The author demonstrates generating random numbers from a normal distribution in R, showing the numbers and calculating their likelihood. Despite being from the correct distribution, the likelihood is extremely small, making it unlikely that these numbers came from that specific distribution. Even when compared to an incorrect distribution, the likelihood remains tiny. The author questions why statistics heavily relies on likelihoods, given that they are consistently small, using the analogy of a turtle being faster than a snail, but both are slow. This challenges the common use of likelihood in statistics, raising a thought-provoking point about the significance of such small probabilities.
https://stats.stackexchange.com/questions/639548/why-is-everything-based-on-likelihoods-even-though-likelihoods-are-so-small