Chapter 34: Key Takeaways

Blockchain Fundamentals for Prediction Markets

Blockchain-based prediction markets offer censorship resistance, global access, trustless settlement, transparency, and composability --- advantages that centralized platforms cannot fully replicate.
These benefits come with trade-offs: higher transaction costs (gas), slower execution (block times), a steeper learning curve, and smart contract risk.
The choice between centralized and decentralized prediction markets depends on your priorities: if trust minimization and global access matter most, blockchain is compelling; if speed and low cost are paramount, centralized platforms may be preferable.

A blockchain is a distributed, append-only data structure maintained by a network of nodes following a consensus protocol.
Blocks contain transactions and are linked by cryptographic hashes, making the chain tamper-evident: altering any historical block invalidates all subsequent blocks.
Merkle trees organize transactions within blocks, enabling efficient verification with O(log n) proof sizes.
Proof of Work secures the chain through computational effort; Proof of Stake secures it through economic collateral (staked tokens). Ethereum uses PoS since "The Merge" in September 2022.
Finality determines when a transaction is irreversible. Ethereum's PoS provides epoch-based finality (~12.8 minutes), which is critical for prediction market settlement.

Ethereum extends blockchain with a Turing-complete programming environment (the Ethereum Virtual Machine, or EVM), enabling arbitrary smart contract logic.
Ethereum has two account types: Externally Owned Accounts (EOAs) controlled by private keys, and Contract Accounts controlled by their code.
Every transaction specifies a gas limit and fee parameters (base fee + priority fee under EIP-1559). Gas is the unit of computational work; the base fee is burned, and the priority fee goes to validators.
Typical prediction market operations cost between 65,000 gas (token transfer) and 2,000,000 gas (market creation). At 20 gwei base fee and $3,000 ETH, a 200,000-gas trade costs approximately $12 on L1.

Smart contracts are immutable programs deployed on the blockchain that hold state, enforce rules, and interact with other contracts.
Solidity is the dominant smart contract language. Key constructs include state variables, functions (external/public/internal/private), events, modifiers, and constructors.
The ABI (Application Binary Interface) defines how to encode function calls and decode return values. Python's web3.py handles ABI encoding automatically.
Contract interactions follow the pattern: build transaction, sign with private key, broadcast, wait for confirmation.

Providers connect web3.py to blockchain nodes: HTTPProvider (most common), WebsocketProvider (real-time), and IPCProvider (local nodes).
Reading data (balances, contract state, events) is free --- it runs locally on the node without consuming gas.
Writing data (sending transactions, calling state-changing functions) costs gas and requires a private key for signing.
Events/logs are the primary mechanism for tracking on-chain activity. They are cheaper to emit than storage writes and are indexed for efficient querying.
Pagination is essential when querying historical events, as RPC providers limit the block range per query.

ERC-20 defines fungible tokens (every unit is identical). Most prediction market collateral and simple outcome tokens use ERC-20.
ERC-721 defines non-fungible tokens (each unit is unique). Less common in prediction markets but used for unique positions or governance rights.
ERC-1155 is a multi-token standard that manages both fungible and non-fungible tokens in a single contract. Polymarket's Conditional Token Framework uses ERC-1155.
The split/merge mechanism is fundamental: 1 unit of collateral splits into 1 YES + 1 NO token; merging reverses the operation. This ensures YES price + NO price approximates 1.

Layer 2 (L2) networks process transactions off Ethereum mainnet while inheriting its security. They reduce costs by 10-1000x and increase throughput.
Optimistic Rollups (Arbitrum, Optimism, Base) assume validity and allow fraud proof challenges during a 7-day window.
ZK-Rollups (zkSync, StarkNet) generate cryptographic validity proofs, enabling faster withdrawals but requiring more computation.
Sidechains (Polygon PoS) have their own consensus but lower security guarantees than rollups.
Polymarket operates on Polygon because the low fees ($0.001 per trade) and fast confirmations (2 seconds) make small bets economically viable.
Connecting to L2 from Python uses the same web3.py API --- only the RPC endpoint URL changes.

Ethereum uses elliptic curve cryptography (secp256k1). Address = last 20 bytes of keccak256(public key).
Never hardcode private keys in source code. Use environment variables or dedicated secrets management.
Minimize hot wallet funds and use separate accounts for trading, testing, and cold storage.
Monitor token approvals and revoke unused ones. Unlimited approvals are a security risk.

Direct contract queries (view functions) provide current state data for free.
Event logs provide historical activity data (trades, market creation, settlement).
The Graph indexes blockchain data and exposes it via GraphQL, enabling efficient historical queries without block scanning.
Data pipelines for serious analysis combine contract queries, event logs, and indexed data sources.

Transactions move through: construction, signing, broadcasting (mempool), inclusion (validator selection), execution (EVM), and confirmation (finality).
Gas price strategies range from "slow" (25th percentile, save money) to "urgent" (high priority fee, time-sensitive).
Transaction monitoring includes checking confirmation, handling reverts, and implementing retry logic with escalating gas prices.
Simulating transactions via eth_call before sending catches reverts early and saves gas on failed transactions.

Reentrancy: External calls before state updates allow attackers to drain funds. The Checks-Effects-Interactions pattern prevents this.
Front-running / MEV: Visible mempool transactions enable sandwich attacks. Mitigations include private mempools (Flashbots), commit-reveal schemes, and slippage limits.
Oracle manipulation: The most critical risk for prediction markets. Decentralized oracle networks, optimistic oracles, and dispute mechanisms mitigate this risk.
Approval risks: ERC-20 approvals grant contracts permission to spend your tokens. Use exact amounts, not unlimited approvals.
Always verify contract addresses, read audit reports, start small, and use hardware wallets for significant positions.