Average case analysis defines expected time complexity by calculating the mean runtime over all possible input instances weighted by their probability, distinguishing it from worst-case guarantees which focus on maximum values to ensure conservative bounds. This concept applies specifically to randomized algorithm design and probabilistic analysis in computer science theory, utilizing recurrence relations and expectation calculations based on uniform distribution assumptions for distinct inputs. It demonstrates that while quicksort has a quadratic $O(n^2)$ upper bound specific to unbalanced partitions or sorted data, its average performance of $\Theta(n \log n)$ is derived from the statistical likelihood of balanced partition splits occurring across permutations of size $n$.
Average case analysis defines expected time complexity by calculating the mean runtime over all possible input instances weighted by their probability, distinguishing it from worst-case guarantees wh…