You have a very valid point regarding bugs. Transpiler bugs definitely happen. When I first tried using Babel - still named 6to5 then - instead of Traceur, it immediately broke my code because of an incorrect TCO implementation in Babel. Babel's maintainer (sebmck) is insanely responsive and the bug was fixed almost immediately, but still.
If I used Babel or any other transpiler for critical production code, I would only enable a whitelist of features that I thoroughly understand: for example, how a fat arrow is translated into a regular function. This vetting would be time-consuming, but that's a price I'd be willing to pay because I enjoy writing ES6 much more than ES5. That's highly subjective though, so I completely understand how some other people would rather just stick to ES5 and get stuff done with the lowest possible risk of weird things happening.
Babel is a completely different transpirer than all the others. It's goal is readable, understandable transpiled code and Sebastian (sebmck) is a genius robot wunderkind. I check the repo everyday and see new updates, bug fixes, etc. It is heartening to see someone give a project so much dedication.
I have an affinity for Babel unlike any other JS project (except React). They team works really hard to keep it interesting.
Yeah, Babel and the Babel team is great. I got very suspicious of the project after my first experience with it broke perfectly valid code, but after reporting the issue and then another and having them fix it almost immediately they won my confidence back. :)
For production code I'd still stick to a whitelist of features that have been released for at least a month though. Babel is aggressively pursuing ES6 coverage and new features might not have all edge cases covered when they're released. Give them one month though and with their responsiveness it's 90% likely it's going to be fixed.
If I used Babel or any other transpiler for critical production code, I would only enable a whitelist of features that I thoroughly understand: for example, how a fat arrow is translated into a regular function. This vetting would be time-consuming, but that's a price I'd be willing to pay because I enjoy writing ES6 much more than ES5. That's highly subjective though, so I completely understand how some other people would rather just stick to ES5 and get stuff done with the lowest possible risk of weird things happening.