The Impossible Equation
We know exactly how many Bitcoin will ever exist. We know exactly what it costs to mine one today. We still cannot tell you what it is worth. Here is why that impossibility is the whole point.
Why read this (9 minute read)
On March 10, 2026, the Foundry USA pool mined block 939,999 and collected 3.125 BTC. Unremarkable block. Except it contained the 20 millionth Bitcoin ever created. Of the 21 million coins that will ever exist, 95.2% are now in circulation. The remaining 5% will take 114 years to produce.
Meanwhile the average industrial miner is losing $16,000 on every coin they dig out of the digital ground today. We have more precise supply data than for any monetary asset in history. We still cannot calculate fair value. This piece walks through exactly why, layer by layer, and why the impossibility is not a gap in our models. It is the most honest thing the asset has ever told us.
The workbench at 11pm
The garage light is off. The BitAxe hums on the workbench: 17 watts of angry plastic running a lottery the industrial operations will never play. My all-in cost for that solo ticket is laughably low. The big operations are not laughing. Average production cost sits at approximately $88,000 per coin. Spot price: $72,000. Every industrial block mined right now bleeds roughly $16,000. The average miner is running a 21% operating loss on every coin produced.
This is not unprecedented. Bitcoin traded below production cost in 2019 and again in 2022. Both times the protocol self-corrected: rigs shut off, difficulty fell, survivors kept the chain alive, price eventually recovered. The protocol does not negotiate with miners. It adjusts every 2,016 blocks regardless of what any balance sheet says.
But here is the question nobody has cleanly answered in 17 years of trying: if we know the supply schedule perfectly, and we know the production cost precisely, why can nobody tell us what one Bitcoin is actually worth?
Five reasons. Each one compounds the last.
One: cost is a floor, not a ceiling
Production cost is the supply side’s pain threshold. Nothing more. It tells us when capitulation happens. It says nothing about demand.
Think of it this way. A barrel of oil has a known production cost. That cost does not tell you what oil is worth when a war closes the Strait of Hormuz. Bitcoin converts real-world electricity through proof-of-work into verifiable digital scarcity. The thermodynamic input is measurable down to the kilowatt-hour. The value of what that thermodynamic input produces is set entirely by what people on the demand side believe it is worth, which is a human and political question, not a physics question.
Every analyst who has anchored a Bitcoin price target to mining cost has eventually been humiliated by the market, in both directions. Cost is useful. It is not a valuation model. It is the first reason fair price is impossible to calculate, because it only closes half the equation.
Two: you are measuring with a broken ruler
Every Bitcoin price you have ever seen is quoted in fiat. That denominator shrinks by design, every year, in every major currency simultaneously.
When Bitcoin moves from $10,000 to $72,000, two things are happening at once: Bitcoin is appreciating and the dollar is depreciating. Most price models do not separate these. They are built in dollars, measured in dollars, and calibrated against historical dollar prices. Asking what Bitcoin’s fair price is in fiat is like measuring a building with a ruler that loses 2% of its length every year. You can get a number. It is not the number you think it is.
Bitcoin’s new supply shrinks with every halving. Fiat supply grows with every policy cycle. The pressure this creates is real and persistent. The exact speed of translation into price depends on how fast institutional adoption accelerates, how fast fiat trust erodes, and how the two interact across different geographies simultaneously. No model has reliably forecast any of those three variables. The broken ruler is the second reason fair price is incalculable. You cannot price something accurately in a unit that is itself losing value on an unpredictable schedule.
Three: history ran this experiment and we ignored the results
One thousand years ago the Song Dynasty invented paper money. Private bankers issued notes called jiaozi backed by iron coins. The government took over in 1023 and printed beyond its reserves. Inflation followed. The Yuan Dynasty doubled down, banned metal coins entirely, and printed until their notes were nearly worthless. By the time the Ming Dynasty took power, paper money was so discredited they abolished it completely in 1455.
The lesson is not that paper money always fails overnight. It lasted centuries. The lesson is that every monetary system built on expandable supply has eventually been expanded past the point of trust. Every single one. Without exception. Across every culture and every era that tried it.
Bitcoin is the first monetary network engineered so that no emperor, no central bank, no emergency decree, and no committee vote can expand its supply. The 21 million cap is not a policy. It is mathematics enforced by every node simultaneously. History makes the long-term directional thesis as clear as any thesis gets. It still cannot tell you the price. The timing between “this cannot continue” and “this has ended” is the one variable no model in history has ever predicted accurately. The Yuan Dynasty printed for decades before collapse. History is the third reason fair price is impossible: it confirms the direction without clocking the arrival.
Four: the supply schedule is perfectly known and therefore prices in nothing
On March 10, 2026, the Foundry USA mining pool added block 939,999 and collected 3.125 BTC. That block contained the 20 millionth Bitcoin ever created. There are now fewer than 1 million Bitcoin left to mine, ever. Paybis
Bitcoin’s issuance schedule was written in code at the network’s inception and has followed that schedule with clockwork precision ever since. BytePower X Seventeen years. Zero deviations. Not one missed adjustment, not one corrupted halving, not one supply rule bent for any government or any price.
