Algorithmic prediction markets fail yet their real achievement is managing complexity.

Evidence

Zuboff (2019) shows how such computational technologies are drawn in Silicon Valley as the driving force of a new field of competition and commodification: the market for the best predictions, which are the most powerful speculations. But even computational algorithms for modeling and predicting the future often fail because of the inherent unpredictability of human behavior. Arguably, the single greatest success of computational technologies has been their harnessing of the multilayered complexity that has enveloped all human activities. — Aris Komporozos-Athanasiou, Speculative Communities, 645

Categorization

• XREF: Reinforces the book’s core thesis that prediction/anticipation — not accurate forecasting — is the operative logic of financialized life; also connects to Zuboff’s surveillance capitalism argument. — connects to: Predictability requires controlling variables, so heterogeneity defies guaranteed outcomes. — David Theo Goldberg, The Threat of Race: Reflections on…, 1895; Simple rules generate complexity without guaranteeing system self-regulation; Market actions based on efficient-market models can create anomalies those models cannot explain — Donald MacKenzie, An Engine, Not a Camera: How Financial…, 1166
• SEED: Angle: the ‘prediction market’ as misnomer — what these systems actually do is absorb complexity, not predict.

Broader Connections

• Simple rules generate complexity without guaranteeing system self-regulation #qualifies (0.78)
• Predictability requires controlling variables, so heterogeneity defies guaranteed outcomes. #complicates (0.75) — David Theo Goldberg, The Threat of Race, 1895
• Market actions based on efficient-market models can create anomalies those models cannot explain #qualifies (0.75) — Donald MacKenzie, An Engine, Not a Camera, 1166