Engagement Metric Regression
Fading fame: predicting the fall after the spike.
Overview
This pillar analyzes unsustainable spikes in celebrity popularity caused by single viral events. It forecasts the inevitable regression to their baseline engagement, providing an edge in markets that overvalue short-term hype.
What It Does
It establishes a historical baseline for a celebrity's social media engagement and search interest. When a viral event occurs, it measures the peak of these metrics and models the decay rate. The model predicts how quickly and to what level their popularity will normalize.
Why It Matters
The public and prediction markets often overreact to fleeting moments of fame. This pillar offers a quantitative forecast for the natural decline in attention, allowing users to identify overhyped assets and position against the long-term impact of short-term news cycles.
How It Works
First, the pillar calculates a 90-day rolling average of a celebrity's follower growth and engagement rates. Second, it detects a spike event, defined as a significant deviation from this baseline. Finally, it applies a logarithmic decay model to forecast the timeline for these metrics to return to their pre-event average.
Methodology
A 90-day pre-event baseline is established for key metrics like follower growth and engagement rate. A spike is identified by a >3 standard deviation increase in search volume or follower growth. The regression is modeled using a half-life decay formula, V(t) = V_peak * (0.5)^(t/T_half), where the half-life (T_half) is estimated from historical data of similar celebrity events.
Edge & Advantage
It provides a data-driven counter-narrative to hype, allowing you to systematically fade public overreactions to viral entertainment news.
Key Indicators
-
Engagement Rate Decay
highThe decline in likes, comments, and shares per post after a peak popularity event.
-
Follower Churn Rate
highThe rate at which new followers gained during a spike begin to unfollow as interest wanes.
-
Search Volume Normalization
mediumThe speed at which Google and social search interest returns to the pre-event baseline.
Data Sources
-
Provides historical follower and engagement data for major social media platforms.
-
Tracks public search interest for celebrities and related topics over time.
-
Monitors content performance across social media to track virality and engagement.
Example Questions This Pillar Answers
- → Will [Celebrity X]'s Instagram follower count remain above [Number] by [Date]?
- → Will the TV show starring [Recently Viral Actor] be renewed for a second season?
- → Will [Influencer Y] win 'Creator of the Year' at the upcoming awards?
Tags
Use Engagement Metric Regression on a real market
Run this analytical framework on any Polymarket or Kalshi event contract.
Try PillarLab