Vier AfD-Kandidaten vor der Wahl gestorben: Zufall? (Wahrscheinlichkeit mittels Poisson-Verteilung)

About this video

• Video Title: Vier AfD-Kandidaten vor der Wahl gestorben: Zufall? (Wahrscheinlichkeit mittels Poisson-Verteilung)
• Channel: Prof. Dr. Christian Rieck
• Speakers: Prof. Dr. Christian Rieck
• Duration: 00:34:19

Overview

This video explores the statistical probability of unlikely events occurring in clusters, using the case of four AfD candidates dying before an election as a focal point. The speaker aims to debunk initial, often exaggerated, probability calculations and explain how to properly assess such events using statistical methods like the Poisson distribution, while also discussing the difference between statistical and game-theoretic thinking.

Key takeaways

• Statistical Probability vs. Perception: The video highlights how seemingly improbable events can occur due to the sheer number of possibilities and that initial calculations of such probabilities are often flawed.
• Defining the Event: A crucial step in statistical analysis is accurately defining the event being investigated, rather than calculating the probability of a very specific, narrow outcome.
• Game Theory and Information: The speaker introduces game theory concepts, suggesting that certain patterns or numbers can convey information or intent beyond random chance, using examples like license plates and financial figures.
• Poisson Distribution and its Applicability: The video discusses the use of the Poisson distribution as an approximation for binomial distributions when dealing with rare events, particularly in the context of death probabilities.
• Plausibility of the AfD Candidates' Deaths: Through estimations and calculations, the speaker suggests that while the deaths are statistically significant, they fall within a somewhat plausible range, especially when considering factors like age and health, though seven deaths would be highly improbable.