A prediction can look precise without capturing everything that could change the outcome. From Election Day polling to sports forecasting, the most important uncertainty is often the hardest to measure.
Eric Bradlow, Adi Wyner, and Shane Jensen explore how polls, prediction markets, and statistical models handle uncertainty and correlated error. They discuss what early results can reveal about an election forecast and why no model should be mistaken for certainty. The conversation also covers the Yankees’ postseason roster questions, NFL parity, the college football playoff race, Conor McDavid’s scoring pace, tennis rankings, and the Sixers’ early-season performance.
","author_name":"The Wharton School"}