Prediction Markets on Bitcoin Cash

Prediction markets enable trading on the outcome of events, crowdsourcing and aggregating vast quantities of information into simple probabilities.

By cutting through misinformation, prediction markets offer the potential to save countless lives, protect homes and businesses from preventable disasters, reduce the power of special interests, root out corruption, and help people better understand and plan for the future.

This is the first of several articles exploring the development of prediction markets on Bitcoin Cash: advantages and applications of prediction markets (PMs), unique advantages of decentralized PMs, unique advantages of building PMs on Bitcoin Cash, and a technical strategy for developing decentralized PMs on Bitcoin Cash.

What is a Prediction Market (PM)?

Prediction markets (PMs) allow traders to bet on the probabilities of events. With “skin in the game”, traders who best predict events are rewarded, while traders who are wrong — or lying to mislead others— are penalized.

PMs reveal the wisdom of the crowd, offering easy-to-understand answers to hard or controversial questions. PMs work like a weighted, secret ballot poll of the most knowledgable experts, revealing what they truly believe.

For example, a PM could answer: “What will be the effects of Law X by 2024?” with:

• Lower unemployment (-.5%)
• Higher median household income (+$1,200)
• Higher GDP (+.3%)
• Reduced revenue for Special Interest Monopoly X (-50%)

This law would likely improve millions of lives, but in today’s world, it could never pass. At the threat of 50% losses, Special Interest Monopoly X would conduct an all-out blitz, pulling strings to have their “experts” appear on the news, commissioning “studies” to demonstrate how the proposed law would “destroy the economy”, and employing lobbyists to ensure the law never made it to a vote.

In a world with well-established, decentralized prediction markets, this misinformation and corruption would be easily exposed. In the same way the printing press made information widely accessible, prediction markets will make deep understanding widely accessible.

Understanding Prediction Markets

For a deep-dive into PMs, read Prediction Markets of all Shapes and Sizes by Paul Sztorc. The paper begins by describing simple markets like “Candidate X Elected in 2024?” where answers are binary “yes” or “no”. It then goes much further to describe complex, multidimensional markets like the “what will be the effects of Law X?” example above.

For brevity, this article avoids detailed discussion of how PMs work. To learn more, read Paul Sztorc’s PM Sequence of papers: learn how prediction markets integrate all other sources of information (and how to assess a PMs result), how prediction markets are extremely resistant to manipulation (would-be manipulators are quickly bankrupted by market forces), and how prediction markets will reduce crime and improve general human welfare.

Applications of Prediction Markets

Prediction markets represent a technological leap forward which will improve almost every field and industry. Here are just a few of the areas in which PMs might change the world.

Health & Medicine

In health and medicine, knowledge equals lives saved.

When Ignaz Semmelweis discovered that hand washing reduced maternal mortality during childbirth by 70% in 1847, it took 20 years for hand washing to become widely practiced. Prediction markets could have guided the medical establishment to recognize the discovery more quickly, saving many thousands of lives.

More recently in 1999, When the VIGOR safety panel recognized the “disconcerting” trend that Vioxx increased the risk of heart attacks, prediction markets could have helped both the FDA and the public to understand the gravity of the findings. It took 5 more years for the drug to be pulled from the market, leading to as many as 60,000 preventable deaths.

While government oversight failures like Vioxx are often more visible, the most costly and deadly failures of medical regulatory bodies are the treatments they prevent or delay from coming to market. Prediction markets could offer oversight which is immune to regulatory capture, making lifesaving treatments available faster and reducing costs.

Scientific Theory & Discovery

Because PMs can offer a sort of real-time analysis of scientific news, PMs can help industries and the public quickly and intelligently respond to new discoveries. For example, developments in quantum computing are notoriously hard to follow, even for professionals in the software industry. Well-designed prediction markets can serve as decentralized, reliable, incorruptible BS detectors.

PMs have also been successfully used to estimate the reproducibility of scientific research. Scientific PMs may even offer opportunities to venture-fund important research.

During the SARS-CoV-2 pandemic, a wide variety of competing studies and meta-analyses have been produced, often with radically different conclusions. Prediction markets could help to integrate this information reliably, giving the public a minute-by-minute, best estimate of important statistics like household secondary attack rate from asymptomatic index cases. Earlier accessibility of information can more appropriately inform family, business, and government responses, minimizing both death and economic destruction. Multidimensional PMs can even offer policy makers advice by region, promoting more effective and less destructive responses.

