This dissertation should be understood as an effort to provide a kind of critique of statistical reason. "A kind", since it has another focus as well: the question of sociality. It is argued that these two topics are closely interrelated, not only, as the short exploration into the prehistory of statistics shows, for historical, but also and mainly for systematic reasons. The first part of this thesis is concerned with methods; not, in fact, so much with different methods as with the idea of methods in general. The second part deals with statistics, its prehistory and structure. It is argued that statistics is inherently causal and always and only understands the world as means (for our power), as it is. The third part explores a socio-logic, a concept meant to capture both the essence of sociality and our understanding of this sociality. And sociality, in the last analysis understood as play or game, turns out to be what cannot be controlled or dealt with in a methodical manner; it can never be reduced to what it is. The fourth and final part discusses and tries to overcome the proposed antithetical relation between statistics (methods) and sociality, and discusses possible consequences of the analysis for the fields of sociology and social thinking. In sum: The dissertation contributes to our understanding of methods, statistics and sociality and their interrelations.