Power Law vs. Stock-to-Flow: Which Bitcoin Model Holds Up?
Power Law vs. stock-to-flow, data-checked: how both models calculate, what they have predicted since 2019 and which one you can trust today.
Power Law or stock-to-flow: which model holds up? The Power Law. Up to 2019, both models fit the past almost equally well. After that they parted ways: stock-to-flow puts bitcoin today at about $846’000, more than eleven times the price. The Power Law puts fair value at about $108’000, and the price is moving within its corridor.
Why you have to choose between two models
If you hold bitcoin for years, you want to know: is the price high or low today? When does it get tight, when is it an opportunity? For that you need a yardstick; otherwise the mood decides.
Two models became known for this: stock-to-flow, which made headlines in 2019 and 2021 with targets of $100’000 and more, and the Power Law. Both can't be right at the same time. For early 2028, the prices they expect are almost five times apart.
What is the stock-to-flow model?
"Stock" is the existing supply: all bitcoin that already exist. "Flow" is the inflow: the bitcoin newly mined in a year. Divide the stock by the flow and you know how many years today's production would need to create the whole stock again.
The ratio comes from the world of commodities. Gold has a stock-to-flow of about 61, bitcoin about 122 since the halving of April 2024. By this measure, bitcoin is scarcer than gold today. Saifedean Ammous made it known in 2018 in his book "The Bitcoin Standard".
On March 22, 2019, the pseudonymous analyst PlanB turned it into a price model. He compared stock-to-flow with bitcoin's market value and found a close relationship: the scarcer bitcoin, the higher its value. Because the flow halves with every halving, the stock-to-flow doubles every four years. The model therefore expects a jump in price after every halving.
The 2019 stock-to-flow formula
Stock-to-flow (SF) = stock ÷ flow per year
Market value in US dollars = e^14.6 × SF^3.3
Model price = market value ÷ stock
We count the flow as the bitcoin mined in the preceding twelve months, with the halvings at their actual dates and 144 blocks per day in between. On Nov 7, 2026: stock about 20.1 million, flow about 164’000, SF = 122. That gives a market value of about $17 trillion and a model price of about $846’000.
Why about tenfold per halving: if SF doubles, the market value grows by 2^3.3 ≈ 9.85.
What is the bitcoin Power Law?
The Power Law does without scarcity. It takes only the time since the genesis block on January 3, 2009 and finds that bitcoin's price has grown roughly with a fixed power of that time. That gives the Power Law line. At 42 % of this line lies the floor, where almost every crash so far has ended, and above the floor the trend as fair value.
The core idea: the trend keeps rising, but by a smaller percentage every year. Today it is about 29% a year, about 2.6 times over four years. How the model works in detail is explained in the basics article The bitcoin Power Law explained.
The Power Law formula
Power Law line in US dollars = 10^−16.493 × days^5.688
Floor = 0.42 × Power Law line
Top band = floor × e^(3.56 ÷ (1 + (days ÷ 4692)³))
Trend = the lower of the Power Law and √(floor × top band)
"Days" counts from January 3, 2009. On Nov 7, 2026 (6’517 days) this gives a Power Law line of about $158’970, a floor of about $66’770, a top band of about $175’790 and a trend of about $108’340. All parameters and their derivation are in the methodology.
Power Law vs. stock-to-flow: the differences at a glance
| Stock-to-flow | Power Law | |
|---|---|---|
| What it derives the price from | Scarcity: stock ÷ new bitcoin | Time since the genesis block |
| Published | March 2019 (PlanB) | 2018 (Giovanni Santostasi), corridor 2019 (Harold Burger) |
| Role of the halving | Drives the price, a jump every four years | Plays no role |
| Expected growth | About ×10 per halving, without end | A little slower every year |
| Lower bound | None | Floor at 0.42 × Power Law line |
| Model price on Nov 7, 2026 | about $846’000 | trend about $108’000, floor about $67’000 |
| Price on Nov 7, 2026 relative to model | 0.09 × model | 0.70 × trend |
The chart: both models since 2010
The chart shows the daily close since July 2010 on a logarithmic scale, so every tenfold rise has the same height (how to read a log chart). Along with it, the stock-to-flow line, the Power Law trend and the floor. The vertical line marks March 22, 2019, the day stock-to-flow was published. Move your mouse or finger over it to see a day's values.
