The competition-colonization trade-off has long been a mechanism explaining patterns of species coexistence and diversity in nonequilibrium systems. It forms one explanation of the intermediate disturbance hypothesis (IDH) for local communities--specifically that diversity should be maximized at intermediate disturbance frequencies, yet only a fraction of empirical studies support IDH predictions. Similarly, this trade-off is also a powerful explanation of coexistence at larger spatial scales. I show, with a microbial experimental system, that the diversity-disturbance relationship is dependent on the relative distribution of species along this trade-off. Here I show that, when species are skewed toward late-successional habits, local diversity declines with disturbance. Yet, despite this trait skew, diversity at scales larger than the patch appears insensitive to the trade-off distribution. Intermediate disturbance frequencies produce the greatest diversity in patch successional stage, thus benefiting the maximum number of species at larger scales.