I'm not sure if there even is any universal measure for how good of a programmer you are. There's probably someone who's really good at optimizing the hell out of a really hot loop that absolutely has to be faster, never mind if the code is comprehensible to mere mortals. And somebody who can design and architect a huge, complex enterprise program well enough that a newbie developer can quickly get an idea of what's going on and make contributions. And somebody who can inherit a huge ball of mud and gradually refactor it into something more reasonable while keeping it working the whole time. And somebody who can start a new project and build just the right amount of infrastructure into it at the right time - not so much that you've got some huge over-designed thing for a product with 2 users, and not so little that it becomes impossible to scale when you hit 10k users. And plenty more sub-field besides those.
It's easy to be really great at 1 or 2 of those, and not so good at the others.