War in the name of "credibility"

Worth noting with respect to the argument that we must occasionally slap around some two-bit dictator to maintain our credibility as a great power, (or bully):

As Christopher Fettweis shows, political scientists are nearly unanimous in finding little evidence for the proposition that the believability of threats depends on the outcome of prior threats—Thomas Schelling’s game theory notwithstanding. Daryl Press’ case studies show that when leaders, Hitler included, consider going to war in the face of deterrent threats, they focus on the balance of power and the threat-maker’s interests. It’s not that past credibility does not matter at all, but that it matters far less than other factors.