We explore an approach to developing Datalog parallelization strategie
s that aims at good expected rather than worst-case performance. To il
lustrate, we consider a very simple parallelization strategy that appl
ies to all Datalog programs. We prove that this has very good expected
performance under equal distribution of inputs. This is done using an
extension of 0-1 laws adapted to this context. The analysis is confir
med by experimental results on randomly generated data.