- Price
- Stock-to-flow
- Trend
- Floor
Left of the vertical line, both models lie close together and the price fluctuates around both. To the right of it, stock-to-flow jumps up about tenfold after each of the 2020 and 2024 halvings. The price follows only the first jump, and only for a few months in 2021.
Until 2019: both models fit
Anyone comparing the two models in 2019 could well have liked stock-to-flow. Over prices from July 2010 to March 2019 it explained about 89% of the variation in log prices, the Power Law about 92%. That was no clear winner.
After 2019: the price followed the Power Law
From March 23, 2019 to November 7, 2026, there were exactly 2’787 new daily closes that neither model knew. This is how they fared:
| Since Mar 23, 2019 (2’787 days) | Stock-to-flow (2019 formula) | Power Law, calculated only with data up to March 2019 | Power Law (this site's methodology) |
|---|---|---|---|
| Typical price (median) | 0.47 × model | 0.73 × line | 0.84 × line |
| Range | 0.07 to 2.27 × | 0.35 to 2.98 × | 0.41 to 3.36 × |
| Typical deviation | factor 3.5 | factor 1.7 | factor 1.6 |
| Since the April 2024 halving | factor 6.9 | factor 1.7 | factor 1.5 |
| On Nov 7, 2026 | 0.09 × | 0.40 × | 0.48 × |
The middle column is the fair comparison. For it, we recalculated the Power Law using only prices up to March 22, 2019, so with exactly the knowledge stock-to-flow had too. Even this 2019 Power Law has been about twice as close to the price on average since then. Since the last halving, stock-to-flow has typically been off by a factor of seven.
In detail:
- 2020 to 2021: stock-to-flow hits. After the May 2020 halving, the model line rose to about $55’000 to $71’000. The price reached it in spring 2021 and closed at $67’525 on November 8, 2021, just below the model. That was the model's big moment.
- From 2022: the price falls behind. In the bear market, bitcoin fell to $15’784 on November 21, 2022, 0.21 times stock-to-flow. For the Power Law it was 0.41 times the line, close to the floor, as in earlier bear markets.
- From 2024: the model takes off. After the April 2024 halving, the stock-to-flow line rose above $800’000 by 2025. The all-time high of October 6, 2025 ($124’673) reached just 15% of that. Measured against the Power Law trend it was 1.54 times, a high a good tenth below the top band, the line through the cycle highs.
Since April 17, 2021, bitcoin has not closed above the stock-to-flow model on a single day.
How we compared them
Data: 5’957 daily closes from Jul 18, 2010 to Nov 7, 2026 (Coin Metrics until Jan 10, 2018, OKX after that), the same ones the methodology uses and which you can download as CSV.
Deviation on a day = price ÷ model; for the Power Law, price ÷ Power Law line. "Typical deviation" is the root mean square of the log deviations, converted back into a factor: factor 1.6 means the price was typically 1.6 times above or below.
The 2019 Power Law is a straight line through log10(price) against log10(days) over the 3’170 prices up to Mar 22, 2019 (least squares): log10(line) = −16.951 + 5.828 × log10(days). Today it lies about 19% above this site's formula.
Share of variation explained (R²) up to Mar 22, 2019: stock-to-flow 0.89, Power Law 0.92. Since Mar 23, 2019, stock-to-flow explains less than a simple average (negative R²), the 2019 Power Law still 0.61.
Why stock-to-flow is off
The error isn't bad luck; it is built into the model.
The halving is known. Everyone knows the date of the next halving to within a few weeks. An event that everyone knows in advance can't multiply the price by ten. Anyone expecting that would have bought beforehand and anticipated the jump.
Scarcity alone doesn't make a price. The price comes from supply and demand. Stock-to-flow looks only at supply. New bitcoin from mining are about 450 a day today; on a normal day many times that are traded. Whether the price rises depends mainly on how many people want to buy and hold bitcoin.
The model grows without end. Because stock-to-flow doubles every four years, the model price rises tenfold every four years. In early 2030 it would be about $8.7 million, in early 2033 about $38 million. After the last new bitcoin around 2140, the flow would be zero and the model price infinite. No asset grows like that.
Gold refutes it by itself. Gold's stock-to-flow has been almost constant for decades, yet its price has fluctuated strongly. The ratio describes scarcity well; as a price formula it doesn't work. More on comparing the two is in Bitcoin vs. gold.
In April 2020, PlanB followed up with an extended version, "S2FX", which expected an average price of about $288’000 for the years 2020 to 2024. That value was never reached either.
