If you have two reflections that are linked (say by a dependent reflection refresh) and the second one is meant to use the first then the second one will only use it if they are not scheduled immediately one right after another. IE there is a race condition between the scheduler and whatever thing is making the reflections be used in query plans.