The nastiness problem in computer science

I wonder if our tendency to consider and plan for worst-case scenarios (time and space complexity analysis for algorithms, covering every edge case in implementations, etc.) is somehow related to this perceived negativity?