Bitcoin's growth follows the same mathematical pattern found in city scaling, biological metabolism, and network adoption. Not a prediction — an observation.
This relationship — proportional1, not exponential — has held across 15 years and nine orders of magnitude, from $0.01 to $100,000. It describes growth that naturally decelerates as the network matures: fast enough to be transformative, slow enough to be sustainable.1 It is not an investment thesis. It is an empirical observation about how Bitcoin scales.
The model is called the Power Law. Its R² exceeds 95%.2 Its structural floor — the boundary below which Bitcoin has never sustained a daily close — has survived every major stress test of the last decade, including the March 2020 COVID crash and the 2022 FTX collapse. And it has held not only in-sample but out-of-sample: first identified in 2014, every subsequent year has confirmed the trajectory. One caveat from the statistical literature: a 2026 audit finds the fitted exponent sensitive to the chosen time origin — the model is stronger as a forecaster than as a physical constant — which is why this site treats the trend as a scenario to reason from, never a promise.
Computed live from Bitcoin's age. The doubling interval is not a fixed number of years — it is a constant share of the network's own age, so it lengthens as Bitcoin matures. Across the documented Power Law exponents (5.57–5.82) it runs roughly 812–851 days; the exponent survey below shows how little that choice moves where the trend sits today, and how much it moves where the trend goes next.
In exponential models, the price doubles at a fixed calendar interval — every four years, regardless of how old the system is. This is mathematically unsustainable: it demands the same absolute performance forever, eventually requiring infinite growth in finite time.
In a power law, the doubling time stretches. It requires a constant percentage of the network's existing age — precisely 12.76% at the reference exponent (21/5.77 − 1), or roughly 12.8% — not a fixed number of years. This means each successive doubling takes longer in calendar time, even though the proportional relationship stays constant.
When Bitcoin was 1,000 days old, it needed 128 more days to double its fair value. At 5,000 days old, it needs 638 more days. At 6,000 days: 766 more days. The growth continues, but the pace naturally decelerates — which is what makes it sustainable.
This is why the conventional metric of CAGR (Compound Annual Growth Rate) can be misleading when applied to Bitcoin. CAGR assumes constant percentage growth per year — exponential by definition. Bitcoin's growth is better described as proportional: it scales as a constant proportion of its own age, producing a trajectory that is robust precisely because it doesn't demand ever-increasing performance.
Because Bitcoin follows a power law, its annualized growth rate naturally declines over time. But even as it decelerates, the projected returns remain substantially higher than traditional equity markets for decades. The table below shows the Power Law's implied CAGR (Compound Annual Growth Rate) from each starting year to 2035, compared to the S&P 500's historical average of roughly 10% per year.4
The key insight: even the decelerating version of Bitcoin's growth outperforms the historical average of the world's most successful equity index. This isn't a prediction — it's what the model's trajectory implies if the 15-year pattern continues. The growth slows, but the slowed growth is still extraordinary by any traditional benchmark.
Driven by scarcity — models price as a function of supply halvings. Gained wide popularity during 2020–2021 but predicts exponential growth that must accelerate forever. Failed to predict the 2021–2022 cycle, missing targets by 50–80%. Implies Bitcoin is valuable because it is rare.
Driven by adoption — models price as a function of network growth over time. Predicts proportional growth that naturally decelerates. Has maintained its predicted range for 15+ years across nine orders of magnitude. Implies Bitcoin is valuable because it is a growing, useful network.
The table below shows when the Power Law trend line first reaches each price level, and when the structural floor — the more conservative and more predictable boundary — secures that level as a likely permanent baseline. Past milestones serve as validation; future milestones are model projections, not predictions.
| Price Level | Trend Reaches | Floor Secures | Status |
|---|---|---|---|
| $1,000 | ~2016 | ~2017 | ✓ Confirmed |
| $10,000 | ~2020 | ~2021 | ✓ Confirmed |
| $100,000 | ~2025 | ~2028 | ● In progress |
| $250,000 | ~2028 | ~2031 | Projected |
| $500,000 | ~2030 | ~2034 | Projected |
| $1,000,000 | ~2033 | ~2037 | Projected |
| $10,000,000 | ~2045 | ~2051 | Projected |
The strongest evidence for any model is its performance on data it was never fitted to. Giovanni Santostasi first identified the Power Law in 2014, and Matthew Mežinskis began charting it publicly through his Porkopolis work. Every year since has been an out-of-sample test — the model was fitted to the data that existed at the time, and subsequent prices have kept tracking the trajectory.
