# Opinions are Priced: Prediction Markets
"Fortune tellers"- Who? I let Prediction Markets speak the future.

## Prologue
Historically, people have been betting on various things like board games, political events, wars, etc. It was legal at times, also illegal sometimes but still people were betting. Betting has been constant but the things people bet on and the use cases of it has evolved. In modern times, people usually bet on events like Financial events, Elections, Sports, etc. Combining this with the power of Decentralization will make the rise of Prediction Markets inevitable.
Prediction markets are judgement-based markets that make predictions for binary events. These markets are really good at predicting results of future events.

## Index - The Path of Prediction
1. The Timeline of Betting
2. Introduction to Prediction Markets
4. Prediction Market & Other Speculation Systems
6. Decentralized Prediction Markets
7. Let's Enter Polymarket!
8. PnP & Uniswap: a simile
9. Prediction Markets: A Deeper Dive

## The Timeline of Betting
Have you come across those "Mayans predicted the world will come to an end on this day." or similar information? Turns out this isn't a suprising phenomenon. Many ancient civilizations across the world had their own Astrological/ Future prediction systems using natural events, horoscopes, etc as a guiding metric before taking future steps. Humans over the course of a long long history had the tendency to predict or speculate about future events.

As societies evolved, this tendency of speculation became the backbone of taking any action, as everything felt unpredictable. Farmers bet that plating seeds will yield good crops, Traders bet that goods will be carried across oceans safely, etc. and even science evolved by betting on trying out experiments based on limited information. Naturally, this speculative trait tagged along and many systems to predict or speculate on the future events, over the course of time, were newly developed and some even got better. Combining this with the added incentive of receiving prize money on predicting correctly, gave rise to betting.

Since, humans have been betting since the dawn of time, just take a look at the timeline to have an idea of it.

### History of Betting Before the 1900s
- **Before Civilization Existed**
-> Barter is the most primitive form of exchange. But more that that, it was a form of betting.
-> Let's say, a hunter was trading a piece of meat with a peasent for some herbs. Both sides are essentially betting on intuition. The hunter bet that the herbs had the medicinal value worth giving up a piece of meat. The peasent bet that the meat would be more worth than the herbs for him in the moment.
-> Even some pre-human species made primitive bets through reciprocal altruism. Like, one individual would share food hoping to be repaid later. If not repaid, the bet failed.
- **Ancient Times, Before 1000 CE**
-> Mesopotamia, Egypt & China -> Earliest evidence of dice, rudimentary gambing tools and chance games.
-> Ancient Greece & Rome -> Widespread betting on gladiator games, chariot races and board games. Despite legal restrictions at many places, it was often socially accepted.
- **Medieval Period, 500-1500 CE**
-> Gambling linked to religious festivities, jousts and tournaments.
-> Legal at some places and illegal at some, depending on the region.
- **Early Modern Period, 1500-1700s**
-> England: Betting on political events, royal births, duels and wars. The nobles wagered on jousts and wars, while peasents bet on the games of chance.
-> India: Indigenous betting games such as various dice games, rooster fighting. Also, sports and horse racing betting started on full-fledged mode.
- **Late 1700s–1800s**
-> South Sea Bubble (1720): South Sea Company, the company's stock price inflated dramatically. It's a speculation mania that ruined many British investors in 1720. Some also consider it the World's first Ponzi Scheme.
-> Paris & London (1800s): Stock gambling, wars and election bettings were happening and were quite common in informal markets.
-> United States (19th Century): Political betting such as for Presedential Elections were taking place. Horse racing emerged as a formal betting market.

### History of Betting After the 1900s
- **20th Century**
-> Early 1900s: Horse racing and off-track betting grew in legal structures globally.
-> 1920s USA: Las Vegas legalized gambing, became a betting hub.
-> Post World War: State lotteries, sports betting and casino gaming expanded in Europe and America.
-> Late 1900s: Rise of computer models, televised sports and betting exchanges, early online betting began in later '90s.
- **21st Century**
-> 2000s: Internet revolutionized access, online casinos, sportsbooks, poker rooms emerged.
-> 2010s: Mobile betting and global regulatory changes happened across the world. In India, Online betting, lottery and horse races betting are legal unlike other forms.

