All of the data was developed through testing, and not all at once. Some of the existing data is quite old. Manufacturers continue to add new data (when new components become available - not just powder but bullets too), and re-test some of the old data.
Modern pressure measurement instrumentation is much faster and more accurate than it was even a decade ago. Occasionally, when manufacturers re-test 'old' loads with new instrumentation, they learn something new. Sometimes they even have an "Oh shit' moment, and decide they need to reduce a supposedly 'time-proven' load because of a previously undetected pressure spike.
Also, as much as we'd like to think so, the testing is not completely exhaustive. Sometimes they find out after the fact that a particular powder is more temperature sensitive (or reverse temperature sensitive) than they originally thought.
Even if 99% of the old data is still good, you run a 1% risk of loading something nasty.