Why do people use "the number of [...] issues and pull requests" to measure the usefulness of an open source project? Shouldn't a project gradually trend toward

Why do people use "the number of [...] issues and pull requests" to measure the usefulness of an open source project? Shouldn't a project gradually trend toward maturity, when the vast majority of bug fixes and big-win features are already part of it? Is that really the best point at which to start spinning it down?