Rather than assert this, the chart below hands you the dial. Drag the fit window to choose how many years of early data the regression may see; the green line refits and re-projects live, and the readout reports the fitted exponent, the trend price it implies today, and how far that trend sits from the one you would get by fitting every year of data. The default cutoff is end-2017 — where the site's own 2026 refit settled, because it brought the out-of-sample bias to near zero and still leaves the better part of a decade as untouched test.
Some windows fit badly, and that is the point of handing you the dial: a very early cutoff (drag to end-2014) has too little data and over-projects the trend roughly four-fold by the mid-2020s — and the gap only compounds after. A recent-only window mostly re-describes the present and proves little. The interesting range is the middle — from about 2016 onward the fitted exponent settles near the reference value, 5.77, and stays there no matter where you stop. That stability is the out-of-sample claim.
It is tempting to read the trend line as the place price belongs — drifting above and below it in equal measure. It doesn't. Measured across every historical sample, Bitcoin has spent more time below the trend than above it. The split below is computed live from — price samples since 2010.
This is not a sign that price is “due” to return above trend. The mean log-deviation ?The average distance from trend, measured the way the chart is drawn — in log terms. is essentially zero (—) — but that average hides the shape: a handful of violent overshoots balance many long, quiet stretches below the line. Below trend is the ordinary condition; the returns arrive in bursts. That is an argument for holding through the quiet stretch, not for timing a return to trend. The Doubling Ladder maps this wave around the trend in depth, using month-end data.
What are the exponent and the coefficients? The trend is just price = a × ageb — b sets how steep the curve is, a sets its scale, and only the pair means anything. They trade off, so a steeper exponent does not mean a higher price. The only like-for-like comparison is the price each fit implies at a fixed date — which is how the table and explorer below rank them.
Highest exponent, yet below the trend. bitcoinretirement.net uses the steepest exponent in the set (b = 5.82) — and its trend still sits about 2% below the reference fit today, because its a is lower. It crosses above only around 2045. Rank these fits by exponent alone and you get the ordering exactly backwards.
Why do different authors use different ones? Because they fit different windows of data, different price series, and slightly different methods — reasonable people fitting the same phenomenon land on slightly different numbers. The How the fit is justified column carries the specifics. The Porkopolis pair every page on this site uses is the one we will call the reference fit.
| Source | a ?The scale — where the fitted curve starts. | b ?The steepness — how fast the fitted trend grows. | Implied trend today ?The price this fit's trend line puts on today. | How the fit is justified |
|---|
Listed but not plotted — exponent published without its paired a, so no curve can be placed: b1m.io / Fred Krueger (b = 5.566) and a naive fit through the full series (b = 5.63, our own). Saying why they can't be drawn is the point: an exponent without its constant is not a model.
| Implied trend price | 2026 | 2035 | 2045 | 2060 |
|---|
Do they change over time? Yes — fits refresh as new data arrives. The two Porkopolis rows are the same author at two moments, and they differ by well under 2%. That is drift, not revision: the model is not being rewritten, it is being re-measured against a few more years.
What do the differences imply? Fitted to the same history, every source agrees within single digits on where the trend sits today — then they fan out with the horizon. Deviation from the reference fit keeps widening, decade after decade: BitcoinPower.law runs about −9% today but roughly −17% by 2060. That is the real disclosure — how much of the long-run return case rests on a parameter that is itself uncertain.
Why do we use the one we use? The reference fit is the longest-standing published one with a disclosed methodology, and the structural floor derived from it has the tightest fit of any band on the chart. A reference line's chief virtue is stability: refitting it every month would quietly move every number on the site and break every saved scenario link. So it stays fixed across the site, reviewed once a year, with its divergence from the other fits shown above rather than hidden.
