Merge Sort vs Quicksort Linked List

Merge sort is normally preferred for a linked list because every step matches what a list does well. I can find the middle with slow and fast pointers, recursively sort the two halves, and merge them by relinking nodes in increasing order. For example, 4 -> 1 -> 3 -> 2 becomes 1 -> 4 and 2 -> 3, and the merge only changes next pointers to produce 1 -> 2 -> 3 -> 4. That gives guaranteed O(n log n) time and can preserve the order of equal keys. I would be precise about space: the common recursive version uses O(log n) call-stack space, while a bottom-up version can use O(1) auxiliary space because it does not need an array buffer. Quicksort can be implemented on a linked list, so I would not say it requires random access. It is simply a less natural fit: partitioning and reconnecting sub-lists is awkward, it loses the cache-friendly indexed partition available on arrays, and a poor pivot can still lead to O(n²). So the preference comes from predictable time and easy pointer-based merging, not from quicksort being impossible on a list.