In a MEAN stack application, you need to perform a MongoDB aggregation that groups documents by a field, calculates an average, and then filters groups where the average exceeds a threshold. Which pipeline stage order is correct?

Remaining questions and full explanations are VIP-exclusive.