast enough. So what matters isn't the starting hash rate at the time that there's a shock. What matters is the recovery curve. Attackers don't need permanent control. They need an attack window. Security is not the depth of the curve, it's the area under the curve. And the danger moment is not low hash rate, it's unstable hash rate. We uh it's 15minute talk, so I'll skip over the math. So three things that uh are not obvious from the hash rate. One is capital. Industrial mining is financed. They have uh equipment loads that are coll or loans that are collateralized by AS6. They have um project finances that are tied to specific sites. They have rate linked uh rate linked um uh loans. And they all have covenants. So debt service coverage ratios, collateral maintenance triggers, uh those are all covenants tied to the hash price and the revenue per terahash per day uh determines whether those covenants are breached. So when the rate drops temporarily, that's not volatility. That those are triggers. the operations have to curtail or liquidate and it's not the system that failed, it's the balance sheet. So temporary protocol conditions lead to permanent loss of capacity. Second is energy. Uh what matters isn't the energy contracted, it's how flexible that energy is. So how do take or pay contracts behave? They behave like fixed cost floors during shocks. The interruptible loads can curtail when prices spike. They can monetize demand responsibly. They can re-energize when conditions improve. It's not the fastest or rather it's not the largest megawatt. It's the fastest megawatt. And also as we said before uh there's coupling networks lock onto uh um common shared dependencies. Uh AS6 supplies concentrate in three vendors. Even those three vendors buy capacity from the same foundaries and they have shared firmware stacks. Large transformers take 18 to 24 months uh for a replacement lead window. Physical destruction is uninsurable in real time under those conditions and geographic dispersion can still mean uh that shared exposure to one grid operator jurisdiction uh and weather conditions um are present. So from an underwriting perspective, the correlation risk looks diverse until it's stressed and then it looks like a single position. >> Welcome to predict. The world is a market. Everything is a market. Every headline moves the line. Every moment is your market. Call the moves. Bet on your instinct, your prediction, your edge. Dual bits predict where everything is a market. Hash rate doesn't wait for difficulty. So when the hash rate drops, collateral uh collateral drops uh um or rather collateral obligations trigger uh credit is tightened, power contract obligations kick in uh and curtailments uh are triggered which cause further hash rate drops. That's the cascade. So on a difficulty uh in proto that's in protocol time 20 days. These things aren't waiting 20 days. These are are responding continuously. Capital markets, power markets all respond continuously. Credit markets respond brutally and insurance responds when claims arrive. So these clocks are not aligned. The protocol is the slowest clock. there's a mismatch between uh the continuous response of the markets and the response of the protocol and they can be moving in opposite directions. So the uh protocol's observed hash rate and the market's observed risk are not correlated in that time and during that gap when they live when they're in opposite directions a 30% shock can become a 40% problem before the protocol catches up. So if I compress everything I said to three points, it's recovery speed. How fast can capacity return when the window opens? Capital halflife. How long can stressed operators survive before they're forced offline? And correlation failure. How much does does uh does the system share underlying exposure? These are not abstract quantities. They're specific, observable, and often priced quantities. Uh hash rate over break even, insurance premium changes, ASIC collateral haircuts, power optionality value, pool concentration, supply lead times. These are measures that forecast the postshock security. Notice that the aggregate hash rate is not in that list. When the aggregate hash rate moves, the other factors have already moved. So lastly, software is not just one layer. Uh hash rate is just one layer. Hash rate measures effort, but security is how the system recovers. Bitcoin solve the trust in software. It didn't solve physics, power transformers, semiconductors, and cooling. It didn't solve capital like debt structures, covenants, liquidity, and it didn't solve coordination, the willingness and ability of independent operators to remain online during times of stress. Those don't run on Bitcoin's clock. They fail independently and recover at different speeds. The network security question is not how much hash is there right now. It's what happens under stress and how quickly does it close the window. And the hash rate is the last signal to know. Thank you. >> Every year, this community comes together to celebrate, to debate, to build what comes next. And every year, the stage gets bigger. Sound money, center stage. So, where do you go to celebrate the next chapter in Bitcoin history? You come home. Nashville, July 2027.