"I almost fell outta my chair as I was reading this . . . because it is gospel. By the time enough data is generated for a good result, the gun will have been worn to the point that it actually shoots differently - go figure."

Not really. Only 50 shots for any given load/gun combination will provide highly reliable grouping performance information that cannot be obtained in any other way. Most so-called Experts have no concept of what statistical analysis involves. But they love to give their opinions anyway.