Bitcoin Spreads Like a Virus¶
Bitcoin Spreads Like a Virus is a research paper demonstrating that Bitcoin's long-term price follows a Gompertz sigmoid growth function -- the same mathematical curve used to model the spread of viruses, bacterial growth, and tumor proliferation. Published in 2019, the paper provides a mathematical derivation linking traditional time-value-of-money concepts to Metcalfe's Law and shows that Bitcoin's adoption proceeds along a predictable biological growth curve rather than through speculative mania.
The Problem of Noise¶
Financial markets generate enormous quantities of noise -- trading activity driven by emotion, misinformation, or speculation rather than by information about fundamental value. As Fischer Black observed, people who trade on noise think they are trading on information. The cryptocurrency market, populated by participants ranging from naive day traders to sophisticated institutional investors, is an exceptionally noisy environment.
The paper set out to demonstrate that Bitcoin's price formation is not a noisy result of emotional investing but is instead founded on the economic principles of network effects.
From Time Value of Money to Network Value¶
The key mathematical insight connects two well-established concepts. The standard time-value-of-money formula relates present and future value through a rate of return compounded over time. Metcalfe's Law states that network value is proportional to the square of users. When these are combined, a striking result emerges: if the number of users grows at a rate consistent with a sigmoid function, then the natural logarithm of price traces a horizontal parabola when plotted against the square root of time.
This "square root time" framework is unusual but mathematically natural. Because Metcalfe's Law relates value to the square of users, and user growth follows an S-curve, the square root transformation linearizes what would otherwise be a complex nonlinear relationship. The result is that Bitcoin's entire price history -- spanning orders of magnitude from fractions of a cent to tens of thousands of dollars -- falls along a clean parabolic arc.
Why Gompertz, Not Logistic¶
Both Gompertz and logistic functions produce S-shaped growth curves, but they differ in a critical way: the Gompertz curve is asymmetric, while the logistic curve is symmetric. In a logistic model, growth accelerates and decelerates at equal rates around the inflection point. In a Gompertz model, early growth is faster and mid-cycle deceleration is more gradual.
Using Franses' parametric selection method on quarterly Bitcoin active address data, the paper found that a Gompertz function is the appropriate choice. This matters because using the wrong growth curve produces substantially different forecasts. The Gompertz function historically describes biological diffusion processes -- viral infection, bacterial colonization, tumor growth, and mobile phone proliferation -- all phenomena where early adoption is rapid and later saturation is gradual.
The estimated parameters for Bitcoin's user growth (a = 14.1492, b = 0.685, c = 0.001) produce a curve that fits observed active address data with high fidelity.
Deviations as Evidence of Manipulation¶
When Bitcoin's actual price deviates significantly from the Gompertz-Metcalfe model, the deviation cannot be explained by user-related factors such as transactions, active accounts, wallets, nodes, or hash rates. The paper identified five notable deviations corresponding to documented periods of price manipulation. These periods were independently corroborated by Gandal et al. (2018), Griffin and Shams (2018), and Peterson (2018).
After excluding these manipulation episodes, the remaining price history conforms closely to the model, reinforcing the conclusion that Bitcoin's organic price formation follows predictable network growth dynamics.
Comparison to Facebook and the Internet¶
The paper validated the framework against two other networks. Facebook's stock price, when plotted against the square root of time since its early user growth, follows the same parabolic pattern predicted by combining Metcalfe's Law with sigmoid user growth. The Dow Jones Internet Composite Index tracks the Metcalfe value of the internet computed from the number of global internet users.
Both Bitcoin and Facebook were innovative but not entirely original (DigiCash preceded Bitcoin; MySpace preceded Facebook). Both faced bans in China. Both received widespread publicity about their adoption. The parallel growth patterns across fundamentally different networks -- one a social platform, the other a digital currency, the third the internet itself -- suggests the underlying mathematical relationship is robust and not specific to any single technology.
Implications¶
The model provides a framework for distinguishing information from noise. Forecasts of extreme prices -- whether sky-high or doomsday -- are almost certainly noise because the probability of such extreme events is very small. User growth, not speculation, drives long-term price.
The biological metaphor in the title is more than rhetorical. Bitcoin's adoption literally follows the same mathematical curve as viral infection. Each new user increases the probability that others in their social and economic network will adopt, just as each infected individual increases the probability of infecting others. The network grows not through centralized marketing or government mandate but through decentralized, peer-to-peer transmission -- spreading, in the mathematical sense, like a virus.