Evolution Of Prediction Markets: From Middle-Ages To Blockchain Era
Prediction markets are markets created off of the idea of predicting the outcome of an unknown event in an exchange-traded fashion. The probability of a binary event occurring is traded among the participants. The market prices, then, reflect what the crowd considers the probability of the event to be ranging from 0-100%.
Payoffs of traders with different predictions are related to the unknown future outcome. Similar to the concept of crowdsourcing, prediction markets most commonly revolve around electoral and sports betting, however, the spectrum of the events to be predicted is widening more and more; especially when distributed ledger technology (DLT) gets involved.
The present article attempts to briefly chronicle the evolution of prediction markets starting from previous centuries and ending to achievements of today, where blockchain gets intertwined with prediction markets to offer newly-defined financial services for the world of betting, while at the end, some blockchain-based use-cases are mentioned and shortly described. Let’s get right into it.
Brief History Of Prediction Markets
The concept of people making predictions about the outcome of a future event dates centuries back. Perhaps since the dawn of Homo Sapiens intelligence, the curiosity, as well as the desire to manipulate uncontrollable events, gave birth to the idea of predicting, unarguably related to the enhancement of survival.
Although a fundamental human concept, later in time prediction-making, accompanied with confidence in one’s belief, became valuable and bred the concept of betting; prediction markets were born. Political betting has been probably the earliest form of prediction markets, as a historical account dating back to Middle-Ages refers to people betting on the papal successor, a practice that even at that time was considered old.
In more recent times, there are records of prediction markets related to election betting in Wall Street in 1884. During the 1988 US presidential elections, the University of Iowa introduced one of the first modern electronic prediction markets, the Iowa Electronic Markets (IEM). IEM is a group of futures markets operated by the University of Iowa Tippie College of Business for educational and research purposes only, which has an exemption from the Commodity Futures Trading Commission (CFTC) and is using real money to predict, among others, electoral and economic outcomes, often with greater accuracy than traditional polls and experts. Nevertheless, the robustness of IEM is questionable as speculators are limited to invest only between $5 and $500.
In 2001, InTrade was launched and run out of Dublin, Ireland, and was probably the most successful private effort, before it ceased trading in 2013 when the CFTC shut it down for financial irregularities. InTrade was a web-based platform that allowed members to trade real money on contracts related to several prediction markets with categories such as business, current affairs, finance, etc. After being shut down, there was an effort for “InTrade 2.0”, which was never able to run officially and by the end of 2014 all accounts were suspended and monies refunded.
Modeled after the IEM, the Victoria University of Wellington in New Zealand, launched PredictIt in 2014. Similarly to IEM, PredictIt is a nonprofit project using real money, while it has secured a no-action letter from the CFTC as it is operating for educational purposes only. The exemption is conditional on the basis that each question is limited to 5,000 traders and the market cap on individual investments per question is $850. In 2020, PredictIt reached more than 100,000 active traders.
Lastly, in 2003 the US Defense Department publicized through the Defense Advanced Research Projects Agency (DARPA) a new website, the Policy Analysis Market, where people could anonymously make predictions about future terrorist attacks in the country. Essentially, the US government was providing speculators with the opportunity to bet on forecasting assassinations, coups, and other terrorist attacks, all this sponsored by the government itself. Although the Pentagon defended this program claiming that similar futures trading had been proven to be effective in predicting other events such as election results and oil prices, the program was predictably shut down due to public outcry.
Accuracy Of Prediction Markets
As previously mentioned, prediction markets have been widely used for quite a long time, either from governments or individuals, and have mostly yielded effective results. Therefore, the key question here is where this accuracy is coming from and whether we should rely more or less on prediction markets.
There are various theories that explain why prediction markets can be accurate. According to James Surowiecki’s extensive work, the accuracy of such markets is based on the concept of ‘wisdom of the crowds’. This concept refers to the idea that aggregated beliefs over a certain matter can be more accurate than several individual beliefs. More specifically, the mean value of a large sample of estimates will likely be closer to the true value than the majority of the individual estimates; namely, collective wisdom.
Hear J. Surowiecki explain “Wisdom of the Crowds” in his own words. https://www.youtube.com/watch?v=pTI6u_gbilY
The power of collective wisdom can be utilized given three different conditions are met. Firstly, the diversity of information is important, as the wider the predictive spectrum is, the more accurate the mean will be. However, this can often be the failure point of prediction markets. It is not rare the case that a specific topic requires the respective specialized knowledge, which the majority of the people do not have, thus resulting in inaccurate judgments.
