📚 Methodology

How the Housing Viral Engine works

Core Principle

The Housing Viral Engine measures YouTube content behavior—specifically the language patterns, sentiment, and hype intensity used by creators when discussing housing topics. This is NOT a prediction tool, forecasting system, or financial advice platform.

⚠️ Important: We measure what creators are saying, not what will happen in the housing market. All "predictions" refer to future-assertive language frequency, not actual outcomes.

Data Sources

  • YouTube RSS Feeds: Free public RSS feeds (no API key required)
  • 30 Canadian Real Estate Channels: Curated list of housing-related channels
  • Time Window: Last 24 hours of published content
  • Data Extracted: Title, description, publish date, channel info

Processing Pipeline

PASS 1: Metadata Analytics

Analyzes video titles and descriptions to classify:

  • Sentiment: Negative, Neutral, or Positive
  • Hype Intensity: High Hype or Low Hype
  • Content Style: Predictive Language or Non-Predictive
  • Narrative Topic: The Crash, The Hustle, The Struggle, or The Flex

PASS 2: Ranking

Calculates normalized metrics and ranks videos/channels:

  • Topic Relevance: Classification confidence from PASS 1
  • Recency: Newer videos rank higher
  • Ranking Score: 60% topic relevance + 40% recency

Panic Score Algorithm

The Doomsday Clock Panic Score (0-100) is calculated deterministically:

  • Base Score: 50 (neutral starting point)
  • +1 per video: Negative sentiment + High Hype
  • -1 per video: Neutral or Positive sentiment
  • Clamped: 0-100 range
  • Panic Zones:
    • 0-32: Vibes Are Chill
    • 33-66: The Noise
    • 67-100: TOTAL MELTDOWN