Meritocracy requires established experts to select candidates based on merit, which is quite different from the masses selecting candidates based on popularity (democracy).
I disagree with the experts part, after all under meritocracy Einstein is the greatest scientific mind of the 20th century however he wasn't even able to get a job in academia until after he produced ground breaking results.
I think that was the implication -- that experts are often wrong. One could go the No True Scotsman route and say that a true expert would occasionally select randomly following a Bayesian Bandit approach for optimal decision-making.