Real-time market sentiment — Stocks & Crypto
Every day at market close, we scan multiple internet sources — social media, video platforms, search trends, and financial news — to capture what people are saying and searching about the markets.
Posts are analyzed using NLP models trained on financial language, combined with a proprietary panic/greed keyword system. High-engagement content is weighted more heavily to focus on what's actually moving sentiment.
Source signals are weighted and combined through a proprietary scoring algorithm into a 0-100 scale. 0 = Extreme Greed, 100 = Extreme Panic. The model is continuously refined as more data is collected.
We record daily SPY and BTC-USD prices alongside each reading to track whether sentiment extremes predict market reversals — the core thesis behind the index.