This is a linkpost for https://dkl9.net/essays/entropic_2t1l.html
The strategy above makes all three statements seem equally unlikely to be true. Mathematically equivalent but with different emphasis would be to make all three statements seem equally unlikely to be false.
i.e. Pick things that seem so mundane and ordinary that surely they must be universally true—then watch the reaction as it is realised that one of them must actually be a lie.
That would be fun in the same way. If your goal in playing includes informing listeners, it's better to use thoroughly absurd facts and an equally-absurd lie; absurdity is low prior probability leads to surprise corresponds to learning.
The classic social game "two truths and a lie" asks each player, at their turn, to say three facts as the title says, which other players, listening, then seek to tease apart. It gets boring if listeners can easily and reliably tell which "fact" is false. To make it more exciting, the natural strategy would be to pick two absurd true facts and one dull fake fact.
With this strategy used at a moderate level, listeners mostly guess the more absurd true fact as the probable lie. At an extreme, listeners correctly recognise the lie as the most plausible-sounding one. We can do better.
Some people (me), by confusing people, derive fun; some witnesses (my friends), by watching, derive collateral fun. Disagreement begets excitement, which, except when violent, likewise adds to fun. Information-theoretic entropy roughly measures confusion and disagreement. The optimal "two truths and a lie" strategy is the most fun strategy is the one making the greatest entropy from the listeners' guesses is that which makes each "fact" seem equally probable as the lie.
This balance becomes numerically apparent if listeners declare their guesses. Even if the game focuses wholly on the speakers, the entropy will become cognitively apparent as listeners notice a similar truth-propensity between all three statements.