People I (Mostly) Admire 117. Nate Silver Says We're Bad at Making Predictions
13 snips
Oct 28, 2023 
Nate Silver, the founder of FiveThirtyEight, is a renowned data scientist celebrated for his election predictions. In this conversation, he delves into the difficulties of making accurate forecasts, particularly in politics and economics. Silver discusses the art and science of election predictions and critiques common statistical models. He also explores the surprising correlation between COVID-19 death rates and political affiliations, illustrating how bias can cloud judgment. His journey emphasizes the importance of data literacy and adapting to the unpredictable nature of human behavior.
AI Snips
Chapters
Books
Transcript
Episode notes
Early Success and Pressure
- Nate Silver correctly predicted 49 out of 50 states in 2008 and all states in 2012.
- This success led to unrealistic expectations, setting him up for perceived failure later.
Correlated State Outcomes
- State election outcomes are correlated due to similar voter demographics, not independent events.
- Nate Silver's model accounts for this, unlike others which overestimate confidence.
Data Science Skills
- Nate Silver's success as a self-taught data scientist highlights the value of common sense and curiosity.
- Formal credentials are less important than good questions and storytelling with data.
