The Bitcoin Power Law: The Pattern in the Price
The Bitcoin power law is an observation that Bitcoin’s price, plotted against time on log-log axesA chart where both scales count by multiples (1, 10, 100) instead of by equal steps. A power-law relationship shows up as a straight line on it., has followed a remarkably straight trend across its trading history, with price rising roughly in proportion to time raised to an exponent near six. It isn’t a guarantee about the future and it isn’t a price prediction. It’s a description of how Bitcoin has behaved so far, and a way to put its extreme volatility into a longer historical context.
The Bitcoin educator Natalie Brunell frames the idea in a single line:
Companies die, cities live forever, and Bitcoin is growing like a city.
Natalie Brunell, Coin Stories
That’s the intuition behind the analogy. Companies are bounded organizations that can disappear. Cities grow through networks of people, infrastructure, and economic activity, and some persist for centuries. Bitcoin is also a network, and its price history has exhibited a power-law shape. Santostasi describes the day-to-day swings as the weather and the long-run power law as the climate.
First, what a power law is
A power law is a simple relationship where one thing grows as another raised to a fixed power, written as y = a times x to the n. Multiply the input by the same factor and the output changes by a predictable factor set by the exponent. The useful trick is what happens when you plot it. On ordinary axes a power law curves steeply and is hard to read. On log-log axes, where both scales count by multiples (1, 10, 100, 1000) instead of by equal steps, that same curve straightens into a line. If the data falls close to a straight line on a log-log chart, a power-law relationship is one possible explanation.
Power-law relationships show up in many systems once you start looking. They have been used to describe earthquake sizes, city populations, metabolic scaling in animals, income distributions, and network growth. They often appear where growth is uneven, multiplicative, or network-driven, where there is no single “typical” size and large events are far more common than a normal distribution would suggest.
Bitcoin’s power law
Around 2018, a physicist named Giovanni Santostasi noticed that Bitcoin’s price did this. Plot the price against time since the network launched, put both on log scales, and more than a decade of history falls close to a straight line. Harold Christopher Burger published a similar log-log analysis in 2019, an earlier look at the same general relationship. Fitting the price directly puts the exponent near six, and the derivations land close to that too: the 2026 model by Santostasi and Perrenod produces an exponent of about 5.69. In plain terms, Bitcoin’s price has grown roughly in step with time raised to about the sixth power.
What makes the pattern interesting is how much of Bitcoin’s trading history it describes. The fit spans roughly six orders of magnitudeFactors of ten. Six orders of magnitude is a millionfold range, here from cents to tens of thousands of dollars. in price, from the early trading era to the tens of thousands of dollars. It has remained broadly consistent through several crashes of more than seventy-five percent and several major speculative peaks. On the log-log chart, those episodes appear as large deviations around the underlying trend rather than obvious breaks in it.
The chart can be turned into a corridor
The underlying model is a single central trend line, the regression fit. Some versions of the chart add an upper and a lower band around that line to show how far price has tended to stray from the trend. Here those bands sit at roughly 0.5x and 2x the central trend.
That distinction matters. The central line comes from the regression. The corridor is an additional way of picturing historical deviations from it, not a statistical confidence interval and not part of the power law itself. It is an interpretive overlay laid on top of the fit.
The corridor is wide, so it says nothing about where the price will be next week or next month. What it gives you is context. Prices during several major bear markets have approached the lower rail, while major bull-market peaks have approached or exceeded the upper rail. Neither is a signal to act. Both are context.
A proposed mechanism
A pattern that fits the past is interesting. A pattern with a plausible explanation is more interesting, but the explanation has to be tested separately from the curve itself, not read off the same chart.
One proposed explanation starts with network growth. Researchers have found a power-law relationship in the number of non-zero-balance Bitcoin addressesBitcoin addresses that currently hold a balance. Their count is used as a rough, imperfect proxy for how many people use the network. over time. That is not the same as counting users, but it gives a measurable proxy for adoption. Researchers then bring in a version of Metcalfe’s lawThe idea that a network’s value can grow faster than its number of users, because each new participant can connect with all the others., which proposes that a network’s value can grow faster than its number of participants as the connections between them multiply. Santostasi and Perrenod combine those relationships in a model that produces an exponent of about 5.69, close to the exponent found by fitting Bitcoin’s price history directly.
That is an interesting result. It is not proof that adoption causes Bitcoin’s price to follow the power law. The adoption side is inferred through proxies, and the link between network size, value, and price stays a modeling assumption. The authors describe their result as an empirically consistent decomposition rather than a demonstrated causal law.
If that explanation is broadly right, the power law would not be a magic property of Bitcoin’s price. It would be the visible result of a network growing in a particular way. That is a claim worth investigating, not one to take on faith.
How to read it without fooling yourself
The power law is most useful as historical context.
In a deep bear market, when the headlines say Bitcoin is finished, the corridor puts the drop in context. A fall toward the lower rail resembles the range where previous bear markets have occurred, not automatically a new and terminal event. In a euphoric bull market, when the same headlines say this time is different, the corridor cuts the other way. Previous major moves toward the upper rail have been followed by substantial reversals. The lens places those moves against the longer record, whether the market is near a previous low or pushing toward a previous extreme.
What it will not do is tell you what happens next in any precise way, and it is not a trading tool. The corridor is wide enough that on any given day the price can be near either rail, so it says nothing actionable about the short term.
