Stock-to-Flow’s Last Window: Can S2F Still Be Right by 2028?

On August 12, 2026, Bitcoin changed hands near $64,000, roughly 49 percent below the record of $126,080 it set on October 6, 2025, according to CoinGecko. For a volatile asset in the back half of a halving cycle, that is an unremarkable number. For the most famous price model Bitcoin has ever produced, it is a quiet emergency.

The stock-to-flow model, known to almost everyone in crypto as S2F, still implies a Bitcoin worth several hundred thousand dollars this cycle. Most takedowns of the model look backward and tally the misses. This one runs the model’s own clock forward. Bitcoin is now past the midpoint of the 2024 to 2028 cycle, which means S2F has one clean, testable window left before the next halving resets the math. The question is not only why the model has missed, but whether the arithmetic leaves it any path at all to be right.

A Model That Refuses to Die

Few forecasting tools in finance have been declared dead as often as stock-to-flow, and fewer still keep getting quoted after each obituary. It survives because it is simple, it is visual, and it is relentlessly bullish. A single rising line promises that scarcity alone will carry Bitcoin to six and then seven figures, and that is a comfortable story to hold through a drawdown. When the price sags, the model does not; the line just keeps climbing off to the upper right, waiting for reality to catch up.

The problem is that comfort is not the same as accuracy. Before burying the model one more time, it is worth doing something its critics rarely bother with: taking it completely seriously, plugging today’s numbers into its own published equation, and seeing what it actually demands of the next twenty months. Only then can we ask whether those demands are remotely plausible, or whether the most-shared chart in Bitcoin’s history has quietly run out of road.

What Stock-to-Flow Actually Measures

Stock-to-flow is borrowed from commodities. Stock is the total amount of an asset already in existence; flow is the amount produced each year. Divide one by the other and you get a scarcity ratio: how many years of current production it would take to reproduce the existing supply. Gold scores high because miners add only a sliver to the aboveground hoard each year. Consumables like copper or wheat score low, because production turns over quickly relative to what is stored.

Bitcoin’s ratio is unusual because it steps up on a fixed schedule. Every 210,000 blocks, roughly every four years, the block subsidy halves, so the flow drops while the stock keeps climbing. According to CoinGecko’s own primer, gold sits near 62 and silver near 22, while Bitcoin jumped from about 27 before the 2020 halving to about 119 after the April 2024 halving. On this one measure, Bitcoin is now scarcer than gold. The table below lines them up.

The intuition is easy: scarcer things tend to hold value better. PlanB’s leap was to claim the ratio does not merely correlate with value, it sets it, and can be turned into a precise dollar price with a single equation. That leap is where the trouble starts.

From a 2019 Blog Post to a Price Religion

The model arrived on March 22, 2019, in a post titled Modeling Bitcoin’s Value with Scarcity, written by a pseudonymous analyst who goes by PlanB and the handle @100trillionUSD and describes himself as a Dutch institutional investor. His original article fit a straight line through the logarithms of Bitcoin’s market value and its stock-to-flow ratio, producing the now-infamous equation: the natural log of market value equals 3.3 times the natural log of the ratio, plus 14.6. He reported an R-squared near 95 percent, a headline statistic that made the relationship look almost deterministic.

A year later, in April 2020, PlanB published a cross-asset version he called S2FX, which folded gold and silver into the same framework and produced a specific forecast: an average Bitcoin price around $288,000 for the 2020 to 2024 cycle. The chart, with its stair-step model line and rainbow of monthly price dots, became one of the most shared images in the entire asset class. It was less a spreadsheet than a banner, and it turned a regression into a rallying cry. Bitcoin never averaged anything close to $288,000 during that cycle; it peaked near $69,000 in November 2021 and then bottomed under $16,000. But by then the model had outgrown its own results.

Running the Equation Forward: What S2F Demands in 2026

Here is where a forward-looking test gets uncomfortable for the model. The equation is public, and so is the current ratio, so anyone can run it. Feed the post-2024 stock-to-flow figure of about 119 into PlanB’s own 2019 formula and it returns a market value north of $15 trillion, which works out to roughly $770,000 per coin at today’s circulating supply. That is not a strawman built by critics; it is the model doing exactly what it was designed to do, with its author’s own coefficients.

