A quick perusal of this thread (and others linked) indicates random assignment, with a spread of 15-72 chests, rough mean range of 30-50 chests, can’t be arsed to calculate confidence interval or deviance, sorry.
Search terms: elemental chests, decrease, logging, log, tracking
@Olgavit:
Greetings.
As people have imperfect memories and perspectives, it’s always a good idea to collect your own data. It’s an easy thing to do: choose a unit of measurement - in this case, the community seems to think (frequency per total chests) rather than (frequency per total days) is more precise, and I agree. How long it takes you to complete a chest varies, and is almost guaranteed to be different from how long it takes me or any other random person.
So set up a simple tally of elemental chest / total chests. An index card works. Compare your data to the observations in that thread. Gather at least 50 chests’ of observations. See where you fall on the interval. If you are noticeably outside the range, send in a support ticket.
That will accomplish a lot more for you than being rude.
Kind regards.