Don't take unnecessary chances! (Q1868148)

Probability started as a theory for describing frequencies of different events; eventually, it was extended into a general theory for describing uncertainty. Several philosophers argue that in many real-life applications of probability theory, we cannot use the frequentist interpretation, because we are dealing with probabilities assigned to unique events. In the reviewed paper, the author argues that while some probabilities are indeed related to unique events -- such as probabilities in quantum gravity dealing with the Universe as a whole -- in most practical applications, what appears to be a unique event probability is, in reality, a frequency. For example, when an insurance company estimates, e.g., the probability of a car accident for a person, it relies on the records of the accident history for people with similar age, background, etc. -- so what appears to be a probability of a unique event is, in reality, a frequency calculated by considering similar events. In some cases, an insurance company may insure something truly unique, e.g., the hands of a famous pianist. The author argues that in many such cases, the corresponding probability is still frequency-related: namely, this probability comes from extrapolating frequencies (probabilities) corresponding to similar situations.