### The Rise of Prediction Markets
- **Late 20th Century: Academic Beginnings**
-> 1988: [Iowa Election Markets](https://en.wikipedia.org/wiki/Iowa_Electronic_Markets), operated by University of Iowa, launched to study election forecasting via low-stakes trading. This pioneered the idea of markets as predictors.
- **2000s-2010s: More Platforms Emerged**
-> Projects like Intrade, PredictIt, Augur, etc. explored real-money or crypto-based event markets.
-> Platforms like [Metaculus](https://www.metaculus.com/) came into existence, which is a community forecasting platform. It's not financially incentivized, it's about prediction accuracy and the scores are ruled on that. This platform is mostly used by researches and academics.
- **2020s: On-chain and Scalable Prediction Markets**
-> Polymarket, Manifold and Kalshi became prominent, offering markets on elections, finance, tech events and world affairs.
-> Polymarket is currently the most popular blockchain-based prediction market which got a big boom during the US Presidential elections.
-> Opinion trading is an emerging trend in India where users wager on outcomes of real-world events, mostly sports on platforms like Dream11.
-> Blockchain enabled decentralized, censorship-resistant markets, often outperforming polls and surveys.

## Introduction to Prediction Markets
### What are Prediction Markets?
Prediction Markets are judgement-based information markets used to forecast the outcomes of future events, often those with binary possibilities like "yes/no" or "will/won't happen".

In prediction markets, people buy and sell contracts based on what they think will happen. The more people believe in a certain outcome, the more valuable that contract becomes. So, the contract's price acts like a live estimate of how likely that outcome is to happen, based on the crowd's combined beliefs. The final contract prices just before the conclusion of the binary event very accurate. These markets harness the crowd's wisdom and often outperform traditional forecasting methods in accuracy.

The contracts here can be a shares, points or any other valuable thing. Basically, it's free money in your pockets, given that, you have the ability to predict the future result of an event.

### Example: Prediction of Election Results
The principle of Prediction Market: Choose the correct binary option (Yes/No) or (Will/ Won't happen), you will be rewarded if it turns out to be correct.
One of the interesting things in the prediction market is that the Market Price of the outcome shares can be interpreted as roughly similar to the probability of that outcome occuring.

Following example is the basic working mechanism, it will be expanded in further sections.


- Let's say, two people are fighting an election, Pariston and Cheadle.
- In this case, the contract is a share.
- Let's say, a share is worth 1 USDC.
- At a moment, Market Prices of outcome shares in the Prediction Market are currently:

| \ | **Pariston** | **Cheadle** |
|-|-|-|
| **Price per Share** | 0.8 USDC| 0.2 USDC |
| **Market's Prediction: Probability of Victory** |~80% | ~20% |

- The share prices indicate: At that moment, the market thinks that Pariston has 80% chances of winning the election & Cheadle has only 20% chances of winning the electon.
- You are a trader. There are 3 simple cases:
1. You think the market is wrong
2. You think the market is correct
3. You go along with the market's predictions

#### 1. You think the market is wrong
- Let's say, you have an extra information in your hand that the market dosen't have or you have considered some extra factors significantly affecting the results of the election that the market didn't consider.
- You think that Cheadle has a 95% probability of winning and the market is wrong.
- In that case, you buy the Cheadle shares at 0.2 USDC each and try to make a profit of 0.8 USDC per share.
- You have a 95% chance of getting the shares you bought at 0.2 USDC per share and a 5% chance of not getting any shares and losing 0.2 USDC per share for all the shares that you bought.
- Let's say, you bet with these market shares and probabilities again and again, you will win most times but also lose sometimes. Hence, on an average you will make a profit of 0.75 USDC per share:
$$Average \space Profit = 0.95 \times 0.8 + 0.05 \times (-0.2) = 0.75 \space USDC$$
- But, what if you go against your logic and decide to buy the Pariston shares even with Pariston's losing probabilities? You do this again and again. You will lose most times. Hence, on an average you will make a loss of 0.75 USDC per share:
$$Average \space Loss = 0.95 \times (-0.8) + 0.05 \times (0.2) = -0.75 \space USDC$$

#### 2. You think the market is correct
- Let's say, you agree with the market on the victory candidate but not on the probabilities.
- You think that Pariston has a 95% probability of winning.
- In that case, you buy the Pariston shares at 0.8 USDC each and try to make a profit of 0.2 USDC per share.
- You have a 95% chance of getting the shares you bought at 0.8 USDC per share and a 5% chance of not getting any shares and losing 0.8 USDC per share for all the shares that you bought.
- Let's say, you bet with these market shares and probabilities again and again, you will win most times but also lose sometimes. Hence, on an average you will make a profit of 0.15 USDC per share:
$$Average \space Profit = 0.95 \times 0.2 + 0.05 \times (-0.8) = 0.15 \space USDC$$
- Let's try the losing bet again. On an average you will make a loss of 0.15 USDC per share:
$$Average \space Loss = 0.95 \times (-0.2) + 0.05 \times (0.8) = -0.15 \space USDC$$

