I found that piece to be not well thought out when it first appeared. For starters, it makes the argument that because Thursday is the best day to post, and noon the best time, therefore Thursday at noon is the best time to post. Which does not follow. Plus it talks about time zones a bit, but doesn't seem to clue into the fact that using a different timezone for your data would actually change your per-day results.
And of course, he didn't do the real analysis: is there any variation in the chance for a post to be noticed depending on time of day/day of the week. He just says more toprated posts were made on Thursday at noon: I wouldn't be surprised if that was simply due to more posts being made on Thursday at noon. How does the percentage of posts that make it to top rating vary by day/time? Midnight on Saturday may actually be you best time for scoring a top rating, if you really care about such.