of 46

# Nate Silver - What Do Poker, Presidential Elections, & Sports Have in Common with Big Data? - Data Summit

Nate Silver, Statistician, Author & Founder, ESPN'S FiveThirtyEight blog His presentation at the 4A's Data Summit on Oct. 16 in NYC. Visit http://datasummit.aaaa.org/ for more information.
Published on: Mar 3, 2016
Published in: Sports      Technology      Spiritual

#### Transcripts - Nate Silver - What Do Poker, Presidential Elections, & Sports Have in Common with Big Data? - Data Summit

• 1. Big Data… Big Problems?
• 2. Big Data… Big Problems?
• 3. Big Data… Big Problems?
• 4. The Lessons of 2012 The 538 Method (Simplified) 1. Average the Polls 2. Count to 270 3. Account for Margin of Error
• 5. The Lessons of 2012
• 6. The Lessons of 2012
• 7. The Lessons of 2012
• 8. The Lessons of 2012 “nate silver” “joe biden”
• 9. The Lessons of 2012 “justin bieber”
• 10. Problem #1: Big Data… Big Bias?
• 11. Problem #1: Big Data… Big Bias?
• 12. Problem #1: Big Data… Big Bias?
• 13. The Signal-to-Noise Ratio
• 14. The Signal-to-Noise Ratio
• 15. The Signal-to-Noise Ratio
• 16. Problem #2: Desperately Seeking Signal http://imgs.xkcd.com/comics/sports.png
• 17. Problem #2: Desperately Seeking Signal
• 18. Problem #2: Desperately Seeking Signal
• 19. The Limits of Artificial “Intelligence” a 8 7 6 5 4 3 2 1 b c d e f g h
• 20. The Limits of Artificial “Intelligence” Kasparov +1 +1 +1 +1 +1 +1 +3 +3 +3 +5 +9 = 29 Deep Blue +1 +1 +1 +1 +1 +3 +3 +5 +5 +9 = 30
• 21. The Limits of Artificial “Intelligence” a 8 7 6 5 4 3 2 1 b c d e f g h
• 22. The Limits of Artificial “Intelligence” a 8 7 6 5 4 3 2 1 b c d e f g h
• 23. The Limits of Artificial “Intelligence” a 8 7 6 5 4 3 2 1 b c d e f g h
• 24. Problem #3: Feature or Bug?
• 25. Problem #3: Feature or Bug?
• 26. Problem #3: Feature or Bug?
• 27. Suggestions
• 28. Suggestions 1. Think Probabilistically 2. Know Where You’re Coming From 3. Try, and Err
• 29. Suggestion #1: Think Probabilistically Flood Prediction: 49’ Levee: 51’
• 30. Suggestion #1: Think Probabilistically Margin of Error: ±9’ Flood Prediction: 49’ Levee: 51’
• 31. Suggestion #1: Think Probabilistically
• 32. Suggestion #2: Know Where You’re Coming From
• 33. Suggestion #3: Try, and Err 100% 80% 60% Accuracy 40% 20% 0% 0% 20% 40% 60% Effort 80% 100%
• 34. Suggestion #3: Try, and Err
• 35. Suggestion #3: Try, and Err 100% 80% 60% Accuracy 40% 20% 0% 0% 20% 40% 60% Effort 80% 100%
• 36. Suggestion #3: Try, and Err
• 37. Suggestion #3: Try, and Err 100% 80% 60% Accuracy 40% 20% 0% 0% 20% 40% 60% Effort 80% 100%
• 38. Suggestion #3: Try, and Err 100% 80% 60% Accuracy 40% 20% 0% 0% 20% 40% 60% Effort 80% 100%
• 39. Suggestion #3: Try, and Err 100% Water level 80% 60% Accuracy 40% 20% 0% 0% 20% 40% 60% Effort 80% 100%
• 40. Suggestion #3: Try, and Err 100% Competitive Advantage 80% 60% Accuracy 40% 20% 0% 0% 20% 40% 60% Effort 80% 100%
• 41. Suggestions 1. Think Probabilistically 2. Know Where You’re Coming From 3. Try, and Err
• 42. Suggestions (Know Your Limitations) 2. Know Where You’re Coming From 3. Try, and Err
• 43. Suggestions (Know Your Limitations) (Consider Your Assumptions) 3. Try, and Err