#### 3. You go along with the market's predictions
- Let's say, you agree with the market that Pariston is winning the election with similar probabilities.
- Then, you buy Pariston shares at 0.8 USDC per share to try make a profit of 0.2 USDC per share with 80% success rate and have a 20% probability of losing all shares and money you put into the market.
- You bet again and again this way. On an average, you will neither make a profit nor a loss:
$$Average \space Profit = 0.80 \times (0.2) + 0.20 \times (-0.8) = 0 \space USDC$$
- This third case indicates that you will need an analysis of your own and your own prediction probabilities for betting on Prediction Markets in order to increase average difference (profit or loss) in the cash flow. So, always betting on the probabilities of market is not the best choice:
$$Trader's \space Probability \neq Market's \space Probability$$

### Check out some Prediction Markets!
- **Metaculus**
-> Metaculus is a community-driven judgement-based forecasting plaform.
-> Continuous and probabilistic predictions take place. One must predict the occurence of an event with a probability percentage instead of choosing from a binary choice contract.
-> Aggregates expert forecasts with scoring incentives with an emphasis on accuracy and calibration.
-> Focuses on science, technology, global risks, geopoilitics, etc. Popular among academics and researchers.

- **Kalshi**
-> Kalshi is a regulated CFTC-approved (Commodity Futures Trading Commission USA) prediction market.
-> Contains binary prediction contracts (Yes/No) or (Will Happen/Won't Happen)
-> Focuses on Macroeconomics, politics, inflation, job reports, etc. contains all the legally tradeable events in the US as it's regulated.
-> It's a prediction market operating with real finance & economic value.

- **Iowa Electronic Markets (IEM)**
-> IEM is a research-focused real money market operated by University of Iowa.
-> It has both Binary and Proportional prediction payouts. Which means, more accurate predictions gets higher payouts.
-> Mostly focused on US politics and economics. Historical track of accurate prediction in elections.
-> IEM forecasted better predictions than mainstream medias during US Presidential Election. We'll discuss one such case, futher down this blog.

- **Dream11**
-> Dream11 can be considered an Opinion Trading Platform involving real-money, marketed as a Fantasy sports platform.
-> It is an online game where users create a virtual team of real-life players and earn points based on the performances of these players in real matches.
-> Focused on multiple games such as Cricket, Football, Basketball, etc.
-> Not exactly a prediction market by definition but it's a skill-based speculation game on sports.

- **Probo**
-> Probo is an Indian-based Opinion Trading Platform with a gamified UI and real-money rewards.
-> It has Binary (Yes/No) Prediction Type.
-> Focuses on Cricket, entertainment, news, politics.
-> It targets casual users rather than researchers or analysts. Hence, it's not marketed as a Prediction Market even if it is a binary judgement-based betting app.

- **PredictIt**
-> PredictIt is a New Zealand-based online prediction market that offers exchanges on political and financial events. But only US citizens can bet on this site.
-> Focus was mostly around the political outcomes of US politics. It was used extensively for forecasting elections.
-> Currently, it's partially operational and in a legal battle with CFTC.

## Prediction Markets & Other Speculation Systems
### Prediction Markets vs Gambling
Comparing opinion trading to gambling is flawed and misleading. Let's first take a look at an example of a Gambling Platform.

#### Stake
-> Stake is an online casino and sports betting platform, operating on both, Crypto and Non-Crypto currencies.
-> It is a Gambling platform (chance-based).
-> Contains an entire marketplace of casino games. It also has sports betting and esports wagering.
-> It's one of the world's largest online casinos with live betting, global brand partnerships, shady regular promotions and VIP reward system.

#### Let's Compare
| **Aspect**| **Prediction Markets**| **Gambling**|
| -| -- | --|
| **Core Principle** | Driven by skill, analysis, and informed judgment|Based purely on luck or chance|
| **Working Mechanism** | Participants use data, trends, and probability to make informed bets |Outcomes are random, chance-based like dice rolls or lottery draws |
| **Main Goal** | To crowdsource collective intelligence for accurate forecasting | Primarily for entertainment or expecting quick monetary gain |
| **Regulation** | Considered a financial product (e.g., by CFTC, SEBI) and regulated accordingly | Falls under national or local gambling laws |
| **Economic Role** | Helps make decisions in government, finance, markets, and society | No broader economic or strategic impact |
| **Centralization Effect** | Both centralized & decentralized markets available and returns reflect real market signals and sentiment| The centralized entity (casino/online gambling platform) has built-in advantage and always profits |

### Prediction Markets vs Mainstream Media
Mainstream media also makes predictions about future events with the help of data, models or pundits. According to studies, prediction markets are often more accurate than other methods of prediction because they combine news, polls, and expert opinions into a single package that represents the market’s collective view.
There are many examples which may indicate Prediction Markets perform better than Mainstream media. Check this popular one.

