it might seem logical, but there is no connection between the job and the backup after the job is finished a job doesn't own backups, nor is it constant/immutable. all the pruning and limiting logic works on the storage/backup "group" level, and covers all backup of that guest, whether they are created by job A, B or C or manually.
In that case I hope Google will index this thread well enough, but even then people will find it only AFTER they discover the problem.
It's a bit like, if I design a car with the steering opposite to what's intuitive, I would argue even documenting it well enough is not ... enough.