Replace mock data with live API integration for arbitrage detection across Kalshi and Polymarket

[Copilot]

This PR completely replaces hardcoded mock data in all four React components with live API integration for real-time arbitrage detection between Kalshi and Polymarket prediction markets.

Overview

The application now fetches live market data, performs fuzzy market matching, calculates real arbitrage opportunities, and provides a complete workflow for cross-platform prediction market arbitrage detection.

Key Changes

Core Infrastructure

• API Layer: New TypeScript modules for Kalshi (src/api/kalshi.ts) and Polymarket (src/api/polymarket.ts) with normalized market data structures
• Business Logic: Arbitrage calculation utilities implementing the formula APR = (1-C)/C * 365/D where C = minYes + minNo
• Fuzzy Matching: Market similarity scoring using title token overlap, category matching, and close date proximity
• State Management: Central store with localStorage persistence for matches and ecosystems

Live Data Integration

• Real-time Polling: Configurable intervals (Polymarket: 2.5s, Kalshi: 5s) with automatic pause on tab visibility change
• Market Normalization: CommonMarket interface standardizing data across platforms with pricing, conditions, and metadata
• Error Handling: Connection status monitoring and graceful fallback for API failures

Component Refactoring

Arbitrage Bets Component: Now displays live arbitrage opportunities calculated from matched markets in the store, showing real APR calculations, profit projections, and detailed condition pricing breakdowns.

Market Matcher Component: Replaced static market lists with live feeds from both exchanges. Implements fuzzy similarity scoring for match suggestions and provides a manual approval workflow with confidence metrics.

Ecosystem Matcher Component: Multi-platform market selection from live data with real-time ecosystem creation. Markets are grouped across exchanges with derived statistics (exchange count, market count, total conditions).

Matched Ecosystems Component: Displays stored ecosystems with live-calculated days until close, expandable details showing constituent markets, and integrated condition mapping interfaces.

Arbitrage Detection Logic

The system now detects arbitrage opportunities when:

1. Markets are matched between platforms with "same" condition mappings
2. Total cost C = minYes + minNo < 1.00
3. Calculates period return as (1-C)/C and annualizes using days to expiry
4. Updates opportunities in real-time as prices change

Data Persistence

• Market matches and ecosystems persist across browser sessions via localStorage
• Central store provides reactive updates to all components
• Condition mappings support 6 relationship types (same, subset, mutually-exclusive, etc.)

Example Usage

// Live market fetching
const kalshiMarkets = await fetchNormalizedKalshiMarkets();
const polymarketMarkets = await fetchNormalizedPolymarketMarkets();

// Arbitrage calculation
const opportunities = calculateArbitrageOpportunities(matches, kalshiMarkets, polymarketMarkets);
// Result: [{ apr: 0.15, profitOn100: 2.34, C: 0.977, ... }]

// Market matching with similarity
const similarity = calculateSimilarity(kalshiMarket, polymarketMarket);
// Result: { overall: 0.94, title: 0.89, date: 1.0, conditions: 0.87, settlement: 1.0 }

The application is now ready for production use with live prediction market data and real arbitrage detection capabilities.

---

💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click here to start the survey.