Original author: Sid, IOSG Ventures
Special thanks to Aravind Menon for his insights
The life cycle of a bet
Investopedia states: “A prediction market is a market where people can trade contracts based on the outcome of unknown future events.” Essentially, it is a betting/gambling market. To better understand the betting market, let’s break down the life cycle of a bet:
belief
At the belief stage, a prediction is just an opinion. When a person turns his opinion into a bet of money, he can reap rewards when the results support that belief.
Beliefs are formed through the interplay of complex cognitive, social, emotional, and environmental factors. Opinions can arise from immediate conviction or thoughtful reflection, and because there is no monetary loss to the person expressing the opinion, opinions are given more freely.
bet
There are two situations in which bets are generated:
Want to profit from ones beliefs
There is a belief-independent but very attractive result
The first type of bet may come from a calculated perspective, while the second type comes from a small bet, big win attitude.
In order for any contract to be successful, there needs to be a side and a counter-party:
To bet $50 on Chelsea to win a game, you need someone (or many people) willing to put a total of $50 on Chelsea to lose (assuming the odds are 50/50)
In margin trading on GMX, the trader opens a long position and GLP is the counter
Casino games such as roulette and blackjack require a banker as the opponent
Sometimes, incentives are needed to attract opposition, because the outcome of an event is not always equally likely. These incentives can include various forms, such as odds, bond curves in AMM (Automated Market Maker), and even funding rates in perpetual/margin trading platforms.
The structural design of market forecasts becomes more complex when focusing on specific types of outcomes. Sports betting, for example, requires unique odds setups because almost no two events will have nearly the same outcome. Furthermore, each major event (e.g. the outcome of a league championship) may involve many smaller events (the outcome of each league game), adding further complexity.
In predicting events, the contract also needs to be executed correctly. What if your opponent refuses to pay? This is why derivatives are essentially legally enforceable contracts. On the blockchain, contracts can be executed trustlessly based on the results.
Therefore, to place a bet, you need:
The event occurs (or does not occur) and publishes the event/game contract
Ensure that enough participants have an opinion on these events (maker demand: market participants provide market orders)
Ensure these participants have counterparties (taker demand: market participants execute existing orders in the market)
ensure settlement
Ensure there is no market manipulation
result
Gambling games promote the illusion of control: the belief that gamblers can exert skill over outcomes that are actually defined by chance. - Dr. Luke Clark
The result is the end of the event stakes. Once the outcome is determined, the bet is complete.
Do prediction markets require Web3?
Let’s look at the necessity of Web3 based on the above-mentioned criteria for creating a gambling market:
Event/Game Creation
There are no clear blockchain use cases here other than permissionless event publishing. Permissionless Posting is a bug, not a feature, as it creates a high degree of redundancy for the same event, thus worsening the punter experience. Bets can be created based on an event, or games like on-chain roulette or blackjack can be created. (Permissionless publishing refers to anyone being able to publish information or transactions without centralized review or permission)
An event can also be price discovery. We’ve seen prediction markets on Aevo for yet-to-be-released tokens, which provide a good indicator of how the market feels about the token’s price.
Parcl is also creating a prediction market for better price discovery in real estate. It provides homeowners with a ballpark figure for what their home is worth and also provides a budget range for buyers looking to purchase real estate in a certain city.
The use case for price discovery is also a function of liquidity in event contracts, which is why the next section is important.
manufacturer demand
Blockchain cannot control manufacturer demand, which is entirely driven by offline behaviors such as marketing or games built into the product.
Those targeting price discovery must focus on generating as much maker volume as possible to obtain the most accurate price for a specific asset.
counterparty
Now we get to an interesting topic. Counterparties can be incentivized to gamble through attractive odds, especially when the outcome of an event is almost certain. As you can see in the image below, it is possible to win $200 with a $0.50 bet due to the huge mismatch in the Polymarket order book.
