Understand the Future Through Prediction Markets
Predictech aggregates and analyzes data from leading prediction platforms to identify trends, sentiment shifts, and emerging signals.
Why Prediction Markets Matter
Market prices reflect probability-weighted beliefs backed by incentives; faster than polls and social media.
- Market prices express real probabilities
- Capital at risk reduces noise
- Markets react faster than traditional narratives
- Collective intelligence can outperform individual forecasts
What Predictech Does
Data Aggregation
Unify multiple platforms into one comparable structure
Market Metrics
Volume, liquidity, spread, open interest, orderbook depth
Trend Detection
Identify topics gaining attention early
One Platform. Multiple Sources.
- Cross-platform comparison
- Normalized metrics across sources
- Category + keyword exploration
- Historical snapshots for trend evolution
Advanced Analysis with AI (Premium)
Premium users can run LLM analysis on aggregated market data by selecting sources and time ranges.
- Select platforms/markets to analyze
- Generate sector-based sentiment summaries
- Spot correlations, anomalies, and emerging narratives
- Ask complex questions on aggregated data
Not predictions generated by AI — analysis generated from real market beliefs.
Who Is Predictech For
How It Works: Technical Architecture
From Prediction Markets to Database: The Data Journey
Data Sources
We use public endpoints, public APIs, or APIs with authentication to extract data from various prediction markets. We filter markets for relevant topics: Economy, Finance, Technology, Politics, Social issues. We select top markets based on various trading metrics such as volume, liquidity, spread, and orderbook depth.
Polymarket
GET https://data-api.polymarket.com/markets
Kalshi
GET https://api.elections.kalshi.com/trade-api/v2/markets
Predict.fun
GET https://api.predict.fun/v1/markets
Filtering and Selection
Data is filtered and selected before being saved to the database to ensure quality and relevance.
// Filter markets by relevant topics
const relevantTopics = [
'economy', 'finance', 'technology',
'politics', 'social'
];
const filteredMarkets = allMarkets
.filter(m => {
const category = m.category?.toLowerCase() || '';
return relevantTopics.some(topic =>
category.includes(topic)
);
})
.filter(m => m.volume > 0)
.filter(m => m.status === 'open');
// Select top markets by multiple trading metrics
const topMarkets = allMarkets
.filter(m => m.volume > 0 && m.liquidity > 0)
.map(m => ({
...m,
score: (m.volume * 0.4) +
(m.liquidity * 0.3) +
((1 - m.spread) * 0.3) // Lower spread = better
}))
.sort((a, b) => b.score - a.score)
.slice(0, 500);Data Aggregation
Data is normalized, enriched with categories and calculated metrics before saving.
// Calculate orderbook metrics
const orderbookMetrics = {
bestBid: orderbook.bids[0]?.[0] || null,
bestAsk: orderbook.asks[0]?.[0] || null,
spread: bestAsk - bestBid,
totalBidLiquidity: orderbook.bids.reduce((sum, bid) => sum + bid[1], 0),
totalAskLiquidity: orderbook.asks.reduce((sum, ask) => sum + ask[1], 0)
};Supabase Database
Data is saved as temporal snapshots in separate tables for each platform, enabling historical analysis.
// Save snapshot to Supabase
await supabase
.from('polymarket_snapshots')
.insert({
snapshot_timestamp: new Date().toISOString(),
market_id: market.conditionId,
market_data: enrichedMarket,
best_bid: orderbookMetrics.bestBid,
best_ask: orderbookMetrics.bestAsk,
volume_24h_usd: market.volume24hr,
open_interest: openInterest
});Dashboard Filtering
Users can filter by category, keyword, volume, liquidity and other metrics directly in the interface.
// Frontend filtering
const filteredMarkets = markets.filter(m => {
if (categoryFilter && m.category !== categoryFilter) return false;
if (keywordFilter && !m.title.includes(keywordFilter)) return false;
if (minVolume && m.volume24hUsd < minVolume) return false;
return true;
});Sync Frequency
Data is synchronized daily to keep information up to date.
Premium Sync
Premium sync enables real-time comparison of new data with previous snapshots, automatically identifying significant events such as anomalous liquidity injections, sudden volume changes, significant spread variations, or other patterns requiring attention. Users receive personalized insights based on their interests and the metrics they monitor.
From Market Signals to Understanding
Predictech transforms raw prediction market data into clarity. We don't tell you what will happen — we show you what the world believes is most likely.