Secondly, the independence of decision influences the predictive accuracy. In some cases, though, personal biases, peer pressure, panic, and/or unreasonable rush may infiltrate and skew the judgments yielding unsuccessful predictors.
Lastly, the decentralization of the organization plays a great role as well. In comparison with expertise decisions, prediction markets are more decentralized by nature and, therefore, can be considered a more valuable source of harnessing collective wisdom with greater accuracy.
An important contributing point to the accuracy of prediction markets is the incentivization of individuals to make a good prediction. Different forms of incentives, such as economic or prize-like outcomes, may produce the right behavior needed for individuals to create powerful and accurate prediction markets.
To sum up, the accuracy of such markets is based on the efficient aggregation of a plethora of information – whether that is beliefs or data – which is obtained in a truthful manner through financial as well as other forms of incentives.
Virtual Markets In The Blockchain Era
So far, we have explored some historical background information of prediction markets and briefly analyzed where their accuracy comes from. In this last section, the relationship between prediction markets and distributed ledger technology shall be discussed, while also some related promising projects will be mentioned.
On the basis that prediction markets are as accurate as they are supposed to be, they can interact with blockchain in very exciting ways. As mentioned in the previous section, the decentralization of prediction markets is an important precondition for the successful predictive power of the market.
Given that public blockchains are decentralized by definition, this condition is met by default in cases where prediction markets are built on some blockchain. Perhaps the most important outcome of this combination is that the long-lived flirting of prediction markets with gambling sanctions can be overruled and censorship is almost impossible, as it must be a tough task for law-enforcement agencies to shut down all activity of the whole network.
This does not mean that companies can/will now exist for profit, but, hopefully, that existing for-profit services will be replaced by automated utilities resistant to censorship.
Another important outcome of the interaction of prediction markets with blockchains is, essentially, the crowdsourced-ness of the ‘truth’. As there is no central authority deciding what actually happened, blockchain-based prediction markets are very difficult to corrupt given that no centralized interests are involved.
Additionally, the combination of DLT and prediction markets accounts for equality of participation – something that is very much absent in the traditional prediction markets. Blockchains enable more efficient, transparent markets as anyone can participate and/or create their own market.
Before mentioning a few prediction markets that run on some blockchain, it is important to discuss the importance of the “oracles”. A short definition of an oracle is a medium that reports off-chain data, that is, information from real-world events, to the blockchain so that smart contracts can be fulfilled.
Prior to blockchain, central reporting entities would play the role of gathering data necessary for deciding an outcome. Today, the market itself running on the blockchain is using the weight of its own utility to compensate a distributed pool of users for acting truthfully.
Once the relevant information is concentrated, the if-then clause is automatically fulfilled by the smart contract and both parties of a contract of a certain market are rewarded respectively without intermediaries.
Now that some of the fruits of the merging between DLT and prediction markets have been discussed, let’s take a look at some use-cases from prediction markets in the blockchain realm.
After crowdfunding in 2015, Augur was launched in 2018. Augur is a decentralized prediction market platform built on the Ethereum blockchain and was developed by the Forecast Foundation. It simply allows users to create prediction markets on their own utilizing the power of smart contracts. Substituting custodial betting platforms, Augur offers a transparent exchange with no limits or requirements, that is community-owned and operated.
Augur is not a prediction market itself, but a peer-to-peer prediction market protocol and a decentralized oracle. The use of smart contracts enables automated processes, such as the migration of off-chain data to the blockchain as well as payouts, without intermediary interference, while the fees are kept around 1%. Users can utilize several industry-standard ERC-20 compatible wallets, such as MetaMask, AlphaWallet, or Trust Wallet, to perform actions and transact within the platform.
In addition, Augur has its own native token called Reputation (REP). REP is used to incentivize users to report truthful information and back their reports with tokens. REP holders must perform work – in the form of staking their REP on correct outcomes- to get rewarded with a portion of the market settlement fees. Those users that vote on incorrect answers get penalized by losing some staked value, whether that is money or REP.
Examples of already existing prediction markets include markets regarding events such as company performance, election results, sports results and even natural phenomena, where users can purchase shares that either support or refute the proposed outcomes and, in turn, fees are returned to market participants in an automated, fair and decentralized manner.