PlanB frames the current cycle a little more modestly than his raw equation. In a widely quoted post captured by Yahoo Finance in March 2026, he wrote that with Bitcoin near $67,000, the model still “screams $500,000 average this cycle,” meaning 2024 to 2028, with the broader framework implying a range from $250,000 to $1 million. So even on the model-maker’s own more cautious framing, the current-cycle target is half a million dollars a coin.

Set those numbers next to reality. Bitcoin traded near $64,000 in mid-August 2026. Even against the conservative $500,000 figure, that is a shortfall of roughly 87 percent. Against the raw output of the original equation, the miss is larger still. The model is not slightly early or a little optimistic; it is off by nearly an order of magnitude, at the exact point in the cycle where its supporters expected the payoff to be arriving.

The Falsification Clock

A price target only means something with a deadline, so let us give S2F the fairest one available: the next halving, expected around April 2028. Bitcoin’s fourth halving landed on April 20, 2024, which puts August 2026 about twenty-eight months into a roughly forty-eight-month cycle, a little past the midpoint. That leaves close to twenty months for the model to come good before the scarcity ratio ratchets up again and the goalposts move.

The arithmetic is unforgiving. Climbing from $64,000 to the $500,000 average target means rising nearly eightfold. Spread across twenty months, that requires compounding at roughly 11 percent every single month, without pause, an annualized pace above 240 percent sustained for almost two years. Bitcoin has produced individual months like that. It has never strung twenty of them together without a serious drawdown, and this calculation allows for none. Reaching even the low end of the range, $250,000, still demands close to 7 percent a month, every month, from here.

The cycle-average framing is harsher yet. PlanB has always described the model value as an average across the whole cycle, not a fleeting spike on one lucky day. Because the cycle’s first twenty-eight months averaged well under six figures, lifting the four-year average to $500,000 would require Bitcoin to average more than $1 million across every one of the roughly twenty months that remain: not tag that level briefly, but sit at it. The table below shows what each version of the target now asks of the next twenty months.

None of these paths is strictly impossible; crypto has humbled everyone who has ever used that word. But they range from historically unprecedented to arithmetically absurd, and the window narrows with every month that Bitcoin trades in the sixties. This is the heart of the matter: the model is not just wrong today, it is running out of time to become right on any reading of its own numbers.

The Tautology at the Core

Run the model forward and it fails on the numbers. Look inside it and you find a deeper problem, one that was there from the very first post. Market value is, by definition, price multiplied by supply. Stock-to-flow is built from stock, which is just supply again. So when the model regresses market value on stock-to-flow, supply is sitting on both sides of the equation. As Bitcoin Magazine has laid out, that makes the celebrated fit partly circular: you are, in large part, correlating supply with a quantity that already contains supply.

The 95 percent R-squared has the same weakness. Bitcoin’s supply grows on a smooth, pre-programmed curve, and its price trended up over the same span of years, so any regression of one on the other will look tight almost by construction. Cory Klippsten, among others, has pointed out that once you adjust for the autocorrelation baked into trending data, the model’s genuine explanatory power collapses toward zero. A number that was presented as near-proof turns out to be an artifact of the method, not evidence of a law.

Spurious by Construction

Statisticians have a name for this trap: spurious regression. Take any two series that both drift upward over time, plot one against the other, and you will usually find a strong correlation even when there is no real link between them. The classic teaching examples pair things like ice cream sales and drowning deaths, which move together only because both rise in summer. Bitcoin’s price and its scarcity ratio both climbed through the 2010s, and that shared trend is enough to manufacture an impressive-looking chart without any causal glue holding it together.

One of the earliest and sharpest versions of this critique came from Nico Cordeiro, chief investment officer at the research firm Strix Leviathan, in a July 2020 report. He called S2F a “chameleon” built on shaky assumptions and argued, in comments reported by Cointelegraph, that its accuracy would likely prove about as reliable as “the astrological models of the past” were at predicting financial outcomes. His bluntest line was that the paper was not proper empirical analysis but something closer to “a marketing piece.” The core objection is simple and has never really been answered: PlanB never demonstrated that a commodity’s market value is caused by its supply rate. He fit a curve that assumed it, then presented the fit as if it were the cause.

