Thinking, Fast and Slow · chapter 22 · id thinking-fast-and-slow-c22-03-expert-political-and-economic-

“Expert political and economic forecasters perform barely better than chance, and pundits with the most media visibility tend to be the least accurate (Tetlock, 2005).”

holdsconfidence: high✓ ZHP-verified + red-teamed

Receipts

Tetlock (2005), Expert Political Judgment, Princeton University PressDOI registry: valid

Over a 20-year study tracking ~28,000 predictions by 284 experts, experts performed barely better than simple extrapolation algorithms. TV pundits performed worse, possibly because fame selects for overconfidence.

Tetlock & Gardner (2015), Superforecasting; GJP results in ScienceFLAGGED: DOI not in registry

The IARPA-funded Good Judgment Project found that a small fraction of forecasters ('superforecasters') consistently outperformed intelligence analysts with classified access. They shared traits: open-mindedness, comfort with numbers, incremental updating.

Kahneman & Klein (2009), American Psychologist 64(6)DOI registry: valid

Political forecasting fails the two conditions for valid intuition: the environment is low-validity (unstable, complex) and feedback is delayed and ambiguous.

This claim is a stable, citable object. If you can falsify a verdict, tell us — corrections are loud here.