If we assume that happiness and unhappiness are accidents that our behavior has only partial influence, then yes, regression to the mean suggests that extreme happiness and unhappiness (suppose that maximum happiness is +3, and maximum unhappiness is 3) should be replaced by something averaged (something near zero). Accordingly, pronounced happy moments should sooner or later be replaced by less happy ones, and pronounced failures in the long term should be replaced by periods of greater happiness.
This reasoning is valid only for the case in which our life is not an accidental throwing from the side aside (random walk), but ordered in a certain way. Basically, I can assume that this is the case, because we adjust our actions in such a way as to minimize failure and maximize success. Some are better at it, some are worse, but we are trying.
"The man decided to swim in the river. He came to the shore and sees a sign:" Beware of the whirlpools "(on a distance of 1 km) ..." He has several options: Swim here; Walk 1 km, spending a certain amount of time and swim there; Go home without swimming. There will be different options for action and different consequences in this story. By neglecting the warning, you can swim and die in the whirlpool. Go home and be annoyed by an unfulfilled desire. Swim in a permitted place, but spend more time, but be satisfied. How does this story relate to the issue ?! People, indeed, very often reap the fruits of their choice! It is important how a person will act in this situation, because his further peace of mind or the continuation of the tests will depend on his actions.