Summary

Highlights

Introduction to Bitcoin Research and Power Law Theory

• The presenter introduces Bitcoin research conducted at Fulgur Ventures to develop strategies for outperforming a buy and hold benchmark.
• A Fidelity chart is presented, showcasing 22 different assets over 40 years, ordered by their five-year cumulative annualized return, highlighting Bitcoin's significant outperformance over the last 10 years, except for 2022.
• The presenter mentions the goal to uncover a framework for understanding and forecasting Bitcoin's behavior using the Bitcoin Power Law Theory.

Explanation of Power Law and its Visualization

• The presenter defines a power law as a mathematical function describing the relationship between two quantities where one grows proportionally with another at a fixed exponent (Y = AX^N).
• Three different space domains are plotted to visualize power laws: linear, log-linear, and log-log charts, with the power law represented by a blue line and compared to an exponential growing function.
• The presenter explains that differentiating a power law from an exponential function is easier in log-linear and double logarithmic charts, where the power law becomes a straight line.

Power Law Relationships within Bitcoin

• The presenter notes that the power law relationship isn't limited to price versus time but also applies to addresses versus time and hash rate versus time.
• The presenter states that price, addresses, and hash rate also form a power law with each other, with an R-squared value of 95% and above, indicating a strong statistical fit.
• The presenter shows the empirical relationships: users grow with time cubed, price is proportional to addresses squared (Metcalfe's Law), and hash rate grows proportionally to the square of the price.

Ubiquity of Power Laws in Various Fields

• The presenter explains that power laws are common in sciences, describing scale-invariant, robust, and stable systems, making them predictable for data scientists.
• The presenter mentions that power law relationships appear in social networks, internet growth, messaging apps, scientific research repositories, and the study of cities.
• The presenter provides examples from physics and biology, such as planetary orbits, gravitational force, electrostatic force, and the relationship between metabolism and animal size.

Bitcoin as the Internet in the 90s: Adoption and Long-Term Play

• The presenter draws an analogy between Bitcoin and the internet, showing a chart of internet adoption since the 90s forming a power law on a double logarithmic chart.
• The presenter suggests that if this data had been available earlier, the growth of the internet could have been predicted accurately.
• The presenter posits that Bitcoin is like the internet in the 90s, emphasizing that it's still very early and a long-term play with the power law at its core.

Cities vs. Companies: Networks and Robustness

• The presenter references Jeffrey West's comparison of cities and companies, noting that cities grow steadily due to intertwined networks, while companies grow exponentially but decline after about 10 years.
• The presenter explains that cities have infrastructure, transport, and other networks that create positive feedback loops, making them robust.
• The presenter draws a parallel between cities and Bitcoin, highlighting the various networks within Bitcoin (blockchain, mining, L2, users, developers, etc.) that exhibit power law relationships.

Power Laws as Essential Signals and R-Squared Value

• The presenter emphasizes that power laws are essential signals and should be studied when identified within a system.
• The presenter shows how the R-squared value changed over time, indicating the goodness of fit of the power law model, stabilizing since 2017.
• The presenter notes that 2017 was when the power law was first discovered, providing about seven to eight years of out-of-sample data.

Decomposing Bitcoin Price into State Regimes

• The presenter explains that a static power law can be decomposed into residuals, showing the deviation of the price from the trend line.
• The presenter uses hidden Markov models to decompose the price into three different state regimes, analogous to fluid dynamics: laminar flow (state one) and turbulent flow (state three).
• The presenter describes the core power law channel (state one) as parallel to the trend line, typically occurring during bear markets, while the transition phase leads to bubble phases.

Bitcoin's Dynamic Power Cycle and Price Predictions

• The presenter represents the data in polar coordinates, with the radius as the price and a full revolution as a four-year cycle.
• The presenter discusses an extension to the Bitcoin Power Law Theory, the dynamic power cycle, suggesting the first cycle was compressed due to market constraints.
• The presenter provides price predictions based on the model: a bear market around $50,000 USD, a fair value around $120,000 USD, and a bull market around $225,000 USD by the end of the year.
• The presenter projects Bitcoin to reach one million per coin between 2034 and 2038, and 10 million per coin around 2046 to the middle of the century.

Inflation Adjustment and Hash Rate Analysis

• An audience member asks if the chart is inflation-adjusted, and the presenter responds that it does not include inflation but can be mapped with gold ounces for roughly the same predictions.
• An audience member asks about hash rate analysis, and the presenter mentions the relationship between hash rate and price (hash rate is the square of the price) but hasn't done explicit exercises with it.

Investment Perspective and Company Analysis

• An audience member inquires about using the model for company investments, considering potential user base or market size.
• The presenter clarifies that companies don't exhibit the same power law behaviors, typically experiencing exponential growth (S-curve).
• The presenter mentions that while the model isn't explicitly used for company analysis, it could be incorporated if a company is related to mining, for example.

Trading Strategies and Diminishing Bubble Peaks

• The presenter suggests a simple trading strategy of riding the waves and DCAing out, noting that bubble peaks are projected to diminish over time.
• The presenter mentions that the decay channel indicates the next bubble in four years, but the returns from the bubbles will decrease relative to the trend line.

Power Laws and Adoption Limits

• An audience member asks if power laws change in nature, and the presenter confirms that systems have constraints, such as adoption limits.
• The presenter mentions other models like the Weibull distribution that collapse into a power law in the beginning.
• The presenter states that the power law is still relevant for the next 10 years for the one million and 10 million predictions.

Entrepreneurial Use Cases and Real Estate Example

• An audience member asks about use cases for entrepreneurs, giving an example of collateralizing Bitcoin to invest in real estate and timing exits based on the chart.
• The presenter admits that this was mainly developed for trading but suggests that if a business revolves around hash rate or Bitcoin users, power laws can map out fair values and potential corrections.

Bitcoin Holder Behavior and Smart Money

• The presenter mentions that analyzing on-chain data can segregate holders into three categories: those holding for more than a year, less than a year, and up to a month.
• The presenter notes that holders for more than a year typically sell at the tops and buy at the bottoms, while those holding less than a year do the reverse.
• The presenter states that the amount of Bitcoin held by long-term holders has gone up by 2x, suggesting following the smart money strategy.

System Updates and Investment Decisions

• The presenter acknowledges that the system needs to be updated daily to track the actual price.
• The presenter mentions that some charting services are showing similar data.
• The presenter concludes that this information can be used to make investment decisions, such as selling when Bitcoin is overvalued.