The Bitcoin “miners are dumping” story is comforting, as simple stories always are. Prices fall, miners run out of oxygen, coins hit the exchanges, and prices are pushed up by one simple villain.
But miners are not a single actor, and sales pressure is not just a mood. It’s math, contracts, deadlines. When stress appears, what matters is not whether miners want to sell, but whether they need to sell, and how much they can sell without disrupting the business they are trying to keep alive.
That’s why the most useful way to think about miner “surrender” is as a thought experiment. Imagine you are now operating a mine in a market where the hashrate ribbon has tumbled into reversal territory and prices are below a rough estimate of the average all-in maintenance cost on a difficulty basis of about $90,000.
At the same time, the total holdings of miners has reached about 50,000 BTC, which is not small, but not bottomless either.
Now you have a simple question that sounds dramatic. If the price stays below the average AISC line for a while, how many coins can you push out in 30-90 days before lenders, power contracts, and your own business practices rebound?
AISC is a moving target, not a single number
All-in sustaining cost (AISC) is a term borrowed from crypto mining and commodities, but it’s been retained to discourage pretending that your electricity bill is your only bill. AISC is basically a number that determines whether or not you can stay in business. It’s not “Can we keep our machines running today?” it’s “Can we keep our operations healthy enough to last into the next quarter?”
Even though different research institutions draw the boundaries differently, AISC for Bitcoin miners can be thought of as having three tiers.
The first layer is the direct operating cash costs that everyone understands. Because at the heart of it all is electricity, and the meter moves whether you’re in a bullish mood or not. Add hosting fees (if you don’t own your own site), repairs, pool fees, network operations, and staff to maintain your facility so it doesn’t turn into an expensive heating appliance.
The second layer is the layer that the meme skips over: capital investment maintenance. This is not a capital investment for growth. Capital maintenance is the money you spend to keep your vehicle from slowly failing. Fans will fail, hashboards will deteriorate, containers will rust, and more importantly, your network will become more robust. Even if your machine is fine, you can lose your share of the pie if you don’t upgrade while others do.
The difficulty level has something to do with it. Bitcoin adjusts the difficulty of mining, so blocks continue to arrive more or less on time. As the hashrate increases, the difficulty gradually increases and the same machine will earn fewer BTC for the same amount of energy.
As the hashrate decreases, the difficulty level eases and the remaining miners have a slightly better bite. The AISC framing we use is explicitly based on difficulty and is a clean way to capture this moving target without requiring dedicated power contracts for every miner.
The third layer is what transforms stress into forced action: corporate costs and financing. Private operators may be primarily concerned with power and maintenance. Debt-laden public miners worry about interest payments, covenants, liquidity buffers, and the ability to refinance.
This is why AISC changes over time and why single number arguments feel foolish. It changes as the difficulty changes or as the composition of the fleet changes (old machines are kicked out and new ones come in).
This will change as the power environment changes, especially for miners who are exposed to spot pricing. It also changes as the cost of capital changes. So, for the same hash output, a miner may appear stable at one point in the cycle and vulnerable at another.
So even if the price drops below AISC’s average estimated price (around $90,000), it doesn’t mean the entire network is immediately underwater, it just makes the center of gravity uncomfortable. Some miners are doing well, others are in trouble, and some are in triage. Stress is certainly there, but there is variation in responses, and that variation prevents the default outcome of “everyone dumping at once.”
There is another reason why the default result is not a dump. In addition to selling BTC, miners can shut down marginal machines, reduce grid payments, roll hedges, and renegotiate hosting terms. And, as mentioned before, crypto slateCurrently, many people are doing side jobs related to AI data centers, which can buffer them from the bad months of mining.
Now the real question arises. It’s about how much sales do you need to do structurally during times of stress?
Dump math: What can you sell without disrupting your business?
Start with one flow that the protocol offers, whether you’re happy with it or not. After the halving, new BTC issuance through block subsidies will be approximately 450 BTC per day, or approximately 13,500 BTC per month.