Environmental Protection

Climate prediction markets have been studied as a more effective strategy for 1) assessing the accuracy and potential bias of climate studies, and 2) disseminating scientific evidence among the public and policy makers. PMs represent a rare opportunity to shift the climate debate away from typical platitudes and group signaling and toward practical analysis, e.g. “How much will the temperature change in X region by 2050, how will that change affect health, agriculture, tourism, etc., how will X policy affect the currently expected outcome, and how much will it cost?”

PMs may be even more useful in analysis of more localized environmental problems, where stronger special interests can often mislead the public about true environmental costs. PMs could help extract the hidden knowledge of local experts and whistleblowers, even if those individuals might somehow otherwise be prevented from publicly speaking out.

Management of Public Corporations

PMs can also be used by public corporations to make better decisions, e.g. a public airline might create a multidimensional market to estimate its profitability in updating or adding new routes. A public market of this type would be able to incorporate information to which the airline might not otherwise have access, e.g. another airline is planning to cancel a route, or large company X is considering opening a new location in a currently-small destination.

PMs may be especially useful in informing public corporation stockholders about voting choices. E.g. a multidimensional PM could offer stockholders the best possible estimates for how selecting a new CEO might affect the company’s revenues and stock price over various time windows. This would enable innovative business models, like funds devoted to assessing CEO performance and trading on their analysis.

Taken further, using PMs for decision making can also be institutionalized by governments or non-profit organizations.

Marketing & Consumer Protection

Because PMs help to filter misleading claims, PMs can also function as a sort of decentralized watchdog. A PM can help the public accept evidence which will eventually become widely acknowledged, e.g. “The United States Surgeon General to issue an official statement, linking tobacco cigarette use to lung cancer and chronic bronchitis, on or before Jan 1st, 1966?” (Example from Paul Sztorc’s PM Applications.)

Insurance & Disaster Prevention

Traditional “risk-pooling” insurance struggles to withstand widespread shocks like floods or earthquakes. PMs offer a viable path to assessing risk and partially-insuring against uninsurable or hard-to-insure disasters, like large earthquakes, pandemics, sea-level changes, supervolcanic eruptions, wars, large meteor impacts, or detonations of government explosive stockpiles.

Perhaps more importantly, PMs can help to prevent or reduce the damage of many disasters. By providing reliable probabilities for various disasters, PMs can help people 1) avoid building homes and businesses in dangerous locations and 2) organize to prevent or mitigate risks. For example, in recent decades, experts have warned and advocated for more focus on fire prevention in California. By measuring risk and helping to guide policy, PMs could likely have saved dozens of lives and thousands of Californian homes and businesses in the past few years alone.

Disaster Response & Recovery

PMs may also help to coordinate disaster response and recovery by aggregating information much faster than can be done by centralized organizations. A stunning example of this type of forecasting was the stock market’s reaction to the Challenger explosion. While the Rogers Commission took 5 months to conclude their investigations, the stock market appears to have identified the problem–and the expected loss of future revenue to the supplier of the defective part–within hours.

With large, well-established prediction markets, these market forces can help to guide quick responses to a variety of disasters.

Software Security

PMs offer a new solution to one of the hardest problems in software security: defenders must ensure all software is secure, but attackers need only find one “hole” to exploit. With prediction markets, security researchers could make money by “betting” that certain pieces of open source software are free of certain types of vulnerabilities. Companies which rely on that software can both incentivize security analysis and partially-insure themselves by betting on the opposite side.

News & Media

By offering a simple, real-time measurement of event probabilities, prediction markets can help people understand the true importance of various news stories. By filtering out noise and reducing hysteria, PMs will elevate balanced, well-researched reporting. PMs may even offer exciting new business models for independent journalists and detectives.

Transparency & Empowering Whistleblowers

Prediction markets can also tip off law enforcement and the public to scandals in near real-time. Investigation of both the Enron scandal and Madoff investment scandal benefited from whistleblowers. Had PMs been available to discern the truth earlier, these scandals may have have been dramatically less ruinous for thousands of investors.

