OP seems to be ignoring sensibilities of maintainers and how seamless getting a PR in is per se, without looking at the changes at all. It may not be the norm,

OP seems to be ignoring sensibilities of maintainers and how seamless getting a PR in is per se, without looking at the changes at all. It may not be the norm, but I've seen occurrences of 3 smaller PRs taking longer to be reviewed and merged (for bogus reasons, not because they touch hot code) than one big chunk (best example would be a maintainer who is arguing about literally everything that is not their own code). Sometimes the process is the actual problem. Maybe it's even worse in companies than in open source projects... On the rest, I am not disagreeing at all, it just might be worth looking at the reasons, but I hope statistics will sort out what I described above.