Estimate MMR changes after matches.
Almost every matchmaking rating is a variant of Elo. Your expected score against an opponent is a logistic function of the rating gap, scaled so a 400-point deficit implies roughly a one-in-eleven chance of winning. The rating change is the K-factor multiplied by the difference between what actually happened and that expectation, which is why beating a stronger opponent pays more and losing to a weaker one costs more.
Expected score
Expected = 1 / (1 + 10 ^ ((opponent MMR - your MMR) / 400))
Rating update
MMR change = K x (actual score - expected score), where actual is 1 for a win, 0.5 for a draw, 0 for a loss
Chess federations use 10 to 40 depending on rating and experience, and multiplayer games commonly sit near 20 to 40. New accounts often get a temporarily inflated K so placement converges quickly.
Because the expected score was already near 1. Beating an opponent far below your rating confirms the prediction, so there is little information for the system to act on.