Another smell: projects with small files will appear to have a low truck factor; whereas those that group code into very large files (edited by multiple people)

Another smell: projects with small files will appear to have a low truck factor; whereas those that group code into very large files (edited by multiple people) will tend to have a very high truck factor. Then theres the fact that irrespective of code quality, not all code bit-rots equally. Some code needs frequent updates to stay with times; some needs minor touches to track small api changes over the years, and some code is pretty much set in stone - it's virtually never touched except for languages changes. This algorithm isn't measuring the truck factor.