PMs may also help to expose illegal activity and improve confidence in public institutions. One example from “Whistleblowing” in PM Applications:

“It to be publicly revealed that the United States Federal Government collects (and retains indefinitely) all emails sent both by foreign and US citizens?”

Edward Snowden could have instead anonymously created this contract, and bet on ‘Yes’, alerting the public to this issue. Snowden could then continue to buy ‘Yes’ shares as they were bid down by an incredulous public or a manipulative government. Ultimately, when his documents were released he would make a fortune.

Similarly, by leaking probabilities of (and even the identities of agents involved in) impending crimes, prediction markets can assist law enforcement, offer protection, and even compensate the victims of theoretical, dystopian “assassination markets”. (Mentioned here because they are often confused with PMs: in reality, if “assassination markets” are possible, they would be neutralized by PMs).

Breaking Propaganda & Reducing Violence

Finally, prediction markets have the potential to alleviate a great deal of human suffering by breaking propaganda and reducing violence.

One of the most famous false flag operations, the Gleiwitz incident, was used by Hitler to establish public support for Nazi Germany’s invasion of Poland. Decentralized prediction markets offer a rare opportunity to foil these kinds of operations via exposure by whistleblowers, defection by co-conspirators, or investigation by financially-motivated outsiders. By harnessing the power of markets to identify accurate information, PMs can reduce the effectiveness of these types of public deceptions.

It is estimated that 17 million people were murdered in the Holocaust era from 1933 to 1945. Almost 2 million were murdered in the 1942 “pulse of death” alone. Yet to the disbelief of many modern observers, by November 1944, less than 25% of U.S. respondents believed that more than 1 million people had been killed in concentration camps.

What other tragedies might be limited or prevented by verifying and disseminating truth faster?

Why Prediction Markets?

In short, prediction markets could save millions of lives in the coming decades. PMs can revolutionize industries and dramatically improve the lives of billions of people. Beyond cryptocurrency, PMs may be the most promising technology we have for expanding freedom and unleashing human potential.

For cryptocurrency developers and investors, PMs represent a massive, mostly-untapped pool of demand. In the future, the world’s largest and most liquid prediction markets are likely to be valued in the trillions (USD, 2020). Because many markets will require the staking of capital for long periods of time, the currency system(s) which dominate the PM market will benefit from increased stability and demand themselves.

Why Decentralized Prediction Markets?

Decentralization offers benefits which cannot be matched by centralized PMs.

Non-Custodial

Centralized prediction markets hold user funds in custodial accounts. This adds time-based risk which taints predictions, making some predictions impossible. E.g. “Will the US declare war on a nuclear-armed country by 2050?” may seem highly unlikely, but on a centralized PM, it’s likely to trade closer to 50% than the real probability. Relatively few users will be interested in the risk and opportunity cost of locking up funds for years when the centralized PM may disappear long before the close of the prediction.

In non-custodial, decentralized markets, these probabilities may be more accurate: there is little risk that a decentralized protocol will stop working so long as some users remain interested.

Global, Uncapped, and Uncensored

In some jurisdictions (like the United States) prediction markets are artificially capped, eliminating their usefulness in rooting out corruption. By reducing the maximum bet of whistleblowers and other high-confidence actors, capped markets lose much of their predictive power.

Decentralized prediction markets operate globally, distinct from any single jurisdiction. Decentralized markets can also support anonymous usage, improving accuracy by allowing whistleblowers to participate at lower risk. By reducing friction, decentralized markets aggregate information more efficiently and more reliably.

Why Bitcoin Cash?

Promising decentralized PM projects already exist outside of Bitcoin Cash, but several properties of Bitcoin Cash make it more compelling for new projects.

Low Network Costs

PM projects running inside the Ethereum network are subject to Ethereum network fees, and the on-chain cost of market activity can become prohibitive during high-traffic times, e.g. $25 USD (2020) per trade. Ethereum layer 2 technologies can reduce these costs with some tradeoffs, but further development is required.