What speaks for the Power Law and what against it
The Power Law has a decisive advantage: from the start, it expects bitcoin's growth to slow. That is exactly what happened. The cycle highs came ever closer to the floor: 37 times the floor in 2011, 6.5 times in 2021, 2.7 times in 2025. The lows stayed near the floor.
It also has weaknesses you should know about:
- No reason within the model. The curve fits, but the formula doesn't explain why. Santostasi's explanation via the network's growth is plausible but not proven.
- The floor doesn't always hold. Since 2010, bitcoin has closed below the floor on 3.7% of all days, most recently in June 2026 at 0.41 × the Power Law line.
- The parameters depend on the data. The 2019 Power Law lies about 19% above our formula today. A study by Baquero and Menezes (2026) shows how strongly the exponent depends on the chosen starting point; over one to two years, the Power Law still forecast better there than the alternatives tested.
- It can fail. If bitcoin isn't above its October 2025 high by December 2028, it falls below the floor. Then the model no longer describes it. Our research paper on the Power Law describes this test precisely.
Each of these weaknesses, checked with numbers, is in The weaknesses of the Power Law.
Worked example: 10’000 euros or francs in both models
If you plan with stock-to-flow, you count on about six times what the Power Law sees as fair value. If you build retirement, a loan or partial sales on it, you are planning with numbers the price hasn't reached since 2021.
What the Power Law gives for the coming years, you can work out with your own amounts in the forecast calculator:
Which model holds up? Our conclusion
The Power Law. Not because it is perfect, but because since its publication it has described the prices that came afterwards. Stock-to-flow has no longer managed that since 2021 and is off by more than tenfold today.
That is why all tools on bitcoin-powerlaw.com calculate with the Power Law: the Power Law page with trend, floor and zones, the calculators and My Goal. We use stock-to-flow only as a measure of scarcity, as in Why there are only 21 million bitcoin.
For you, that means: plan with the floor as the cautious case and the trend as the middle case. Don't count on price targets from models that promise a tenfold rise per halving. How low it can go in a bad case, you can check with the floor calculator.
Frequently asked questions
What is the difference between the Power Law and stock-to-flow?
Stock-to-flow derives bitcoin's price from scarcity, that is from existing bitcoin divided by the new bitcoin of one year. The Power Law derives it only from the time since the genesis block. Stock-to-flow expects a jump of about tenfold after every halving, the Power Law growth that slows a little every year.
Has the stock-to-flow model been disproven?
As a price model, it hasn't held up. Since April 17, 2021, bitcoin hasn't closed above the model price on a single day; on Nov 7, 2026 it was at 0.09 times it. As a measure of scarcity, stock-to-flow remains useful: by it, bitcoin is scarcer than gold today.
What price does stock-to-flow predict today?
By the 2019 formula, the model price on Nov 7, 2026 was about $846’000, about €763’000 or CHF 705’000. After the 2028 halving it would rise to several million dollars. The actual price on the same day was $75’868.
Can the Power Law fail just like stock-to-flow?
Yes. The Power Law is also fitted to past prices. But it has held up much better since 2019 and names its own test: if bitcoin isn't above its October 2025 high by December 2028, it falls below the floor, and the model counts as failed.
Which bitcoin model does bitcoin-powerlaw.com use?
The Power Law with trend, floor at 0.42 × the Power Law line, top band and five zones. Formula, data and our own check are openly available in the methodology. We don't use stock-to-flow for prices.
Sources
- bitcoin-powerlaw.com: methodology, formula, regression and data
- PlanB: Modeling Bitcoin's Value with Scarcity, Mar 22, 2019
- PlanB: Bitcoin Stock-to-Flow Cross Asset Model, Apr 27, 2020
- Giovanni Santostasi: The Bitcoin Power Law Theory
- Carlos Baquero, Raquel Menezes: Bitcoin's Power Law: Weak Structure, Strong Forecasts, arXiv, May 2026
- Saifedean Ammous: The Bitcoin Standard, Wiley 2018
- Satoshi Nakamoto: Bitcoin whitepaper (2008)
- Coin Metrics: community data, reference price until Jan 10, 2018
- OKX: BTC-USDT daily closes from Jan 11, 2018
- European Central Bank: euro reference exchange rates
Not financial, legal or tax advice. Models are no guarantee.
The Bitcoin Power Law newsletter
An email when we publish a new article or calculator. Every issue has a one-click unsubscribe link.