Scarcity Is Only Half of a Price

Every price in every market is set where supply meets demand. S2F models only the supply half, and Bitcoin happens to be the one asset where that half is already completely settled. The issuance schedule is fixed in code; miners will produce the same coins whether the price is $16,000 or $160,000. What actually moves the price is the other half of the equation, the part S2F leaves out entirely: demand. A model that stares only at supply is, in effect, watching the one lever nobody can pull.

The last two years made that impossible to ignore. When the SEC approved spot Bitcoin exchange-traded products in January 2024, it handed traditional money a straightforward way in, and inflows, not any change in scarcity, drove the run to the October 2025 record. Then, across stretches of 2026, that same channel ran in reverse, and a supposedly ultra-scarce asset fell anyway, because the marginal buyer had stepped back. Scarcity did not change during the decline; demand did. Readers weighing the rest of this year can see the demand side laid out in our guide to the September FOMC and the packed regulatory calendar, alongside the rulebooks now shaping institutional flows on both sides of the Atlantic, from the SEC to Europe’s MiCA regime. None of those forces appears anywhere in the stock-to-flow equation, and all of them have moved the price more than the halving ever did.

2021 and the Broken Promise

The model’s credibility problem is not only statistical; it is also about how its author handled being wrong. During the 2021 bull market, PlanB promoted a floor model, a version that traced a rising path of supposed price floors running from the mid-five figures early in the year toward six figures by December. On June 23, 2021, he made it concrete and testable, stating that he would call S2F invalidated if Bitcoin had not reached $100,000 by December of that year. It was, to his credit, a real falsification condition, exactly the kind most forecasters avoid.

Bitcoin never came close. It spent the back half of the year sliding and closed 2021 well under $50,000, never once touching six figures. When the self-imposed deadline arrived, PlanB did not honor the pledge. As Protos documented, he brushed the invalidation off, posting that he actually liked being at the lower bands. A model that cannot be falsified even by its own creator’s stated test has quietly stopped being a forecast. It has become a belief, and beliefs do not have deadlines.

The Mockery and the Defense

By mid-2022 the gap had become a spectacle. On June 18, 2022, the model implied a Bitcoin price of $67,175 while the asset traded below $19,000. Ethereum co-founder Vitalik Buterin weighed in publicly, writing that “financial models that give people a false sense of certainty and predestination that number-will-go-up are harmful and deserve all the mockery they get,” as reported by The Block. It was a widely shared verdict from one of the few people in crypto with the standing to deliver it, and it captured a real hazard: a chart that promises the number can only go up teaches people to size their risk as if losses were impossible.

Not everyone agreed, and fairness demands the other side. Blockstream chief executive Adam Back, a cited influence on Bitcoin’s original design, pushed back on Buterin and defended the underlying scarcity thesis, arguing that the long-run case for a supply-capped asset stayed intact even when the short-run chart wandered. That is the strongest steel version of the S2F position, and it contains a genuine truth: Bitcoin’s fixed cap is unusual and does matter. The leap the model makes, and the one Back’s defense does not rescue, is the jump from “scarcity matters” to “scarcity sets this exact price on this exact date.” The first is defensible. The second is what keeps failing.

The Rivals Waiting in the Wings

S2F is not the only single-line model competing for attention, and its cousins share its family weakness. The power law model, associated with physicist Giovanni Santostasi, fits Bitcoin’s price to time raised to a power and has tracked long-run troughs better than S2F has; but its defining exponent shifts depending on which start date you choose, which undercuts the claim that it captures a genuine law of nature rather than a flattering curve fit. The global liquidity thesis, popularized by former hedge fund manager Raoul Pal, argues that Bitcoin mostly follows the world’s money supply; yet liquidity expanded through parts of 2026 while Bitcoin went sideways, straining the correlation the thesis leans on.