The model does not return a number, it returns a corridor — and the width of it is the finding. Floor, trend and ceiling for any year are three different statements about the same date, not one forecast with error bars. Slide to any date and read all three.
The floor represents the most conservative estimate (R² > 0.99); the trend is the median fair value; the ceiling marks the historical upper boundary.
The Power Law is an empirical observation, not a physical law. It fits 15 years of data with remarkable consistency, but it is not derived from first principles the way gravitational equations are. Its creator, Giovanni Santostasi, describes Bitcoin as a stochastic system that oscillates around its trend — the model describes the trajectory, not any specific price on any specific day.
Projections become less reliable as they extend further into the future. The prediction corridor widens with the horizon, so projections carry more uncertainty the further out they run. Some econometricians argue that the high R² could partly reflect the fact that both price and time trend upward. The out-of-sample track record since 2014 is the strongest counter to this critique, but it remains a legitimate statistical concern.
There are two ways this model could break, and both belong on this page. Downward: a sustained daily close below the structural floor would disprove it outright. This falsifiability is a strength, not a weakness — the model makes a testable claim rather than an unfalsifiable assertion — and to date the floor has survived every major stress test, including the COVID crash and the FTX collapse.
Upward: the model could also break in Bitcoin's favour. Matthew Mežinskis has argued that if Bitcoin's size and importance relative to fiat destabilize in a good way, adoption could turn genuinely exponential and pull price above the power-law channel rather than tracking along it. We grade this as speculation, not forecast — but monetary regime transitions, where a new money supplants an old one, have historically been fast and nonlinear once they tip. A site willing to say the model may break against us has earned the right to say it may break for us; both are stated with the same humility. Past adherence guarantees nothing in either direction. This page presents the Power Law as a structural framework, not as investment advice.
Creator of the Bitcoin Power Law Theory. First published power law relationships in Bitcoin's price, hash rate, and address growth in 2014. The intellectual originator of the theoretical framework.
Author of the most widely cited Power Law implementation, published 2018. Originator of the characterization of Bitcoin's power law growth as “proportional and sustainable.”1 The coefficients used on this page (a = 1.6×10−17, b = 5.77) are from his work.
Mathematician, Bitcoin investor, and co-author of Bitcoin One Million. His work with the b1m.io dashboard and public discussions with Santostasi brought the Power Law framework to wider attention.
Pseudonymous poster whose early logarithmic regression chart of Bitcoin's price history inspired Mežinskis's subsequent power law work.
Physicist whose research on universal scaling laws in cities and biological organisms (Scale, 2017; Bettencourt & West, Nature, 2010) provides the broader scientific foundation for why power laws emerge in complex systems.
This page is a high-level summary intended to make the Power Law accessible. For the full mathematical treatment, the definitive references are Porkopolis Economics and Giovanni Santostasi's foundational essay. For an interactive dashboard with live model tracking, see b1m.io.
Power laws don't describe assets. They describe systems — complex, self-organizing structures whose growth is governed by internal feedback dynamics rather than external market sentiment. Stocks fluctuate with earnings and psychology. Commodities respond to supply and demand. But systems — cities, organisms, networks — scale according to structural rules that produce predictable mathematical relationships across many orders of magnitude.
The fact that Bitcoin follows a power law is evidence of what it actually is: not an asset to be traded, but a system to be understood. Its growth is proportional1 — scaling as a constant fraction of its own age — and sustainable1 — decelerating naturally as the network matures. This same proportional, sustainable pattern appears across biology, urban science, and network theory. Understanding the precedents makes Bitcoin's trajectory structurally expected rather than surprising.
In 1932, biologist Max Kleiber discovered that metabolic rate across mammals scales with body mass raised to the power of 0.75. A mouse burns far more energy per gram than an elephant. This isn't a coincidence — it's a structural law that emerges from the branching architecture of biological distribution networks: circulatory systems, respiratory trees, and neural pathways all follow the same fractal branching geometry, and that geometry produces power law scaling.
The key insight isn't just that the relationship exists — it's that it's proportional and sustainable. Larger mammals don't need exponentially more energy; they achieve greater efficiency through structural scaling. The growth is governed by physics, not by intention. Bitcoin's hash rate scales with price following an analogous relationship — security increasing predictably with value, governed by the same mathematics that governs how elephants breathe.
