Yes, I think we may have to agree to disagree.
I generally think that the designer has rated leaders and units based upon their performance 'on the day' rather than on 'past performance'. I think that is the correct way to do it. Past performance can never guarantee what that performance will be tomorrow. Whether it be a football team, a horse in a race, or a man in battle you just don't know. [I do alright on the horses but if I knew how they would perform 'on the day' (instead of their past performance) I would be a very happy man.] Fortunately, thanks to history we know how specific units and leaders performed on the day of the battle. The designers would be remiss not to use that information and to instead use the less reliable information of past performance.
We know the performance of Iverson and his brigade at Gettysburg. We know their performance prior, and subsequent to that battle. We don't know for sure if Iverson was drunk or why his men failed to send skirmishers forward and thus avoid the ambush. However, we do know they performed badly and that is a fact. For whatever reason they did not have the cohesion that one may have suspected based on overall past performance. [I did come across some allegations of Iverson's brigade 'shirking their duty' at Chancellorsville but that does not seem to be the opinion of other commanders at the time (
https://leefamilyarchive.org/9-family-papers/336-battle-report-of-the-chancellorsville-campaign-1863-september-23)].
As for Shiloh, well, past performance is again not used for the ratings. Why? Because we have the more reliable information of how they actually performed 'on the day'. Was it due to factors such as ill health, being surprised, or some sort of groupthink by the Union (Peabody and 25th Missouri excepted) that there were
not large numbers of Confederates nearby? We don't know for sure but we do know they performed below previous past performances. How they performed on the day is what should be reflected,
not past performance, because that is what will best simulate the battle in the scenario.
If the Union ratings are adjusted to reflect past performance then in most games of that scenario the CSA will have no hope of pushing forward on Day 1. Indeed it may well be that the CSA is pushed back and possibly destroyed on Day 1. That is not Shiloh. Other changes, such as moving the Objective hexes to CSA control could have been made so that the Union is the attacker. But, again, that is not Shiloh.
The designers did the only thing they could do to somehow simulate the conditions
at the time and show the flow of the battle. The only thing in the game engine that would enable them to do that was to use unit/leader ratings. Otherwise, the scenario would be flawed and in no way reflect the battle itself as it occurred. The scenario, as designed, seems to work well with the Union generally winning a close-run thing. That tells me the designer probably got things pretty right.