FWIW, since J_R_B is not an average 4.0 (nearly bumped to 4.5 last year) and the 4.5s aren't necessarily top 4.5s and thus the ratings delta isn't a full 0.5, I thought it would be interesting to look at these three specific matches to see how far off the results were from expected.
Match 1 - J_R_B was effectively a decent 4.5 at this point and his partner was a good 4.0, and they played two average 4.5s. So they were supposed to lose a competitive match, but won convincingly.
Match 2 - J_R_B was higher now and his partner (same as before) went up too, so since their opponents (one the same as before) were weak 4.5s at this point, J_R_B and partner were supposed to win a fairly close 7-6,6-4 type of match but won more easily.
Match 3 - J_R_B was even higher now, and his partner was also above 4.0, so playing a good 4.0 and decent 4.5, J_R_B was supposed to win about 3 & 4 and won a little be more easily.
So J_R_B and partner certainly did better than expected in each match and some of that is likely due to doubles vs singles specialists and/or familiarity J_R_B had with his partner, but these 4.5 matches were expected to be a lot closer than the "average 4.0 vs average 4.5" type of scenario.
--
NTRP Ratings FAQ