Ratings are complex. They look at the difference between your rating and your opponent's rating and also look at statistical measures of how accurate your rating is considered (based on how much you have played recently in that account in that time control and other things.)
You had only played 8 games in that other account in the rapid time control and no games in the last 2 1/2 months. As a result, your rating was still considered possibly inaccurate. As you saw, the change per game dropped off as you played more games..
Just curious, but after a game ends a certain amount of elo is given out to the winner and deducted from the loser, draws not affecting much. I was recently playing some games (on an anonymous account) to test out my sicilian dragon which I have been learning recently. I won some games. I have played on that account in june and haven't touched it since. I played 3 rapid games a few days ago on the account and won all 4. I beat a 1960, 2030, 2100 and got about 35 elo from each of those games. On the fourth game, the elo gain reduced down to 9. I am curious as to how this works since I still only have 12 rapid games on that account and I thought the elo gain stagnates after I play a certain amount of games with a win/loss ratio that is reasonable. Is it the change really this drastic or is something else going on? Is it because of the elo I was at? (2030). Account name is howdoyouplaychess12423