An estimated 2.3 to 3.7 million BTC are permanently inaccessible due to lost private keys, forgotten passwords, and deceased holders. BeInCrypto Add Strategy’s 766,970 BTC in cold storage with no sell mandate, growing ETF custody, and long-term holder accumulation at multi-year highs, and the effective liquid supply is substantially tighter than the nominal circulating figure. Miners now hold just 0.5% of circulating supply, compared to Strategy alone at seven times that amount. Decrypt
The next halving lands around April 2028. Block reward drops to 1.5625 BTC. Daily new supply halves again from 450 coins to 225. Many publicly traded miners are expected to exit Bitcoin mining entirely by 2027 and 2028, pivoting to AI and high-performance computing for more predictable revenue. Decrypt
Here is the paradox that makes supply analysis alone useless as a pricing tool: because the schedule is perfectly known and publicly visible, it is already priced in. Every market participant who is going to act on the next halving has had years to position. The supply curve is the most precisely known fact about any monetary asset in history. That precision does not help you calculate fair value. It means the supply side contributes zero surprise to the price. Everything is demand. And demand is where the models go to die.
Five: demand has no precedent, no model, and now new wildcards
Metcalfe’s Law values Bitcoin by the square of active users. Stock-to-Flow bets purely on scarcity ratios. Power Law traces long-term adoption curves. MVRV Z-Score, Puell Multiple, and Realized Price bands try to show when Bitcoin is statistically cheap or expensive relative to its own history.
They all work. Until they don’t.
Stock-to-Flow predicted $100,000 by end of 2021 and roughly arrived there. Then predicted continued acceleration that stalled. MVRV flagged the 2022 bottom correctly and flagged the 2025 top late. Every model captures some version of the past and fails at some version of the future, because adoption does not follow a smooth curve. It moves in waves driven by narratives, regulatory events, geopolitical shocks, and generational wealth transfer. None of those inputs are quantifiable from on-chain data.
Now layer on what the post-ETF market has done to short-term price behaviour.
Since spot Bitcoin ETFs launched, institutions gained clean, regulated access to futures, basis trades, and options. Major market makers run large options books. When those books accumulate significant open interest, negative gamma creates mechanical pressure: dealers sell into weakness and buy into strength, amplifying intraday moves well beyond what supply-demand fundamentals justify. This is not manipulation. It is the mathematics of hedging large derivatives books. The effect is the same either way: short-term Bitcoin price has become more susceptible to institutional flow dynamics than at any prior point in its history.
The old four-year halving cycle is evolving into something driven more by ETF inflows and macro liquidity than by supply shocks. CoinDesk Media narratives accelerate this: FUD during dips moves ETF outflows faster than any on-chain metric ever could. Euphoric coverage during rallies does the reverse. The signal-to-noise ratio has not improved with institutional participation. It has gotten noisier at a faster pace.
And the final wildcard: most of humanity still cannot explain what Bitcoin actually is. Not the price. The thing itself. A non-sovereign, thermodynamically scarce, globally accessible monetary network whose supply policy is enforced by mathematics rather than institutions. For most people it remains a number on a screen. The transition from that perception to genuine understanding, at the scale of billions of people across dozens of regulatory environments over decades, is the demand variable that determines Bitcoin’s generational price trajectory. No model can clock that transition. Nothing like it has ever happened before.
Why the impossibility is the point
Stack the five layers together.
Production cost closes the supply side’s pain threshold but says nothing about value. The fiat denominator is broken so the measurement itself is unreliable. History confirms the directional thesis but cannot time it. The supply schedule is perfectly known and therefore already priced. The demand side has no historical precedent and defeats every model eventually.
This is not a failure of analysis. It is an accurate description of what Bitcoin is: the first asset in history that sits at the intersection of a fixed, perfectly known supply and a demand function driven by human psychology, geopolitical disruption, monetary distrust, technological access, and generational wealth transfer simultaneously. No single model holds all five variables. No model ever will.
The miners bleeding $16,000 per coin right now are not making a pricing mistake. They are making a specific bet: that the demand variables will eventually overwhelm the cost pressure, as they have after every prior period when production cost exceeded spot price. The protocol does not validate or invalidate that bet in real time. It adjusts difficulty every 2,016 blocks, keeps blocks landing every ten minutes, and moves on.
Over the last decade, Bitcoin’s value has climbed more than 15,000%. Fortune That return was produced by an asset whose supply schedule was publicly known the entire time, whose production cost was trackable the entire time, and whose fair value no model ever calculated correctly in advance. The return came not from modelling the price but from understanding the asset well enough to hold it through every period when the models said it was broken.
The 20 millionth coin has been mined. One million remain. The average miner is losing $16,000 per block. The denominator is broken. History has run the experiment before. The demand side has no precedent.
The BitAxe hums. The difficulty adjusts. The protocol does not wait for a consensus on fair value.
Neither should you.
Bitcoin 🍊 runs on signal. So does the one100milsats newsletter. If this piece added value, stack a few sats my way:
⚡ Lightning Address: one100milsats@blink.sv
Any amount. No middlemen. Instant.
Notes
Production cost figures ($88,000 all-in average, approximately $16,000 to $19,000 per-coin loss at current prices) sourced from Checkonchain difficulty regression model, CoinDesk March 22, 2026, Blocklr March 2026, and pbxscience.com March 2026. Bitcoin price approximately $72,000 as of April 10, 2026 per Fortune and Yahoo Finance. 20 millionth Bitcoin mined March 10, 2026 at block 939,999 by Foundry USA confirmed by on-chain data, reported by Fortune, Decrypt, BeInCrypto. Lost coin estimates 2.3 to 3.7 million BTC from Chainalysis and River Financial. Strategy 766,970 BTC holdings and miner 0.5% supply share from Decrypt March 2026. Miner pivot to AI from Decrypt analysis of Needham and Company research. Song Dynasty monetary history from standard monetary history sources. ETF gamma hedging mechanics from 2025 to 2026 institutional market commentary. All analysis is educational content only. Not financial advice.