#### IEM (Iowa Electronic Markets) vs Nate Silver's Forecast
During the 2008 U.S. Presidential Election, a comparison was made between Nate Silver’s polling-based statistical model and the Iowa Electronic Markets (IEM) prediction market. It was observed that the prediction market demonstrated higher accuracy than the poll-based model.

Nate Silver’s model relied on data collected through public opinion polls and used statistical techniques to forecast outcomes. In contrast, the IEM prediction market derived its forecasts from the behavior of traders. While a statistical model is limited by the number and type of variables it can include, a prediction market aggregates a broader range of information. Traders can consider, not only poll data but also their own individually calculated affecting factors and other factors, such as last-minute news, biases the model may contain, when placing trades in the market. Because of these, the judgment of market turned out to be better than the polls.

#### Let's Compare!
| **Aspect** | **Prediction Markets** | **Mainstream Media** |
| ----------------------- | ------------------------------------------------------------------ | ----------------------------------------------------------------------------- |
| **Source of Forecasts** | Collective intelligence via real-money bets or opinion contracts | Editorial teams, analysts, or pundits |
| **Incentive of Prediction** | Traders trade for predicting the outcome correctly, they lose money for being wrong | Often incentivized by attention, ratings, or ideological bias |
| **Motivation of Forecasting** | Designed for probability estimation and aggregation of beliefs | Designed for reporting, commentary, and narrative shaping|
| **Accuracy** | Tends to be high due to wisdom of crowds and monetary risk | Can vary, depends on statistical models or even biased on purpose |
| **Transparency** | Prices are public and reflect real-time consensus | May lack transparency in sourcing or reasoning |
| **Dynamic Updating** | Market prices shift with new information instantly | Forecasts may lag behind on recent breaking narratives|
| **Bias** | Lower, since money is on stake| Higher, due to editorial agendas, political leanings, or clickbait incentives |
| **Examples** | Polymarket, Kalshi, IEM, Metaculus | CNN, Fox News, The Times, NDTV, Republic|

### Prediction Markets & Opinion Trading
Opinion trading is a form of Prediction Market. While opinion trading and prediction markets share functional similarities, their differences lie in regulatory status, structural design, and intended purpose.

| **Aspect** | **Prediction Markets** | **Opinion Trading** |
| -- | -- | --- |
| **Regulation** | Often regulated as financial instruments (e.g., by the CFTC in the U.S.) | May operate without formal regulatory oversight, leading to legal ambiguities |
| **Structure** | Structured trading mechanisms with clear contract specifications | May have less formal structures and a gamified UI|
| **Purpose** | Designed to aggregate information for forecasting and decision-making | Often positioned as entertainment or informal betting platforms |
| **Examples** | Kalshi, Polymarket | Probo, Better Opinions |

## Decentralized Prediction Markets
### Centralization vs Decentralization
It's assumed that you know the definitions and difference between Centralization & Decentralization. Still, take a small look.

| **Centralization** | **Decentralization** |
| ---------------------------------- | ---------------------------------- |
| Users connecting to a hub | Network of peers (no middleman) |
| One big hub, all control in center | Everyone is connected directly, peer-to-peer |
| Single point of failure | No single point of failure |

#### Visual Representation

### Prediction Markets, Centralized vs Decentralized
In order for the market to be more accurate and trustworthy, it needs to be decentralized. Because, if the market is centralized, there's a chance that it may be subject to manipulation or censorship.

#### Let's take a look at the differences!
|**Aspect**| **Centralized Markets** | **Decentralized Markets** |
| -|------------------------ | ---------------------------------------- |
|**Product Model**| Run by a single company with centralized authority| Run by smart contracts + communities, making the market decentralized|
|**Trust Model**| Require user trust towards the platform or company| Trustless execution on blockchain, hence no problem of trust issues|
|**Control**| May restrict access, meaning that the user dosen't have full control| Usually open to all with a crypto wallet providing freedom and control to users|
|**Censorship**| Can censor/close markets or tamper with the market statistics| Resistant to censorship and hence making the market more trustworthy|

Hence, Decentralization of Prediction Markets must happen.
Some decentralized prediction markets are Polymarket, Augur, Omen, [PlotX](https://plotx.io/), etc.

