How to Design Indisputable Prediction Markets

A comprehensive guide to designing prediction markets that don’t trigger disputes.

Co-written by Alex Gaines, Robbie Klages and Aydan Arroyo.

Tldr; FORE Protocol has integrated with UMA’s Optimistic Oracle to launch user-generated, onchain prediction markets on Arbitrum.

Prediction markets are an important aspect of DeFi, but can be a source of major controversy. Creating effective markets requires clear market rules and reliable data sources that minimize the chance of an outcome being ambiguous.

UMA’s Optimistic Oracle has been used to resolve prediction markets for years, with support from our Verification Program. This article shares practical guidelines and real-world examples from the UMA Verification Program to help FORE users design effective and trustworthy markets.

Key Takeaways

• FORE Protocol is live on Arbitrum and secured by UMA’s Optimistic Oracle.
• Anyone can create a prediction market on FORE and earn revenue for generating user participation in their market.
• The success of a prediction market hinges on clear rules, reliable data, specific wording, binary outcomes, and considering edge cases.

FORE Protocol has integrated with UMA’s Optimistic Oracle to bring user-generated, onchain prediction markets to Arbitrum. Anyone can create their own markets, participate in existing ones, or validate the outcomes. In this post, we will explore how to design and create effective prediction markets on FORE.

Creating Markets on FORE

When you create a market on FORE, you essentially set up your own decentralized prediction market that people can speculate on. As a market creator, you contribute to the protocol’s vibrancy and earn a share of the total market payout. Your earnings are directly tied to your market’s popularity and activity. It’s in your best interest to effectively promote your market and attract more participants.

The protocol is very user-friendly and designed to be accessible. Anyone, including first-timers, can easily create distinctive, scalable markets with the flexibility to tailor them as desired for enhanced rewards. Additionally, FORE now offers a referral program. You can share your unique referral link and earn 1% from market positions and NFTs minted via your link. Your referrals also get $10 off their first market position and a 5% discount on their first NFT mint. As you accumulate referral rewards, you can either claim them immediately or wait for greater benefits.

If you want to learn more about creating prediction markets on FORE, check out their official guide here.

Creating markets is easy, but creating them properly can be difficult.

UMA Verification Program: What We’ve Learned

We’ve learned some important lessons from the UMA Verification Program — a team of UMA community members committed to ensuring proper market resolution. These team members contribute evidence and flag assertions that should be disputed to help ensure the accuracy of market outcomes. Polymarket, an industry-leading prediction market platform, relies on UMA for outcome verification. Our experience resolving Polymarket markets has sharpened our skills and revealed several important practices for creating verifiable prediction markets.

We’ve distilled our findings into four easy to follow guidelines for designing prediction markets on FORE (pun intended). These guidelines are meant to help the market resolve smoothly and avoid disputes:

1. Establish Clear, Specific Descriptions and Rules: Ambiguity in market descriptions and rules can lead to controversial market resolutions that lower user confidence, so it’s crucial to be specific. Unambiguous market descriptions are fundamental for smooth market resolution so the verification team can provide unequivocal evidence to support the market outcome. Users ought to know exactly what they are putting their money into, and any misinterpretation of the market rules could lead to confusion and thus user losses. Avoid unintended losses by explicitly defining novel terminology and clarifying any areas of potential confusion. If a dispute happens, it will go through UMA’s Data Verification Mechanism. When this happens, UMA validators interpret your market’s rules to decide which outcome is true. The clearer the rules you set, the easier it is for them to decide.

Example: SUI Airdrop by May 15? — This is an example of a prediction market that lacks clarity in terminology. The market was whether SUI would do an airdrop. SUI gave away tokens, but didn’t call their token distributions an “airdrop”. This led participants to dispute the outcome of the market. In this case, the market creator should’ve defined “airdrop” in the rules, specified whether or not the term “airdrop” must be used publicly for it to count, and clarified whether or not the distribution of tokens without the use of the term “airdrop” counted. The market was proposed again and ultimately resolved as “Yes”. Here is the result of the market shown in UMA’s Optimistic Oracle dapp.

