[jpeters1734]

There seems to be some confusion that different people’s tests are producing different results, but that’s not really the case. The results are actually remarkably consistent:

- Players with 1 current / high potential (e.g., 500–600) perform way better than they should in AAA.
- Players with 1 current / 1 potential get obliterated, as expected.
- Players with 1 current / moderate potential fall somewhere in between.
- The same 1/600 player performs terribly in MLB.

This confirms the key point: potential ratings are overly influencing statistical output in the minors, but not in the majors.

That’s the problem. Not whether the test used a real draft class. Not whether the scenario is “realistic.” The tests isolate the engine behavior and the behavior is broken.