I suppose an increase in life could cause it to be harder to notice a lowering of your int value by testing.
Though, using (Ld*10+100)Ld, I am getting results that show absolutely no difference until I pass the threshold set at (Ld*10+150)Ld. Almost immed. past that threshold, changes are obviously apparent. I would be inclined to believe it was possibly just the health increase, but it staying so consistent across different Int and leadership values suggests otherwise.
Also tested at extremely high int values so losing ~5% int becomes more noticeable. Tested at low int values also, and wave sizes/leadership values from 75-288.
You've spent more time on it I'm sure, so I'm willing to defer to your judgement on it, it's just not what I'm seeing in testing. Perhaps different servers are running slightly differently in regards to leadership?

