Stock-to-Flow: The $100K Model That Broke

In June 2021 a pseudonymous analyst called PlanB posted a list of five numbers on Twitter: $47,000 for August, $43,000 for September, $63,000 for October, $98,000 for November and $135,000 for December. They were floors, the lowest prices he expected Bitcoin to close each month, and a large online following treated them as a timetable. By PlanB's own count, Bitcoin met the first three. On 30 November 2021 it was at about $57,000. On 31 December it closed at about $46,200.

The list came from a family of charts built on one idea, stock-to-flow, which for two years had been the most shared picture in crypto. Its collapse is a lesson in what happens when a pattern that fits the past is sold as a law of the future.

A chart built on scarcity

PlanB, who posts as @100trillionUSD, published "Modeling Bitcoin Value with Scarcity" in March 2019, with Bitcoin near $4,000. The idea came from commodities. Stock is everything already mined. Flow is what gets added each year. Divide one by the other and the ratio says how many years of current production it would take to double the supply. Gold scores about 60, which is part of why it holds value: no mine can flood the market.

Bitcoin's flow is fixed in code. New coins are paid to miners, the companies that run the computers securing the network, and roughly every four years the payment is cut in half in an event called the halving. It works like a gold mine that is legally required to halve its output on a set date. Each halving doubles Bitcoin's stock-to-flow ratio.

PlanB took Bitcoin's monthly prices from December 2009 to February 2019, 111 data points, plotted them against the ratio and found a tight fit. The model, he wrote, pointed to a market value of $1 trillion after the May 2020 halving, which translated to a price of about $55,000. In April 2020 he published a wider version, called S2FX, which compared Bitcoin with gold and silver and put the next phase at $288,000 per coin and a $5.5 trillion market value, which "could play out by 2024".

For a share trader the closest comparison is valuing a company only by how few new shares it issues, without looking at whether anyone wants to buy them.

The run that made it famous

At first it worked. Bitcoin passed $55,000 for the first time in February 2021, reaching about $58,350 that month. The coloured chart, with its rainbow of dots marching up a line, spread through Twitter and YouTube. "The model says $100K" became a slogan, and PlanB's monthly floor list in June 2021 turned it into a deadline.

Then November missed. Bitcoin set a record of $69,000 on 10 November 2021 and slid from there. PlanB called the November shortfall a "black swan" and an "outlier", as AMBCrypto and Cryptoslate reported, and separated the floor list from the main model. After December also missed, he posted that the stock-to-flow model was "intact" and set a new test: if Bitcoin stayed within one standard deviation of the model, the normal band around its line, for the next two and a half years, the model was still valid, according to FXStreet.

June 2022: the floor gives way

It took six months. On 18 June 2022 Bitcoin fell to about $17,600 and broke below the model's lower band for the first time, as FXStreet and Forklog reported. Three days later Vitalik Buterin, the co-founder of Ethereum, quote-tweeted Anthony Sassano, an Ethereum commentator who had called the model "such an epic failure". Buterin wrote that stock-to-flow was "really not looking good now", and that models giving people "a false sense of certainty" that the number will go up "are harmful and deserve all the mockery they get".

PlanB replied that the model "certainly had a good run" from March 2019, with Bitcoin at $4,000, to March 2022, with Bitcoin at $45,000, and that either Bitcoin was "extremely undervalued" or stock-to-flow would be "less useful in the future". After the FTX exchange collapsed, Bitcoin hit $15,476 on 21 November 2022, more than 85% below the $100,000 the model's fans had expected a year earlier.

The criticism had been there from the start. In June 2020 Nico Cordeiro of the trading firm Strix Leviathan called the model "fatally flawed", pointing out that gold's stock-to-flow ratio had stayed near 60 for more than a century while the value of gold reserves swung from $60 billion to $9 trillion. Analysts at the exchange Huobi argued that it ignored interest rates and the US Federal Reserve. The flaw is simple to state: supply is known years ahead, so if supply set the price, the price would already reflect it. What moves the price is demand, and the model did not measure demand at all.

Four years late and still waiting

Bitcoin did cross $100,000, on 5 December 2024, about three years after the December 2021 deadline, after the United States approved spot Bitcoin exchange-traded funds, stock-market funds that hold real Bitcoin, in January 2024. It peaked at $126,080 on 6 October 2025, according to CoinGecko.

That was still nowhere near the model. In November 2024, after the April 2024 halving, PlanB posted that stock-to-flow pointed to an average price of $500,000 for the 2024 to 2028 cycle, within a band of $250,000 to $1 million, as Benzinga and CryptoPotato reported. Bitcoin never reached the bottom of that band. It fell below $60,000 in mid-2026 and traded at about $82,000 on 16 October 2026, according to CoinGecko.

PlanB still posts on X and still defends the model, and some of his followers still share the chart. Many professional analysts had stopped using it by 2022, and it is now mostly cited as an example of overfitting: a formula tuned so closely to past data that it describes history perfectly and the future badly.