"In short, programmers may sometimes avoid the same features that make OCaml such a pleasant language to program with. It is possible to address this problem by using more aggressive optimization techniques (e.g., whole-program optimization as is used in the MLton Standard ML compiler (MLton, 2007)). Unfortunately, it does not seem likely that such optimization will be available any time soon for OCaml."
Then why OCaml over Standard ML using the MLton implementation?