W ostatnich latach popularny jest koncept obstawiania pieniędzy na wystąpienie jakiegoś zdarzenia.
https://www.theatlantic.com/technology/2026/01/america-polymarket-disaster/685662/ (bez premki: https://archive.is/gFUry)
Sam w sobie ten koncept nie jest niczym nowym, bo oryginalnie było to np. obstawianiem meczy czy innych głupotek, jednakże w ostatnich latach zakres tego co można obstawiać znacząco się zwiększył.
Można obstawiać wynik wyborów, liczbę tweetów Elona Muska w danym tygodniu czy np. czy przed 20 stycznia USA zbombarduje Iran.
Przykładowo Polymarket
Pojawiłą się ciekawa analiza przepływu $$ na takich zakładach i zastanawia mnie na ile to ma sens / jest rzetelne
https://www.jbecker.dev/research/prediction-market-microstructure
A central question in zero-sum market analysis is whether profitable participants win through superior information (forecasting) or superior structure (market making). Our data strongly supports the latter. When decomposing maker returns by position direction, the performance gap is negligible: makers buying "YES" earn an excess return of +0.77%, while those buying "NO" earn +1.25% (Cohen’s d ≈ 0.02). This statistical symmetry indicates that makers do not possess a significant ability to pick winners. Instead, they profit via a structural arbitrage: providing liquidity to a taker population that exhibits a costly preference for affirmative, longshot outcomes.
This extraction mechanism relies on the "Optimism Tax." Takers disproportionately purchase "YES" contracts at longshot prices, accounting for nearly half of all volume in that range, despite "YES" longshots underperforming "NO" longshots by up to 64 percentage points. Makers, therefore, do not need to predict the future; they simply need to act as the counterparty to optimism. This aligns with findings by Reichenbach and Walther (2025) on Polymarket and Whelan (2025) on Betfair, suggesting that in prediction markets, makers accommodate biased flow rather than out-forecast it.