Recording with 'problems' emphasise the distortion of the playback - consider it something like, compound distortion. Ever notice how a "bad" recording sounds very different, on different systems? Now, if each system was doing the right thing, and perfectly replicating the recording, then that poor sound would be the identical poor sound, on each - AB testing should show them almost indistinguishable ... so, if they sound very different, one system is possibly more accurate to the recording than the other - okay, now which one is "better"?Potholes are the right thing to test a suspension and the highway is not as you say. By the same token, the right audio sample needs to be picked which makes it easy to find characteristics of audio gear. Music that is so full of distortion/degradation itself will make it hard to find the distortion in gear.
But you go ahead and play that music on your system and report back as to whether it becomes one of your top diagnostic tracks for hardware.