DDM, when used properly by disciplined statisticians, has shown its capability of exploring the trial data flexibly and, in the meantime, protecting the trial's scientific integrity.