HiLine
So I've read a dozen online articles on how GMAT section scores are determined, but I could not find any information on how the section scores are combined to determine the 200-800 GMAT score. If the same score of each section indicates the same level of proficiency in the skill, why would the combination of two identical section scores not result in the same overall score, provided that the two exams are administered at about the same time to eliminate historical trends?
Identical scores do not necessarily imply identical skill levels. Even if we don't consider measurement error, the GMAT actually scores test takers on a
continuous scale. The total score that you see is from scores that are derived from this continuous scale. The score that test takers see for a particular section is separate from even those scores (the ones used to determine the total score). So the answer to your question is that the sectional scores exposed to test takers are just not as accurate. This is not much of a problem in verbal, because the percentiles there are very close to each other. But it is a problem in quant, where the difference between (let's say) 49 and 50 (or between 50 and 51) is actually quite large.
Just keep in mind that schools are not as worried about this as you might be. They tend to focus on the total score anyway, and if they need to take a look at the sectional scores, they go with the scaled scores.
Congratulations on the 770!