From the GitHub blog:
24 Pull Requests is an annual community-organized event encouraging developers to give back to open source projects over the holiday season. Be part of this year’s giving by submitting 24 pull requests between December 1 and December 24.
Using pull requests as a metric is a bit odd, because not all pull requests are created equal. This is akin to saying, “Track 24 runs on Strava.” Using that analogy, a more interesting metric would be to track the total number of miles run on Strava, or total amount of time spent running. If someone’s completed 24 runs on Strava, I have no idea what their level of output is compared to someone else who’s completed 24 runs.
It’s harder to quantify, but I think time spent is a more helpful metric for software development. After all, that’s how employers track it. If I spend 8 hours making 4 pull requests and you spend 8 hours making 1 pull request, we’re even. I’m not 3 ahead. You just had a harder problem to solve, and that needs to be acknowledged somehow.
Of course, in this particular initiative, the goal isn’t to be comparing oneself to other contributors, but rather to be making a concerted effort at improving open source software. Counting pull requests is a convenient way to do that. I get why this was the chosen metric; I just think it has a limited use case.