## Let's Enter Polymarket!
### Introduction
Meet [Polymarket](https://polymarket.com/)! It's an American cryptocurrency-based decentralized Prediction Market and also the world's largest Prediction Market. And due to it's decentralized nature, Polymarket is way more reliable and trustworthy than other centralized prediction markets. Polymarket uses USDC, a stable coin backed by the US Dollar on the [Polygon](https://polygon.technology/) network ensuring fast and reliable transactions for trading. It offers a platform where investors can place bets on predictions of various events ranging from politics, economics, awards, weather, movies, legislative outcomes, etc.

In 2024, the outcome of U.S. elections became the most active market on the platform. This event combined with it's decentralized nature gave a big boost of money flow and popularity to the platform.

### Sequence of Functioning
1. Market Creation Proposal
2. Market Creation
3. Trading & Prediction
4. Event Occurs
5. Market Resolution Proposal
6. Market Resolves
7. Share Payout

### Market Creation in Polymarket
In Polymarket, the markets are created by the markets team by taking inputs from users and community. Market suggestions can be given on discord or X. Market creation may be controlled but betting is permissionless. This is slightly contrasting to Omen, in which, even markets can be created by anybody on any topic in a completely permissionless way.

The market creation is kept in control by Polymarket to:
- Maintain clarity and trustworthiness in resolution of the market and to avoid any conflict or ambiguity in the resolution.
- Avoid spammy, low-quality or even illegal markets which can be risky.
- Ensure bare minimum regulatory compliance (especially in the US), so that the platform is not shut down by government.

Some things must be taken care of when proposing one on Polymarket:
- The Market Title should be provided.
- Evidence of demand for trading in that market.
- The outcome of the market must be known by its resolution source for clarity+quality of resolution and resolution date/time. If the resolution source is ambiguous or resolution time is unpredictable the market may be considered invalid.
- If a market can have different interpretations leading to different outcomes, the necessary clarifications should be provided.
- Include as much important information as possible.

Even after taking these measures to create a market, sometimes, there are disputes in the market resolution which need to be handled.

### Calculation of share prices in Polymarket
The shares are traded by order book method. In short, in a order book method, buyers bid some amount of money to buy a share and sellers ask a certain amount of money for the share they are selling, when bid-ask prices match or buyer & seller agree on a common price, the trade executes.

Prediction Markets try to keep this condition:$$Price \ of \ a \ share = Probability \ of \ that \ outcome$$

**Let's start from the creation of a market!**
- Initially there are zero shares and no predefined prices of any shares.
- The first trade placed is a limit order (a trade which dosen't execute till the price demands are met and executes automatically once met). The users placing limit orders are called Market Makers.
- When the sum of both orders, 1 YES and 1 NO add up to $1.00, the order is matched and converted to 1 YES share and 1 NO share
- This becomes the initial market price and probability!
- Prices on Polymarket are a function of time.
- Prices displayed are midpoint of the Orderbook's bid-ask spread.
- If the spread is wider than $0.10, the last traded price is used and shown as the probability.

**Example:**
- Title: Will Pariston become the new Chairman over Cheadle?
- There are 2 shares -> YES and NO.
- Traders place their demands
-> Trader 1: YES share for $0.50
-> Trader 2: NO share for $0.80
-> Trader 1: YES share for $0.60 <-
-> Trader 3: NO share for $0.50
-> Trader 2: YES share for $0.40 <-
- As we can see, Trader 1 & Trader 2 get their sum equal to $1.
- Trader 1 bought a YES share for $0.60 and Trader 2 bought a NO share for $0.40.
- The initial market price and probability is set.
- Some time passes, results are becoming unpredictable, the bid-ask gap spreads, least bid by buyers for YES share is $0.7 & highest ask by sellers is $0.85. The difference gets wider than $0.10.
- The last trade, YES share was traded at $0.76. This information is put into use by the market. The market will forecast the current probability of the event happening as 76%. The price of NO shares will be approximately complimentary to the YES shares: $$Price \ of \ NO \ Shares \approx 1-(Price \ of \ YES \ Shares)$$

### Market Resolution
Market Resolution happens when a real-world event reaches a definitive outcome. Market resolves to either a YES or NO. Markets are resolved according to the market’s pre-defined rules. When the market is resolved, trading is halted, people holding winning shares get $1 per share and the losing shares will become nothing. Markets are resolved by the UMA Optimistic Oracle (A service which processes real-world data to smart contracts), a decentralized smart-contract based optimistic oracle.