1. Provide a Reliable Validation Source: Trustworthy data ensures accurate outcomes and directly reduces the risk of a dispute. Multiple data sources may give varying results based on small differences in implementation or internal bias. When creating a prediction market, you must provide a validation source that is clearly defined and preferably as independent and unbiased as possible. Market participants and verifiers will determine the market’s outcome based on the validation source. If a dispute arises, the process will be much smoother if everyone has access to information they can trust. Here’s a helpful tip to keep in mind when choosing a validation source: Instead of asking yourself, “Did X happen?” ask yourself, “Did X happen according to (validation source)?”

Example: Will ETH hit $2,500 by EOY? — This is an example of a prediction market that provides a verifiable source of data to resolve the market. The market creator explicitly states that the price depends on Coinbase’s exchange and provides a link to their platform as the validation source. This made it very easy for market participants to verify whether ETH hit $2,500 on Coinbase. (here is the result of the market in UMA’s Optimistic Oracle dapp)

1. Make Sure the Outcome is Binary: You must ensure to provide binary (Yes/No) options for market participants. Although UMA’s Optimistic Oracle can verify many different types of information and outcomes, FORE Protocol is currently limited to Yes/No outcomes. However, they have plans to introduce more options in the future. It’s important that the binary outcome is as objective as possible, for subjective public perceptions may lead to disputes. The best way to prevent this is by following tip #1 and being extremely specific.

Example: Will Harvard get fewer applicants this year? — This is an example of a prediction market that relies on a binary outcome. The number of applicants for the graduating class of 2028 can either be lower, equal to, or higher than the number of applicants for the graduating class of 2027. If lower, the market will resolve to “Yes”. If equal to or higher, the market will resolve to “No”. Since this outcome is quantifiable and verifiable from Harvard’s official information sources, the chances of a dispute round are very low.

1. Consider Edge Cases: Edge cases are problems or situations that occur only at extreme operating parameters. When designing prediction markets, you must acknowledge scenarios that may complicate or delay the resolution of the market and outline how these edge cases will be handled if they occur. Some common edge case examples include election delays, sports event cancellations, athlete injuries, company rebrands, unexpected deaths, etc.

Example: Will Aston Martin beat Mercedes in the 2023 F1 season? — This is an example of a prediction market that acknowledged edge cases. The market creator explicitly acknowledges the possibility of the race being canceled or rescheduled and how it should be handled. They explicitly state, “if for any reason the 2023 F1 season is rescheduled to end at a different time and date on or before December 31, 2023, the same market conditions will apply.” They also clarify, “if the season is canceled or rescheduled to a time and date after December 31, 2023, this market will resolve according to already-available data”. Click here to see this market’s results in UMA’s Optimistic Oracle.

Neglecting these design guidelines can result in frequent disputes, eroding user trust, and additional costs for dispute resolution. Conversely, adhering to these guidelines can lead to smoother market operation, higher revenue, reduced disputes, cost savings, and a thriving, engaged FORE community.

Conclusion

With the help of UMA’s Optimistic Oracle, FORE is blazing the trail for user-generated prediction markets. The ability for anyone to create, participate in, and validate markets not only democratizes financial speculation but also enriches the ecosystem with diverse perspectives. By adhering to the lessons learned from the UMA Verification Program and the practical guidelines outlined in this post, you can ensure that the markets you create are both effective and trustworthy.

The success of a prediction market hinges on clear rules, reliable data, community engagement, and adaptability to the ever-evolving landscape. As we have seen through various examples, your attention to detail in market design can significantly impact your market’s integrity and profitability. Embracing these principles will undoubtedly lead to more vibrant and successful prediction markets, benefiting the entire FORE community.

Begin your FORE journey and create your first market here.

Good luck!

Huge thanks to UMA Discord members Kurapika, Manuel, Obito, aenews, and Bonded for providing research and Polymarket examples.

If you need help, join FORE’s Discord server. Their team is awesome and always eager to help.