The pattern is consistent across all of them. Any model that explains a reflexive, sentiment-driven asset with one variable will look brilliant on the sample it was fit to and turn fragile the moment fresh data arrives. That is not a knock on scarcity, time, or liquidity as real factors; each one plausibly matters. It is a warning about mistaking a tidy back-fit for a forecasting engine. The map is not the territory, and a smooth line drawn through the past is not a promise about the future.

What Bitcoin Actually Shipped

Step away from the price line and Bitcoin’s actual decade looks very different from its forecasting culture. While S2F chased dollar targets, the protocol did unglamorous, verifiable work. The Taproot upgrade and its Schnorr signatures moved from novelty toward normal use, a shift we track in our Taproot scorecard; the mining industry professionalized into a capital-intensive business whose economics you can actually audit, walked through in our field guide to miner margins. Those are things that either happened or did not, and they can be measured without a regression.

This matters because it points to a healthier way to judge the asset. You can measure adoption, hash rate, fee revenue, custody arrangements, and liquidity, all of which are real and improving unevenly. What you cannot do responsibly is convert a scarcity ratio into a $500,000 price and treat the output as destiny. The strongest case for Bitcoin was never a stock-to-flow chart; it was the boring, remarkable fact that the network keeps running exactly as specified, block after block, through every crash the price models failed to see coming.

What This Means for US Investors

For readers in the United States, the practical takeaway is about hygiene, not doom. The SEC has never endorsed a Bitcoin price target of any kind; when it cleared spot products in January 2024, then-chair Gary Gensler went out of his way to call Bitcoin a speculative, volatile asset and to stress that approval was not an endorsement. Any chart promising $500,000 or $1 million on a fixed schedule should be read as marketing, and often as someone else’s marketing, rather than as analysis you can lean on.

That does not mean scarcity is meaningless. Bitcoin’s capped supply is a real, unusual property, and over long horizons it plausibly supports the asset. But it does not set a 2026 price, and it certainly does not set one to the dollar. The sober posture is to size positions for a wide range of outcomes, to watch the demand-side signals that actually move markets, and to treat any model that has already failed its own creator’s stated test as a study of human conviction rather than a map of the future. Stock-to-flow is worth understanding. As of August 2026, with the clock past the midpoint and the required returns bordering on fantasy, it is not worth trusting.

Frequently Asked Questions

Has the Bitcoin stock-to-flow model failed?

By almost any measure, yes. The model projected an average price near $500,000 for the 2024 to 2028 cycle, yet Bitcoin traded around $64,000 in August 2026, more than 80 percent below that figure. It also missed the 2021 top, the 2022 crash below $16,000, and the 2024 to 2026 range. Supporters keep it alive mainly by stretching its timeline, which is exactly what makes it so hard to test.

What price does the stock-to-flow model predict for Bitcoin in 2026?

PlanB has said the model implies an average near $500,000 for the current cycle, with a range often cited between $250,000 and $1 million. Feed the current post-halving scarcity ratio into his original 2019 equation and it returns an even higher number, above $770,000 per coin. Actual prices have sat near $64,000, so the gap is enormous no matter which version you use.

Who is PlanB, the creator of the stock-to-flow model?

PlanB is the pseudonym of an analyst who says he is a Dutch institutional investor. He published the stock-to-flow model in March 2019 under the handle @100trillionUSD and followed it with a cross-asset version, S2FX, in April 2020. He has remained anonymous throughout, and his real identity has never been publicly confirmed.

Why does stock-to-flow ignore demand?

The model uses a single input, the ratio of existing supply to new issuance, and treats everything else as fixed. Price, however, is set where supply meets demand, and Bitcoin’s supply schedule is locked in code no matter what buyers do. That is why demand shocks like spot ETF flows, interest rates, and regulation move the price sharply while the scarcity ratio barely changes.

What models do analysts use instead of stock-to-flow?

Serious desks lean on demand-side and macro tools: net flows into spot Bitcoin ETFs, realized cap and on-chain cost basis, global liquidity measures, and options positioning. Some still watch scarcity-based charts like the power law, but most now treat any single-variable price model as a rough guide to sentiment rather than a genuine forecast.

Written by Priya Reddy, Editorial Lead at HOGE Wire, covering Bitcoin markets and the models people build to predict them.