#### Market Resolution Process
- The trading in market has been halted as the event has come to an end and the results are out and declared.
- The market is resolved as per the pre-defined rules. Hence, a proposer (any user of the platform can choose to be a proposer) must propose an outcome which involves staking $750 bond in USDC with the contract.
- Once the bond is put up, it will enter the verification process. UMA oracle will ensure it's proposed properly. If proposal is accepted, the proposer will receive the bond money back plus a reward.
- Without any disputes, the market is auto-resolved and the outcome proposed is accepted and the market is finalized and everybody gets their share of money as per the results.
- Now, what if someone disputed? Once a market is proposed for resolution, it goes into a challenge period of 2 hours. If someone disputes by proposing the same amount of $750, UMA oracle starts a formal dispute process. This begins a period of dispute for 24-48 hrs time frame.
- Discussions happen during this period, voting happens every other day and evidences are also contributed to the discussion. After this debate period, one of the four following outcomes take place.

#### Outcomes
- Proposer wins
-> Proposer gets bond amount back + 50% of disputer's bond as bounty.
-> Disputer loses their bond.
- Disputer wins
-> Disputer gets bond amount back + 50% of proposer's bond as bounty.
-> Proposer loses their bond.
- Too Early
-> Event hasn't occured yet (e.g.-> result not declared, match still ongoing)
-> Disputer wins
- Unknown / 50-50 Outcome
-> Rare case where no clear outcome is possible.
-> Market resolves to 50% Yes / 50% No.
-> Disputer wins

#### Example:
- Let's say,
-> Market Title: Will Pariston become the new Chairman over Cheadle?
-> Market Condition: Win Election & Become Chairman.
-> Resolution Source: Election Results
- Finally, election results are out, Pariston won the election and went on for the formal process of being appointed as a Chairman.
- A resolution is proposed by a proposer for YES as Pariston won the election. Dispute window opens.
- But, a twist of events happened. Pariston, during the Chairman appointment process, passed the Chairmanship to Cheadle despite winning the election. This lead to confusion among people, because it was assumed that winning the election meant automatic Chairmanship.
- Current Conditions Outcome:

| Conditions | Pariston | Cheadle |
| - | :-: | :-: |
| Win Election | Yes | No |
| Become Chairman | No | Yes |

- A dispute is raised by a disputer that market should resolve YES as the second condition was not met and the market goes into dispute period.
- Some traders argue Pariston won the election so it should resolve in YES and some argue Cheadle became the Chairman plus the market conditions weren't satisfied, hence NO.
- The market conditions were not met as shown in the above table. Pariston didn't become Chairman. Proposal misinterpreted or ignored part of the market condition, and the outcome is actually NO.
- Outcome type: Disputer Wins
-> Proposer loses bond.
-> Disputer gets bond back + half of Proposer's bond.
- Another thing to notice: Although the market correctly predicted that Pariston will win the election, as most traders implicitly assumed victory as Chairmanship, it didn't consider the nuance of Pariston voluntarily passing the Chairmanship to Cheadle. This signifies, even if the probability forecast is accurate based on available information, unusual or exceptional events can shift the final outcome.

## PnP & Uniswap: a simile
### The Simile
**PnP (Predict and Pump)** aims to make prediction markets permissionless, removing the centralization around:
- Market creation
- Liquidity provisioning
- Market settlement or resolution (by using decentralized oracles)

Just like **Uniswap** made spot trading permissionless, removing the need for centralized listing or order book management.

### Functional Analogy
| Function | Uniswap (Spot Trading) | PnP (Prediction Markets) |
| ---------------------------------- | ------------------------------ | ------------------------------------------------ |
| **Permissionless Market Creation** | Anyone can create a token pair | Anyone can create a prediction market |
| **Decetralized Settlement or Resolution** | Swaps are settled via AMMs (Automated Market Makers) | Prediction outcomes settled via decentralized oracles |
| **Liquidity Provision** | LPs (Liquidity providers) provide capital, earn fees | LPs stake on both sides, earn on trading spreads |
| **Price Discovery** | Done via token trades & slippage | Done via market-based probability (YES/NO pricing) |
| **KYC / Censorship Resistance** | No | No |
| **Trustless Incentive** | Fees to LPs & arbitrageurs | Profits to skilled traders & LPs |

### Why is this transformative?
- Opens up new markets fast by removing the centralized mechanism which makes things slow.
- Avoids centralized moderation which can be biased or manipulative.
- Crowdsources prediction from globally distributed, informed and skilled participants.

