String theory is believed to have an enormous amount of vacua. Rather than to try to construct all of these explicitly, it seems more useful to take a statistical approach and study their numbers, distributions and properties indirectly. In particular one would like to know for example if there are many vacua with small cosmological constant, how precisely one would have to measure physical quantities to be able to single out one vacuum, and which properties are "natural" in the theory. In this talk I will present results on how such a quantitative statistical analysis can be performed, focusing in particular on type IIb N=1 flux compactifications. I will also discuss more general results on the distribution of meta-stable supersymmetry-breaking vacua with positive vacuum energy.