Toward Superforecasting®: Lessons from the Eli Lilly Probability Assessment Panel

Accurate forecasts can make or break a business. But as we learned in our September webinar with bestselling author Dr. Phil Tetlock, who wrote Superforecasting®: The Art and Science of Prediction, most predictions – even those from so-called experts – frequently miss the mark. How can we know if our probability assessments are as good as they can – and should – be?

Back in September, Phil identified four best practices for building a Superforecasting® team. And no one has done a better job of making predictions than the probability assessment panel at Eli Lilly. Now led by Charles Persinger and created by recently retired Jay Andersen, this team has kept meticulous records of its forecasts and compared them with the actual outcome over a period of 20 years. There are many similarities between the best practices laid out by Professor Tetlock and the Lilly team’s processes and practices, leading to results that are head-and-shoulders above those of most business forecasters.

In this webinar, Charles and Jay describe the approach they established at Lilly and what they have learned over two decades. Carl Spetzler builds on Tetlock’s research findings and Lilly’s approach to propose a best-practice design for forecasting panels.


Charles Persinger, LillyCharles Persinger is Senior Research Advisor for Portfolio Strategy and Decision Sciences at Eli Lilly & Company. He leads Lilly’s internal decision consulting practice.


Jay Andersen, Lilly

Jay Andersen is retired from Eli Lilly & Company after a career of 28 years as a senior research fellow. During his tenure at Lilly, he established the probability assessment panel.

Carl Spetzler, SDG

Carl Spetzler is an author, speaker, and consultant with more than 40 years of experience working with top management and boards to make value-creating strategic decisions in the face of uncertainty.


Superforecasting is a registered trademark of Good Judgment Inc.