The reason Kleiber's Law works is that biological organisms distribute resources through branching fractal networks — the same self-similar pattern that appears in river deltas, tree root systems, lightning bolts, and the bronchial passages of lungs. These networks follow a universal scaling principle: each level of branching divides into a predictable number of sub-branches, with the diameter and flow rate at each level following a power law.
This is the deep structural insight: power laws aren't coincidences that happen to appear in unrelated domains. They emerge from a universal architecture — branching distribution networks that optimize the transport of resources (blood, water, information, value) across a system. Geoffrey West's work at the Santa Fe Institute showed that this branching geometry is the common root of Kleiber's Law, urban scaling, and network growth. Bitcoin's transaction and mining networks follow the same branching topology, which is why the same mathematical form appears.
Physicist Geoffrey West and colleagues discovered that cities exhibit a remarkable dual scaling pattern. Physical infrastructure scales sublinearly (exponent ~0.85) — a city twice as large needs only 85% more roads and pipes, not twice as many. But socioeconomic outputs — GDP, patents, wages, innovation — scale superlinearly (exponent ~1.15). Every new person added creates more potential connections, driving disproportionate productivity.
The Bitcoin parallel is direct: the ~12.8% proportional doubling in Bitcoin echoes the ~15% superlinear scaling in cities. Both are systems where each new participant creates more potential connections, driving disproportionate value growth. And notably, cities are among the most durable human institutions — they outlast companies, governments, and currencies. Companies follow exponential S-curves and eventually die. Cities follow power laws and persist. Systems that scale like cities have the durability characteristics of cities.
The telephone network provides the earliest modern example of power law scaling through adoption. In 1908, AT&T's Theodore Vail observed that a telephone without a connection at the other end is useless — its value depends entirely on the number of other telephones connected. Robert Metcalfe later formalized this as Metcalfe's Law: the value of a network is proportional to the square of its users.
The fax machine's trajectory is even more instructive. In the early 1980s, fax machines were a niche curiosity. But as installations grew from thousands to millions, the value of each device surged — because the expanding pool of potential recipients made every existing machine more useful. Total network utility scaled roughly as the square of connected machines. Fax crossed a threshold from "niche tool" to "essential business standard" not because the technology improved, but because the network grew.
Bitcoin's adoption follows the same structural dynamic, but with a critical difference: while telephone and fax adoption followed exponential S-curves (fast start, rapid acceleration, eventual saturation), Bitcoin's adoption grows as a power law — curbed by the difficulty adjustment (which prevents premature completion of the emission schedule) and by investment risk (which moderates the speed at which new participants enter). These curbing mechanisms transform a potentially unstable exponential adoption spike into a proportional, sustainable power law. The growth continues indefinitely but at a naturally decelerating pace — which is why the Power Law has held for 15 years and counting.
Metabolic networks in biology. Branching distribution systems in rivers and vasculature. Social networks in cities. Communication networks in telephones. Transaction and mining networks in Bitcoin. The specific exponents differ, but the underlying mathematical form is the same: complex systems that scale through networks produce power law relationships.
The pattern extends even further. Earthquake magnitudes follow the Gutenberg-Richter power law — many small tremors, few large ones, at a mathematically precise ratio. Word frequencies in every human language follow Zipf's Law — the most common word appears twice as often as the second most common, three times as often as the third. Wealth distribution across populations follows a Pareto power law. In each case, the same mathematical form emerges from systems where many small events and few large events are governed by the same structural dynamics.
Bitcoin's adherence to this ubiquitous pattern is not coincidental. It is the expected behavior of a network system scaling through adoption — the same mathematics that governs how elephants metabolize, how cities innovate, and how languages organize themselves. This is why the framing matters: if Bitcoin is an "asset," its price is unpredictable. If Bitcoin is a system, its scaling behavior is structurally governed — proportional, sustainable, and consistent with mathematical laws observed across every complex network ever studied.
The Power Law isn't just a curve that happens to fit Bitcoin's price history. There is a structural explanation for why it emerges from Bitcoin's protocol design — an explanation that connects adoption dynamics, network theory, and the protocol's own built-in constraints.
