bayesianism
Explanation
The Joke
A bald man with glasses explains Bayesian probability to an audience. He notes that according to Bayesianism, every thesis, no matter how ridiculous, has some probability of being true -- but the sum of all theses multiplied by their probability must still be one. He has therefore created "The Bayesian Overloader": a device that generates opposing theories at tremendous speed (hundreds or trillions per second). By flooding the probability space with alternative theories like "everyone living will not die" or "only pumpkins die" or "nobody has ever died, they're all just sleeping," the device forces the probability of any undesirable theory (like "I will die someday") to shrink, since each new theory gets "some slice of the probability." After about thirty seconds, the initial theory becomes "vanishingly unlikely." He concludes triumphantly: "And thus, I am immortal."
In the final panel, a woman in the audience says: "See, that's why I'm a frequentist."
The Humor
The comic satirizes a common misunderstanding (or deliberate misapplication) of Bayesian probability. In actual Bayesian reasoning, you cannot simply dilute the probability of a well-supported theory by inventing absurd alternatives -- the prior probabilities assigned to nonsensical theories would be essentially zero, so they would not meaningfully reduce the probability of established theories. The man is treating probability as if it were a finite pie that can be sliced thinner by adding more plates, when in reality the new "slices" for absurd theories would be infinitesimally small.
The punchline -- "that's why I'm a frequentist" -- plays on the real philosophical divide in statistics between Bayesians and frequentists. The woman's response suggests that this kind of absurd reasoning is exactly what makes her reject Bayesianism, even though the flaw is in the man's reasoning, not in Bayesian theory itself.
References
- Bayesian probability: A framework in which probability represents a degree of belief, updated as new evidence arrives via Bayes' theorem.
- Frequentist probability: An alternative framework that defines probability strictly as the long-run frequency of events in repeated experiments, avoiding the notion of subjective "degrees of belief."
- The Bayesian vs. frequentist debate is a genuine and sometimes heated philosophical disagreement in statistics.