There are so many posts out there on minimax and tic-tac-toe. So many posts and youtube videos and they all have their strengths and weaknesses. Mostly, they focus on the implementation of the algorithm, which is super important, but I found an excellent MIT lecture on the whys behind minimax and I want to share it.
I think I watched this video at least four times. Every time I wasn't quite sure how to get something to work (or why a thing was working) I knew where to find it in the video and I was like "ooooooh".
I probably still need to watch it a few more times, but understanding why we use minimax to build an unbeatable tic-tac-toe and why (and how) alpha-beta pruning works and see someone kind of run through the algorithm really helps me figure out how to write things well. I like to understand stuff. I also really like how there's no pseudocode (or actual code) in the lecture. I'd like it less, I'm sure, if there weren't dozens of other resources out there to help me get that part down too, but by decoupling the explanation from the implementation for a bit, I find I'm better able to focus on what I need.
So thanks, Prof. Winston!