**If Uniswap democratized trading, PnP democratizes forecasting.**

## Prediction Markets: A Deeper Dive

### Conditional Prediction Markets
How will prediction markets help an organization or even the government in better decision making for taking new initiatives and making new policies? Presenting you, Conditional Prediction Markets!
Conditional Prediction Markets can assist the government as decision making tools.

Prediction markets forecast the outcome of future events. But their utility extends beyond mere prediction, especially in the form of Conditional Prediction Markets. Conditional Prediction Markets allow participants to trade on outcomes under specific conditions or scenarios. Unlike standard markets that predict the likelihood of a single event, Conditional Prediction Markets compare two scenarios or conditions for the same event and hence creates two markets which checks the probability of same event occuring with YES and NO possibilities.

#### Why Use Conditional Prediction Markets?
1. Compare Impact of Multiple Choices
2. Reduce Decision Making Risk
3. Understand Real-Time Demand of Crowd
4. Encourages Transparency of Decisioning
5. Democratize the Decision Process

#### Example
- Let's see a market which helps the government for decisions in taking initiatives.
- Problem: It's an upcoming metro city and the city traffic is rising rapidly. The traffic is exhausting and slow people down, causing functional issues for organization and government. Authorities consider two options:
-> Expand metro rail network or Expand road infrastructure
-> Will the average commute time drop by atleast 40 mins within 2 years?
- The Market and the probabilities are:

| Event & Market Probabilities | Expand metro rail network | Expand road infrastructure |
| :-: | :-: | :-:|
| **Event:** Will the average commute time drop by atleast 40 mins within 2 years? | YES \| NO | YES \| NO
|**Market Probabilities**| 70% \| 30% | 50% \| 50%

- The authority takes the probabilities of this market into consideration and based on higher market confidence (70%, greater than all), takes the further step by choosing to: Expand metro rail network
- Now, the Expand road infrastructure market is closed. The people who placed their bets on Expand road infrastructure, both YES and NO, get their money back. The market then becomes a standard Prediction Market of Expand metro rail network.
- This way, Conditional Prediction Market has helped the government in decisioning.

### Exploring users of Prediction Markets

#### Zero-sum Game
Ideally, Prediction Markets are considered a Zero-sum Game. In a zero-sum game, one participant's gain is exactly equal to another's loss. The total amount of value in the system remains constant, it just shifts between participants. So by design, everyone can't win! Adding upon this, prediction markets are conducted on a platform which leads to charging of platform fee on users. Hence, it's a slightly negative-sum game for the users.

Each share is settled for $1 if correct & $0 if incorrect. $$Price \ of \ NO \ Shares \approx 1-(Price \ of \ YES \ Shares)$$ Suppose, a user buys a YES share for $0.40 in a market and the event dosen't occur, he loses the $0.40 share. On the counterpart, the user who bought a NO share in the same market for $0.60 gets the $1 share.

To compensate for the negative-sum nature of prediction markets, traders and plaform can take some measures.
- Individual Traders
-> Place informed bets, not opinionated. Use Probabilistic Thinking
-> There are arbitrage opportunities & users can also earn from bid-ask spread profits.
- Platforms
-> Governments, organizations, etc. can pay or provide liquidity to create market. Increasing profit increases trader's incentive to trade.

#### Types of Users
There are variations in types of traders betting on Prediction Market.
- Retail participants often bet on intuition, bias or emotion.
- Experts, researchers place sharp bets using data-analysis and the exclusive information they have. Over time, the experts win and pull the value from the retail participants.
- Some users act as Liquidity Providers (LPs). Instead of betting on the outcomes, they earn from bid-ask spreads.
- Arbitrageurs trade on short-time price discrepancies in same market or even between different markets or platforms to make profit.

#### Biases
Traders are faced with biases while placing a trade. Favorite-Longshot bias is a very popular one.
- Confirmation Bias -> Traders tend to give more weight to information that confirms what they already believe, while ignoring contradictory evidence.
- Availability Heuristic -> Assessing the probability of an event based on how easily it comes to mind, rather than statistics.
- Favorite-Longshot Bias -> Traders betting too much on unlikely outcomes (Longshot) because of the high payoff it they win and underbetting on likely outcomes (Favorite) because of less reward.
- Bias from environmental factors -> Factors, such as weather and atmospheric conditions can create biases that sometimes significantly affect their predictions.

