Neither metric really captures proficiency accurately. Not sure if that's possible.
Years of experience is difficult to measure. If I have a node.js build script that I update occasionally, do I have one year of node.js experience after working on the project for a year?
A pure front-end developer will know Javascript much better than a back-end dev who started using it at the same time.
I also spent 3 years working with MySQL before actually learning to use joins or grouping.
Doing several independent project instead of a large one only makes it more impressive, but doesn't indicate greater skill.
Sometimes I wonder if I should add small half-baked projects to my portfolio. I may only have worked on them for two days, but they would make me look super prolific. Do you think that's a good idea?