Calling it silly seems a bit harsh, it's just statistics. We are interested in knowing the frequency of earth-like planets existing. If we assume that the existence of earth is a pre-condition for life existing and performing the measurement, then the existence of earth tells us nothing about that frequency. Regardless of the frequency in question the count of earth-like planets starts at 1, because in scenarios where the count is at 0 the measurement is not being performed. There could be millions of earth-like planets in the milky way, or there could be only one, or earth could even be the only life-supporting planet in the universe. There is no mechanism given only the existence of earth to reason statistically about the distribution of earth-like planets.

Whereas finding a second one tells us a lot. So that's why I say we are in more like the N=0 case, and that finding a second earth-like planet puts us in the N=1 case.