The Power Law exponent of ~5.8 isn't arbitrary. It can be derived from two established principles working in combination.
Step 1: Bitcoin's user adoption grows as a power of time: Users ∝ t³ (cubic growth, curbed by risk and the difficulty adjustment).
Step 2: Network value follows Metcalfe's Law: Value ∝ Users² (each new user creates connections with all existing users).
Step 3: Combining: Price ∝ (t³)² = t⁶ ≈ t5.8 (empirically measured).
Three steps, each grounded in an established principle, producing the observed exponent. This is the closest the model has to a first-principles derivation — and while it doesn't constitute formal proof, it suggests the Power Law emerges from Bitcoin's structural properties rather than being a coincidental curve fit.
Every 2,016 blocks (approximately two weeks), Bitcoin recalibrates the difficulty of mining to maintain a consistent block time of roughly 10 minutes. This mechanism is the single most important structural reason Bitcoin follows a power law rather than an exponential curve.
Without it, a surge in mining activity would produce blocks faster and faster, completing the emission schedule prematurely and destabilizing the network's carefully calibrated supply curve. The difficulty adjustment acts as a thermostat — when the network heats up, difficulty increases, naturally moderating the pace. When activity cools, difficulty decreases, preventing collapse. This built-in governor transforms potentially unstable exponential dynamics into the stable, self-regulating power law we observe.
Bitcoin's growth trajectory is sustained by three interconnected feedback loops that reinforce each other:
Adoption → Value: As more people use Bitcoin, network value increases (Metcalfe's Law). Each new participant creates more potential connections, driving disproportionate value growth.
Value → Security: Higher value attracts more mining power, increasing the hash rate and making the network exponentially more expensive to attack.
Security → Trust → Adoption: Greater security makes the network more reliable and trustworthy, attracting new users — restarting the cycle.
These three loops are self-reinforcing but not self-accelerating — the difficulty adjustment and investment risk act as natural governors that keep the system in the power law regime rather than tipping into unsustainable exponential acceleration.
A counterintuitive finding: the Power Law's structural floor has a higher R² (often exceeding 0.99) than the median trend (~0.95). The floor is more predictable than the average. Why?
Market peaks are driven by top-side noise — speculation, leverage, psychology, and media cycles that vary wildly in magnitude from one cycle to the next. The 2013 peak was far more extreme relative to trend than the 2021 peak. This variability at the top introduces noise into the median regression.
Market bottoms are governed by bottom-side physics — the marginal cost of mining production and the price at which long-term holders refuse to sell. These constraints are rooted in energy expenditure and game theory, not in psychology. The minimum valuation is structurally determined, making it far more mathematically predictable than the average or peak valuations. This is why we default to the floor scenario in our calculator — it's not just the most conservative estimate, it's the most reliable one.
As Bitcoin matures, the magnitude of its price cycles compresses. Each successive cycle sees smaller percentage deviations from trend:
Cycle 1: Price peaked at roughly 8× over trend.
Cycle 2–3: Peaks at roughly 4–5× over trend.
Cycle 4: Price peaked at roughly 2.5× over trend.
This narrowing is a natural consequence of power law dynamics — as the network grows and matures, the system becomes more resistant to speculative swings. The same amount of new capital produces proportionally less price impact. Volatility doesn't disappear, but it compresses, transitioning Bitcoin from a nascent, hyper-volatile network toward a stable global monetary asset. Each cycle is less dramatic than the last.
Most monetary systems require continuous expansion to function — fiat currencies must grow their supply to service compounding debt, creating an inherently inflationary dynamic that physicist Geoffrey West warns leads toward structural instability.
Bitcoin has zero monetary entropy. Its supply is fixed at 21 million. It does not require open-ended growth to maintain its value or its network. This property means Bitcoin's power law trajectory is not dependent on perpetual monetary expansion — unlike fiat systems, which must inflate or collapse. Bitcoin's growth is driven by adoption, not by monetary dilution. When adoption eventually plateaus, Bitcoin doesn't need to find new sources of growth to sustain itself. It simply is.
