Some of the above about how normalization might affect this gets a bit out of my comfort zone, but overall, I think we're basically coming back around to the idea that there may be more variability in deltas in specific metabolites. Exemplified by this amazing illustration:

Scenario #1 with ME/CFS, where there is a much smaller mean change, and a similar spread, is what would make you more confident about a "lack of change". In scenario #2, where even though the overall mean change is zero, there is a wildly larger spread in changes. Making you think it's not simply a "lack" of change, but more like some patients are changing just as much as HC, while others are changing just as much in the opposite direction.
And there still might be an interesting difference between groups (or maybe not super interesting), but where the difference is more like the ME/CFS group's metabolites are dysregulated and go all over the place while the HC group's metabolites consistently increase, as opposed to a "lack of adaptation".
Getting closer?

Scenario #1 with ME/CFS, where there is a much smaller mean change, and a similar spread, is what would make you more confident about a "lack of change". In scenario #2, where even though the overall mean change is zero, there is a wildly larger spread in changes. Making you think it's not simply a "lack" of change, but more like some patients are changing just as much as HC, while others are changing just as much in the opposite direction.
And there still might be an interesting difference between groups (or maybe not super interesting), but where the difference is more like the ME/CFS group's metabolites are dysregulated and go all over the place while the HC group's metabolites consistently increase, as opposed to a "lack of adaptation".
Getting closer?