### Evil Actors
Prediction Markets are not completely free of manipulation. Some people may manipulate the market for the sake of their profit or any other personal benefit, making the market prediction inaccurate as a result. Let's take the above example of city traffic and observe the evil actors roll their own dice.

#### Example 1
- A big shareholder, of the company which will take up the project of expanding metro rail networks if the market predictions are in their favour, wants the project for his company so that he can bag a big chunk of money.
- He wants the prediction market to be in his favour, to achieve that, he buys insane quantity of YES shares, in favour of metro rail. This skews probability distribution of the market.
- By doing this, he ensures that the case of metro rail looks promising and hence, higher chance of getting selected as the new initiative by the authorities.
- After securing the project for his company, nothing else matters to him. Even if the prediction about the reduction in commute time turns out to be wrong (which is unlikely to be wrong) and he loses all the money he invested in buying YES shares, the profit he'll make from the project will easily cover the loss.
- Due to this evil actor, the prediction of the market was manipulated.

#### Example 2
- Now, let's check out another small shareholder of a construction group. His company was confident that the road infrastructure expansion initiative will be taken by the authorities, so they made all the preparations internally. If the road infrastructure expansion project dosen't get selected, the company will face a huge blow.
- The prediction market dosen't look in his favour, it's very skewed. This scares him of a possible financial loss.
- He decides to prepare for both cases, road infrastructure expansion & metro rail expansion. He bought some YES shares of road infrastructure expansion. Instead of buying just the YES shares of road infrastructure expansion, he buys the YES shares of metro rail expansion too. Is it confusing?
- Actually, he is trying to control the financial damage caused, in the case of metro rail expansion getting selected. He is trying to secure some money from the YES shares (highly probable of happening) of the rail network expansion market for compensating some amount of the loss (small shareholder) that he'll possibly face by his company not getting the project. This is called financial hedging.
- Financial hedging is an investment strategy aimed at minimizing potential losses by seeking some favorable returns.
- Likewise, this shareholder also contributed in manipulating the market for minimizing his overall loss.

Prediction market designers are working on finding a solution to prevent such manipulations from taking place.

## Summary
- Historical Roots of Betting:
- Humans have always had a natural inclination to predict the future and bet on outcomes.
- This urge has evolved from ancient bartering to modern sports and various events, driven by potential profit.
- Introduction to Prediction Markets:
- They forecast binary events (e.g., Yes/No outcomes).
- Participants buy/sell contracts, with prices indicating the collective probability of an event.
- Accuracy through "Wisdom of the Crowd":
- Prediction markets aggregate diverse information and incentivized forecasts.
- They often surpass traditional polling and expert predictions in accuracy.
- Variety of Platforms:
- Examples include academic (Iowa Electronic Markets), regulated (Kalshi), and blockchain-based (Polymarket).
- Opinion trading apps (Dream11, Probo) are simpler, often entertainment-focused versions.
- Skill vs. Chance:
- Unlike gambling, prediction markets emphasize skill, analysis, and informed judgment.
- They are frequently regulated as financial instruments due to this nature.
- Superior to Mainstream Media:
- Monetary stakes lead to less bias and faster updates than traditional news forecasts.
- Demonstrated higher accuracy, for instance, in the 2008 US election vs. poll-based models.
- Opinion Trading:
- A less regulated form of prediction market.
- Often features simpler structures and a framing as entertainment rather than formal finance.
- Decentralized Prediction Markets (DPMs):
- Blockchain-based for trustworthiness and censorship resistance.
- Operate via smart contracts, no intermediaries.
- Polymarket:
- Leading DPM, uses USDC on Polygon.
- Team creates markets for quality; trading is permissionless.
- Share prices reflect probabilities (order book).
- UMA Optimistic Oracle resolves markets with dispute process.
- PnP (Predict and Pump) Vision:
- Aims for Uniswap-like permissionless prediction markets.
- Allows anyone to create markets, provide liquidity, and use decentralized settlement.
- Transforms forecasting by speeding up market creation and removing bias.
- Conditional Prediction Markets:
- Predict outcomes under specific scenarios (e.g., policy impacts).
- Aid in decision-making, risk reduction, and crowd-sourced governance.
- Market Dynamics:
- Generally a zero-sum game (or slightly negative due to fees).
- Users include retail, experts, liquidity providers, and arbitrageurs.
- Vulnerable to biases (e.g., confirmation, favorite-longshot).
- Susceptible to manipulation (e.g., large players skewing odds) and hedging.
- Designers are actively working to mitigate these issues.