There are two main types of market sizing questions: the complex ones, which require many assumptions, bottom up. These will be precisely wrong, fine.
The other type are very high level, think a couple of top down assumptions, roughly right. This is what "guesstimate" would be, also called "back of the envelope" or "baby math". FWIW, I've seen a former boss, (ex-McK, CFO) successfully run a $5B company using only baby math. The idea here is to get to the essence of the problem, without getting lost in the noise. Very powerful stuff.