Yep, sorry, it quickly became apparent after rereading this that the original analysis was not worst-case. I have revised my comment to also include how I view calculating the actual time complexity since I didn't actually do that initially.