If miners sell 100% of their new issuance, that is a clean ceiling for flow sales. In reality, miners are not coordinated and everyone does not have to sell everything they mine. However, as a thought experiment, the maximum new supply that can be brought to market without touching existing inventory is 450 BTC/day.
Bring in your inventory now. Because that’s what the scary headlines say. It depends on Glassnode’s estimate that miners have around 50,000 BTC on hand. A stockpile of 50,000 BTC sounds like a lot until you convert it into a time series. Over a 60-day period, 10% of the inventory would be 5,000 BTC, or approximately 83 BTC per day. Spread over 90 days, 30% is 15,000 BTC, or approximately 167 BTC per day.
This is the basic form of minor forced distribution in a stress window. So unless the stress is severe enough for inventory to become your primary tool, flow sales will do most of the work, and inventory sales will add a small but still meaningful amount.
So let’s set up three price paths under the toy model: $90,000, $80,000, and $70,000. We then relate them to three intermediate regimes that map how miners behave when margins thin.
In the basic case, miners sell half of the issued amount and do not touch the inventory. That’s 225 BTC/day. In 60 days, the total issuance of 13,500 BTC is multiplied by 50%, resulting in 6,750 BTC. 10,125 BTC in 90 days.
In a conservative stress case, miners sell 100% of their issued amount and also sell 10% of their inventory in 60 days. This is a total of 450 BTC/day from issuance and 83 BTC/day from inventory, approximately 533 BTC/day.
In case of severe stress, miners will sell 100% of their issued amount and 30% of their inventory in 90 days. This is 450 plus 167, or about 617 BTC per day.
These are upper bound sketches that answer the narrower question of what the market will tolerate.
To understand how much attention the market will pay, we will use the simplest comparison tool that our readers already understand: ETF flow days measured in BTC equivalents. ETF outflows amount to only about 2.5% of BTC-denominated assets under management, or about $4.5 billion. crypto slate I previously described them as being more technical than belief-driven. You don’t even have to litigate the motivation for using comparisons, because it’s the scale that matters.
At $90,000 per coin, a day of $100 million is equivalent to approximately 1,111 BTC. For $80,000, it is 1,250 BTC. Assuming $70,000, it would be approximately 1,429 BTC. Suddenly, the minor numbers look less like monsters under the bed and more like something that can be put on the same shelf as the flows that the market is constantly digesting.
A strict miner distribution diagram, say 600 BTC/day, would equate to about half a day of $90,000 for a $100 million ETF. If it’s thrown into a thin book, or weekend liquidity is weak, or concentrated in a few ugly hours, the price can still fluctuate. But the pushy narrative of miners flooding the market hits two ceilings: issuance and a finite inventory that miners are willing to liquidate.
There are also execution details that are more important than people want. Many of the miners’ sales are not market orders incorporated into the public order book. It can be routed through an OTC desk, structured as a forward sale, or handled as part of broader financial management. That doesn’t eliminate the selling pressure, but it does change how it prints on the tape. When people expect a visible waterfall and it slows down, the impact on the market weakens.
So what takes this from an orderly drip to something even uglier? It certainly involves more than just lowering the price below ASICs. The trigger is when the finance layer takes over decision making. Inventory changes from optional to necessary when miners need to adhere to minimum liquidity, meet collateral requirements, or deal with refinancing barriers in bad market conditions.
That’s the sobering answer to the viral question. Even under stress and the ribbon flipping, there is a real limit to how much a miner can dump in a month or quarter. If you want a practical upper limit, the thought experiment keeps pulling you back into the same zone. A few hundred BTC per day during mild stress, and around 500-650 BTC per day during severe stress windows with stock taps. The exact number depends on power requirements and debt constraints that can be incorporated later.
And if you’re trying to guess what makes the tape move, the punchline is frustratingly unromantic. Markets tend to value rhythm, venue, and surrounding fluidity more than sellers’ narrative labels. Miners can add weight to down weeks, but the idea that there is an infinite trapdoor beneath the price does not survive when it comes into contact with balance sheets.