What it doesn’t tell you
Any serious treatment of the power law has to spend real time on its limits, because plenty of people wield it as a crystal ball. It is not one.
A fifteen-year fit is impressive, but it doesn’t prove a thirty-year fit. Networks can saturate, technologies can be displaced, and rules can change. Fitting a curve to the past is far easier than predicting the future, and critics reasonably point out that the exponent was measured after the fact rather than forecast in advance. The proposed adoption explanation is plausible but not settled. And a serious enough disruption, from regulation to something unforeseen, could break a pattern that has no law of physics forcing it to continue.
The clean fit may reflect how the model is built. Baquero and Menezes (2026) find that the fitted exponent varies by nearly a factor of three across reasonable shifts of the time origin, which is a lot of wiggle for a number people quote to two decimals, and they argue the statistical structure is weaker than the tidy chart suggests. Notably, they also find the simple power law is the best forecaster at horizons beyond about seven months, even a naive “tomorrow equals today” model beats it at shorter ones. It is a substantive challenge to the model, and it belongs next to the proponents’ case, not buried under it. See Bitcoin’s Power Law: Weak Structure, Strong Forecasts.
That’s also why you won’t find a price target in this article. The model can be extrapolated into large future numbers, and many people do exactly that, but a projection built on the assumption that the past pattern holds is a projection, not a fact. The power law is far more useful as a way to understand the behavior you’ve already seen than as a promise about the price you haven’t.
The real takeaway
Strip it down and the power law does one valuable thing. It turns Bitcoin’s price history from apparent chaos into something with a describable shape, and it ties that shape, tentatively, to the growth of a real network rather than to pure speculation. That reframes a few of the most common objections. Volatility (Myth #2) looks different set against a longer-term trend rather than as day-to-day randomness. The bubble charge (Myth #4) has to reckon with an asset that has repeatedly suffered enormous drawdowns and later gone on to make new highs. And the “backed by nothing” and “greater fool” objections (Myth #14 and Myth #15) face a more complicated record: Bitcoin’s price has risen alongside the growth of a functioning network, though that does not tell us how much of the price is explained by adoption, speculation, or other factors.
None of that is proof, and the model could still break. Its value is narrower and more defensible: it gives Bitcoin’s price history a measurable shape and a hypothesis about why that shape might exist. The evidence is interesting. The forecast is still unknown.
The power law is interesting because it describes the past unusually well. The hard question is whether anything makes that pattern durable.
All Roads Lead to Bitcoin
Burger 2019; Santostasi & Perrenod 2026
As of 2026
Log-log fit, 2010 to 2026
Power laws have been used to describe earthquakes, city populations, and even how animals burn energy relative to their size. Bitcoin’s price history has also been modeled with a power law, which is one reason the pattern has drawn attention from physicists and network researchers.
Keep going
The power law is the mathematical spine behind one of the most common objections. Here’s how the bubble charge holds up.
Common questions
What is the Bitcoin power law?
It’s the observation that Bitcoin’s price, plotted on log-log axes against time, has followed a remarkably straight trend across its trading history, with price growing in proportion to time raised to an exponent near six. It describes past behavior and is not a price prediction.
Does the power law predict Bitcoin’s price?
No. The model can be extrapolated into future numbers, but that’s a projection resting on the assumption that a fifteen-year pattern keeps holding, which nothing guarantees. Its more defensible use is to put volatility into historical context, not to forecast a price.
Why do some researchers think Bitcoin follows a power law?
Some researchers point to Bitcoin’s network growth. Researchers have identified a power-law relationship in the number of non-zero-balance addresses over time, and have modeled how that growth might translate into network value. But the explanation remains a hypothesis, and the same pattern should not automatically be assumed for other crypto assets.
Is the power law proof that Bitcoin will keep rising?
No. It’s a fit to the past with a plausible but unproven explanation, and it carries real caveats: extrapolation is speculative, the statistical structure is debated, and a large enough disruption could break it. It describes the past. It doesn’t guarantee the future.
Go deeper
- Research. Santostasi & Perrenod (2026), Nonlinear Science, deriving the exponent (about 5.69, data 2010 to 2026) from address growth and Metcalfe-type network scaling. With Harold Christopher Burger’s earlier log-log regression essays (2019), an earlier analysis of the same general relationship.
- Counterpoint. Baquero & Menezes (2026), Bitcoin’s Power Law: Weak Structure, Strong Forecasts, an arXiv preprint (not yet peer-reviewed): the fitted exponent varies by nearly a factor of three across reasonable time-origin shifts and the statistical structure is weaker than the clean chart implies, though the simple power law still forecasts best beyond roughly seven months.
- Proponent’s case. The Physics of Bitcoin / “The Bitcoin Power Law Theory”, Giovanni Santostasi. The theory and its proposed network-growth explanation, presented by its main advocate rather than as settled science.
- Framing. Bitcoin Is for Everyone, by Natalie Brunell (Harriman House, 2025), host of the Coin Stories podcast, whose “cities live, companies die” analogy opens this piece.
- On this site. How we test every claim, and the Bitcoin myths, explained.
Everything on this site is for educational purposes only. It is not financial, investment, tax, or legal advice. Bitcoin carries real risk. Prices move, sometimes sharply. Do your own research, think for yourself, and speak with a qualified professional before acting on anything you read here.