Philip Tetlock - Footnote 73

NoteThe claim states that early analyses noted that foxes only modestly exceed simple benchmarks like extrapolation algorithms, but the source says that foxes sometimes exceed extrapolation algorithms, not that they only modestly exceed them. The claim raises questions about whether the framework sufficiently distinguishes skill from noise, but the source does not explicitly raise this question.