The fact that strong static types makes the code self-documenting is the reason I vastly prefer it.

It makes me so much more productive, by orders of magnitude. And I say this as someone who has worked extensively in several different languages which covers large parts of this spectrum, like C, C++, Java, Python, JavaScript, TCL and others.

It makes it much easier for me to reason about code I have not recently worked on, be it in the current project or in dependencies. It also means I can be way more focused on the problem at hand, since I'm not constantly having to go on tangent goose chases to figure out what exactly I can do with this object that some function returns.

Sure there's the good, fuzzy feeling when the compile doesn't error out. But that's secondary.