One way is like Augur Turbo, where each market is an independent market running on Balancer AMMs. LPs (liquidity providers) here serve as counterparties in different markets. While this structure does a good job of avoiding overreliance on odds calculation (or fetching), it makes the experience of publishing predicted events worse.
For price discovery order books like Aevo, if there is no liquidity, the platform will sometimes have to act as the counterparty itself. This is not ideal, especially when the bottom of the market is unknown.
Another approach is to create a counterparty LP pool like “The House”. Just like what Azuro and WINR have done. There is a liquidity pool that will serve as the counterparty to bettors. Parcl has a USDC liquidity pool that serves as the counterparty for traders to long-term or short real estate prices in different cities.
Both protocols have proven their effectiveness:
Revenue generated by Azuro’s LPs on Polygon (Source: Dune)
WINRs LP token (WLP) value has grown from $1 to approximately $1.27 (indicating a 27% return if LP starts around July 1, 2023)
(Source: Dune)
These models demonstrate some good product-market fit, where the front-end only needs to focus on punters placing bets on the platform without having to manage order books or make the trade-offs that come with AMMs.
You can think of these models as Uniswap v4, with different frontends using the underlying liquidity (similar to hooks).
The WINR protocol has a casino betting front-end and another margin trading protocol that offers up to 1000x leverage, which ensures high pool utilization but can be very dangerous for the pool.
ensure settlement
Once the event is complete, bets need to be settled. In the AMM structure, everything is on-chain and settled on the contract. For the Polymarket order book model, the order book is maintained off-chain. Polymarket can block withdrawals if needed. For Azure frontends like Bookmaker.xyz, no deposit is required. Each bet is considered an independent transaction. The only off-chain components are the calculation of odds and the data source.
Make sure there is no manipulation
If there is a centralized data provider and this data source is manipulated by the provider, this can adversely affect outcomes for market makers and takers. This is one of the main reasons why most Web3 prediction markets use oracle systems like Chainlink. There is a trade-off between latency and data integrity when using oracles. When choosing an oracle, platforms can choose between first-party and third-party oracles, which involves latency trade-offs. In fast-moving events, whether there is a delay is a very important factor.
In casino games, it is crucial that the randomness is complete and its fairness cannot be affected by its source.
Chainlink and other oracles like Supra and Pyth minimize the possibility of manipulation through aggregation, but the authenticity and reliability of data sources remains an issue in the vast market. These oracle systems strive to protect markets from improper manipulation by aggregating multiple data sources to provide reliability and reduce the risk of a single point of failure. Still, ensuring the authenticity of data sources and preventing manipulation remains an ongoing challenge in prediction markets.
Successful and failed existing applications
When we look at crypto markets and prediction markets, the more successful examples are cryptocurrencies being used as assets for staking on sites like Stake.com and Rollbit.
(Blue numbers are predicted numbers)
Although applications like Polymarket have had some success, it is not a platform that can maintain consistent trading volumes because there is a huge gap between the event environment and the platform.
Source: Dune
The product market fit (PMF) of cryptocurrencies and prediction markets has initially emerged in “House” pool systems like Azure and WINR. An obvious application scenario is that a new front end focused on a specific type of prediction market only needs to focus on the demand side. They can leverage systems like Azure and WINR, which in turn provide stablecoin holders with best-in-class yields (40-60% annualized at current rates).
In most countries, regulation of gambling apps and online casinos is very strict. Protocols like Azuro and WINR may also face lower regulatory pressure than companies like Rollbit.
There will be as much engagement in the crypto market as the front-end provides. There are currently no fully permissionless and trustless crypto prediction markets.
What we look forward to seeing is the potential success of applications like Parcl, which bring transparency to a fairly illiquid asset class. From fundamental principles, it appears to have the right structure to achieve its price discovery objectives.
The main application scenarios of Web3 include counterparty pool structures that support the construction of various prediction markets, and the successful application of prediction markets for better price discovery.
As cryptocurrency market caps grow and more people have disposable capital on-chain, the prediction markets industry could be profitable, or at least useful.
