When I travel I am faced with a difficult choice: which of the toilet booths to use in offices or hotels. I always try to guess which one is the least used.
Harford responds by appealing to the efficient markets hypothesis that suggests that there will be an equilibrium whereby all toilets will on average be equally clean. Of course, this equally suggests that there is nothing special about traveling: toilets will be on average equally clean at your own office too. Note that with a sufficient amount of use this will be true even if there are toilet choosing biases of the form Harford speaks about. After all, it takes a number of rational choosers to balance this out.
What he hints at but fails to see the gravity of is the insufficient incentives for people to investigate toilet cleanliness and the potential externality poses by dirty toilets being even occasionally overused. Anyhow, as ever, the issues here have not been fully worked out — certainly by hard nosed academic researchers. (Yes I intended the pun!) May be a good opportunity for someone to signal their lack of interest in Harvard.