While I think your response is justified in this specific case given the small sample size, I don't think making two "contradictory" complaints is necessarily w

While I think your response is justified in this specific case given the small sample size, I don't think making two "contradictory" complaints is necessarily wrong in this context. If you make a lot of Type I errors, it might be the cost for reducing Type II errors, and if you make a lot of Type II errors, it might be the cost of reducing Type I errors, but if you make a lot of Type I and Type II errors, then maybe whatever you're doing just sucks.