Lastly, despite Augur being a user-friendly platform that also has its own trading app, there is a downside in that the entire system is built around REP holders. Consequently, this might create scaling issues, as the number of REP holders is inherently limited compared to the potentially infinite number of different prediction events, therefore yielding an overall lower efficiency due to the length of time needed to reach consensus after a prediction event has occurred.
Similarly to Augur, Gnosis has been also built on the Ethereum blockchain and is utilizing the power of smart contracts. However, it is not the token holders that are responsible for judging the event outcomes, but a centralized oracle.
Essentially, it is the platform itself that is accountable for the accuracy and efficiency of the judgments, having the potential to grow to a decentralized version of Google where users’ knowledge from a universal pool gets aggregated to elicit truthful results.
Recently, the team launched the Gnosis Protocol. That is a fully permissionless decentralized exchange that is enabling ring trades for liquidity maximization. These innovative tools are, basically, order settlements sharing liquidity across all orders instead of a single token pair.
Ring trades are therefore uniquely suitable for trading in prediction markets as they empower users to create, trade, and bet in simple as well as combinatorial markets with the use of conditional tokens. Such tokens allow for trading on event outcomes that are conditional on other outcomes with low transaction costs and seamless experience.
For example, let’s suppose there is a market that’s asking whether in a soccer match team A will win against team B. In a separate market, there is a question regarding whether there will be more than 3 goals scored in the match between team A and team B.
Given the two questions are correlated, the value of the conditional tokens is dependent on both outcomes of their respective questions. That way, users can speculate and trade on both the combined outcome as well as hold separate balances for each outcome.
Bodhi is a prediction market platform that is focused on the Chinese market. Like other prediction markets, Bodhi allows users to create and trade on markets for any real-world event, while at the same time, it offers the possibility to hedge against certain risks.
Contrary to previously mentioned prediction market platforms, Bodhi is built on the Qtum blockchain, however, a version on the Ethereum blockchain is currently under progress. Additionally, the team is claiming that the project was put forward to overcome the shortcomings of Augur and Gnosis platforms, that is, scalability and information accuracy issues.
A third-party oracle is set to be in charge of determining the prediction results in an autonomous manner. Making use of the native token (BOT), token holders can take over the voting process in case the oracle fails to deliver accurate judgments.
Finally, a unique innovative feature of Bodhi is the replaceable oracle scheme. It allows any user to challenge and even revoke bet results that were set by individuals. These aspects empower Bodhi to offer more independent and efficient prediction markets.
Being designed as a sidechain of Bitcoin, Hivemind (or Bitcoin Hivemind) is an open-source p2p oracle protocol, which creates and manages prediction markets in a scalable, censorship-resistant and trustless manner, so that Bitcoin users can speculate in on-chain prediction markets. Similarly to other prediction-market platforms, anyone can create a market about anything, avoiding the interference of intermediary parties.
Although most prediction market platforms are focused on all kinds of markets, Hivemind aims to provide valuable insights for governmental issues related to multi-factor decision making. In its whitepaper, the project claims that the platform is a combination of three existing ideas, that is, a software protocol, an event derivative, and a “future Wikipedia”.
More specifically and according to its creators, Hivemind can be useful in several ways, such as a decentralized payment system, a global source of reliable information, a platform where personal experience and expertise can be utilized to generate profit, and a community effort to create a more rational and scientific world.
Hivemind has its native token, VoteCoin, which voters use to resolve event outcomes and come to agreements, while similarly to Augur, voters get rewarded for reporting accurate results but also punished for inaccurate reporting.
In general, people’s tendency to forecasting and trying to make money out of it is not likely to disappear, not even get reduced. At the same time, the need for more transparently accurate information is greater than ever and it will become even greater.
Blockchain-based prediction markets offer a new solution, where markets are unconditionally more accessible and anyone can take the role of validating truthful events by getting incentivized with financial rewards.
And all of this, in a decentralized fashion.
According to the former Coinbase CTO, “Blockchain-based prediction markets may be the one force strong enough to counterbalance the spread of incorrect information on social media. They give people a financial incentive to seek the truth and then protect them with the twin shields of pseudonymity and decentralization.“
Although the future is unknown, the benefits of prediction markets have already proven to be valuable. Putting forward the collective wisdom of mankind, a more prosperous future lies ahead and it’s in our hands to experience it.