The misses are never counted, so the hit rate is never actually measured

The argument

This argument holds that the claim of prescience has never been tested, because the denominator has never been counted. To say that a programme is good at predicting the future is to make a claim about a rate, namely hits divided by attempts. Proponents of this position point out that only the numerator is ever assembled. Compilations of apparent predictions circulate widely, while no comparable effort exists to count the absurd claims that went nowhere, and without that figure the rate cannot be calculated even in principle. The asymmetry is not accidental, the argument continues, but built into how such material spreads. A gag that matches a later event becomes shareable and is preserved indefinitely, while a gag that matches nothing is forgotten within a week of broadcast. Over a run of many years this filtering operates continuously, so the surviving sample is selected precisely on the property under investigation. Any sample assembled that way will show a high success rate regardless of whether the underlying process had any predictive skill whatsoever. Proponents draw a further conclusion about the burden of proof. Because the evidence offered is a curated list rather than a measured rate, the reputation rests on a selection effect rather than on a finding. Anyone wishing to claim real foresight would need to specify in advance what counts as a prediction and what counts as a match, then apply those criteria to the whole body of material rather than to the examples that already stand out. Until that is done, the argument holds, the honest description is not that the show predicted the future but that nobody has checked, and that the checking is made difficult by the very enthusiasm that produced the lists.

Premises

Counter-arguments

Critics of this argument hold that it sets a standard of proof that would make the question unanswerable rather than answered. Predictions in ordinary life are rarely specified in advance in the manner the argument demands, and people nonetheless make reasonable judgements about who anticipates events well. They also object that treating every absurd gag as an attempted prediction inflates the denominator artificially, since most of the material was never offered as a claim about the future at all, so counting it as a failed prediction is as arbitrary as counting only the hits. A third objection is that the argument is compatible with real foresight rather than an alternative to it. Even a programme with genuine insight would have its successes collected and its filler forgotten, so the selection effect the argument identifies cannot by itself distinguish between skill and chance.

Rejecting the premises

[Rejecting P1] Critics hold that the standard of proof demanded would make the question unanswerable, since ordinary judgements about who anticipates events well are never specified in advance either. [Rejecting P2] Treating every absurd gag as an attempted prediction inflates the denominator arbitrarily, because most of the material was never offered as a claim about the future at all. [Rejecting P3] A programme with genuine insight would also have its successes collected and its filler forgotten, so the selection effect does not distinguish skill from chance. [Rejecting C] The argument establishes that the rate is unmeasured rather than that the apparent foresight is illusory.