If you are basing that off so called dynamic range databases, keep in mind that dynamic range may be based on a "time period", not instantaneous (well fraction of seconds - second) dynamic range. In a given frequency range, I would expect that to be even more the case. My understanding of it is that it shows peak to average of a given time period, not peak to minimum which would be more critical to this discussion though perhaps both are.
I know many here hate ASR, but it would be good to collectively push them to do more frequency response measurements at varied power levels, and perhaps at different frequency sweep speeds to induce this issue. I personally don't see Stereophile doing that and I definitely don't see suppliers going out of there way to highlight problems.