This connects directly to the arguments made elsewhere on this site: in The Half-Life, fiat's purchasing power decays continuously because inflation is structurally necessary. In The Melting Ice Cube, holding cash is an active decision with a quantifiable cost. The Power Law provides the mathematical framework for understanding why these dynamics exist — and why Bitcoin, with its fixed supply and adoption-driven scaling, follows a fundamentally different trajectory.
This page presents a high-level structural summary. For the full mathematical treatment and ongoing research, the definitive resources are:
Giovanni Santostasi: The Bitcoin Power Law Theory — the foundational essay presenting the full theoretical framework.
Porkopolis Economics: The Chart — the most widely cited implementation with interactive visualizations and methodology.
b1m.io — Fred Krueger's live dashboard tracking the Power Law in real time with current statistics.
Fulgur Ventures: Executive Summary — the best structured academic-style overview of the theory.
For the broader scientific foundation of power laws in complex systems: Geoffrey West, Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Companies, Cities, and People (Penguin, 2017).
The Power Law isn't a single price line — it's a channel. A structured corridor with a conservative floor at 0.42× trend, the central trend itself, and an upper boundary at 3× trend. Bitcoin oscillates inside this channel: hugging the floor in deep drawdowns, spiking against the upper band in euphoria, returning to trend across the cycle. The chart below shows the channel across Bitcoin's lifetime, with daily price overlaid.
Toggle the time axis below to see the same model two ways. Linear time shows the bands as expanding curves — the lived experience of Bitcoin's price. Logarithmic time collapses them into straight lines — the underlying mathematics. Both views are correct. They are the same Power Law rendered in different coordinate systems.
The Power Law channel framework on this page — the structural floor at 0.42× trend, the upper band at 3× trend, and the daily-close-as-stress-test convention — was developed and refined by Matthew Mežinskis at Porkopolis Economics. This visualization is built on his coefficients (a = 1.6×10−17, b = 5.77) and band ratios. For the live version with ongoing analysis, see Porkopolis: The Chart directly.
A power law is a relationship where one quantity scales as a power of another. For bitcoin, price scales as days-since-genesis raised to the power of 5.77. On a linear-time axis with a logarithmic price scale, that relationship looks like a curve: each successive price doubling takes longer in calendar time. On a log-log axis — both scales logarithmic — the curve straightens into a line. The slope of that line is the exponent.
The two views serve different mental models. Linear time matches how we experience markets — long periods near one price level, then sudden moves to the next. Log-log shows the underlying mathematics — a clean linear relationship that has held for fifteen years across nine orders of magnitude. Toggling between them is itself the educational point.
The trend is price = a × age^b, where age is days since the genesis block. b is the exponent — how steep the curve is — and a is the coefficient that sets its scale. Only the pair (a, b) means anything: the two trade off, so a steeper exponent does not by itself imply a higher price.
Because different authors fit different windows of data, different price series, and slightly different methods, so reasonable people land on slightly different coefficients. The like-for-like comparison is not the bare exponent but the price each fit implies at a fixed date, since a and b trade off.
Yes — a fit refreshes as new data arrives. Two Porkopolis snapshots of the same fit differ by well under 2%. That is drift, not revision: the model is re-measured against a few more years, not rewritten.
Fitted to the same history they agree within single digits on today's trend, then fan out with the horizon. BitcoinPower.law runs about 9% below the reference fit today but roughly 17% below by 2060. The disagreement is mostly about the far future, not the present — which is the real uncertainty to disclose.
The Porkopolis reference fit, a = 1.6 × 10⁻¹⁷ and b = 5.77 — the longest-standing published fit with a disclosed methodology, and the one whose structural floor fits tightest. It stays fixed across the site and is reviewed once a year, because a reference line's value is stability: refitting it monthly would move every number on the site and break every saved scenario link.
Get the essays and tool updates
A few emails a month when something new ships — an essay, a new tool, a meaningful update. No funnel, nothing for sale, unsubscribe anytime.
Every page on this site has been improved by someone pushing on it. Ask a question, flag an error, or suggest what’s missing — it goes straight to the author, never published.
I’m also available for one-on-one conversations and open to collaborations — Work with me →