How do two (programming) methods for capping daily rep stack up?

Question

I wrote a supersuccessful answer to a question on an SE site that got (at least) 21 upvotes in a 24-hour period. In theory, I should have gotten 210 rep points, but there is a daily reputation limit of 200 points (20 upvotes) on a single site, so my daily score was capped at 200. This much I understand.

Then I received one downvote on the question. I would have expected that the -2 would be applied against my 210 score for a net score of 208, which would still be above 200. In other words, I would have expected an "override" programming command of something like, "If daily rep > 200, then daily rep = 200." Put another way, I'd be using a "truncation" function under which my raw score (of 208) is retained by the system.

Instead, the downvote reduced my score from 200 to 198. I looked it up on meta, and they said that the program "stopped filling the bucket" at 200, then the downvote depleted the "bucket." Put another way, the actual program appeared to be using an "accumulating" function that stopped at 200, meaning that my extra upvotes (data), were lost.

Apart from what I've said, what are the merits of the two different approaches to this problem, and how do people choose one over the other?

Answer 1

In general, people choose the simplest version to implement - especially stuff like the repcap which is a sort of experiment where nobody knows how it will impact reality.

Simple implementations are quicker, less prone to error, more robust when other things change, and tend to have a smaller impact on the overall design. And the difference of a few imaginary points in these corner cases isn't worth the added complexity.