Within the class of Bitcoin-derived cryptocurrencies, Bitcoin Cash’s solution to the quadratic hashing problem is required by all transactions, meaning its “base block” is easily scalable to 32MB. With the federated sidechain solution outlined below, it would already be possible for many decentralized PMs to settle withdrawals back to the main Bitcoin Cash chain in a single block, and users can expect to pay transaction fees of less than $0.01 (USD, 2020). While other networks (BTC, LTC, etc.) could implement Bitcoin Cash features to support PMs, higher costs of covenant-updating transactions reduce the reliability of the two-way-peg described below.

Reasonably-Good Privacy

Privacy is crucial for many important applications of prediction markets: experts, detectives, and whistleblowers must be able to share information without fear of retaliation. Between UTXO-based accounting, low on-chain fees, and widely-used privacy technologies like CashFusion, Bitcoin Cash has some of the best potential for privacy in the cryptocurrency space, particularly among currencies with the flexibility to support PMs.

Deflationary

Prediction markets can be biased toward better accuracy in sooner-resolving markets. Traders can only bet on so many markets with a fixed amount of capital: if a trader buys a prediction resolving in 10 years, they must forgo placing that same capital in many other investments in the intervening years. As such, strategies involving multiple, shorter-term buys are often superior to long-term buys, even if the trader has significant, non-public knowledge about the longer-term prediction. (This scenario is simplified. Other market forces can counteract this bias in well-establish markets.)

In prediction markets which trade in inflationary currencies (like USD), this short-term bias is magnified by inflation such that traders must also have an expected gain greater than the expected inflation (and risk thereof) over the course of the long-term prediction. For example, in 10 year predictions, at 2% annual inflation, a trader expects their otherwise-dormant capital to lose ~18% to inflation.

In practice, this bias might be resolvable by markets offering interest payments on active contracts. However, this is difficult in currencies where expected inflation dramatically exceeds interest rates (e.g. USD). Deflationary currencies are vastly superior in this respect: all else equal, holders of deflationary currencies experience a net gain in purchasing power commensurate with economic growth over any period of time. As such, deflationary currencies like Bitcoin Cash are ideal for PM contracts.

Building Prediction Markets on Bitcoin Cash

The rest of this post series will explore a new strategy for enabling prediction markets on Bitcoin Cash.

In short, PMs could use a new kind of Bitcoin Cash covenant, an address where users would deposit funds for use on the market and later withdraw funds during withdrawal periods (much like Drivechain, but only requiring participation from token holders). Bitcoin Cash would likely host several competing Hivemind-like PMs, each with slightly different development roadmaps and sets of arbitrating token holders.

Enabling this kind of covenant would require two changes to the Bitcoin Cash virtual machine. Both changes benefit other major ecosystem efforts, but this post only summarizes their importance for PM covenants. The next posts will describe some early proposals for how each feature could be implemented.

Feedback & Discussion Group

If you have questions, feedback, or you’re interested in contributing to future Prediction Market-related projects, please join us on Telegram:

Bitcoin Cash Prediction Markets on Telegram ➔

Future Development

The following developments are necessary for decentralized prediction markets to be built on Bitcoin Cash.

Scalable Inductive Proofs

Inductive proofs would allow token systems to be designed as Bitcoin Cash covenants. These tokens could support SPV wallets, decentralized minting, and even validation by other covenants, allowing tokens and covenants to interact. While inductive proofs are already technically possible on Bitcoin Cash, every successive transaction for a given token is larger than the last, growing quadratically in size and quickly exceeding virtual machine limits. With some strategy for avoiding this quadratic growth, Bitcoin Cash could support many more decentralized applications, including PMs.

• PMv3: Build Decentralized Applications on Bitcoin Cash ➔
• PMv3-Based CashTokens: Contract-Validated Tokens ➔

Update: February 2022

The PMv3 proposal has been replaced by CashTokens, a set of BCH virtual machine improvements proposed for deployment in May 2023:

Larger Virtual Machine Integers

The Bitcoin Cash virtual machine currently represents numbers in a format which limits the maximum, practically representable value to 2,147,483,647 satoshis, or about 21 bitcoin cash (~$8000 USD, 2020). Many prediction market designs will allow users to deposit and withdraw funds using a specific address on the network, crediting and debiting the user within the decentralized PM’s network. As such, Bitcoin Cash contracts will need to be able to perform math operations on much larger sums of money, at least up to the total deposited in the PM (billion or trillions in 2020 USD).

Update: June 2021

A full specification for larger virtual machine integers is now available.