There has been a problem since the late '90s with mastering, ever since the release of the Waves L2.
Waves is a digital audio company, who makes what are known as 'plugins' for audio recording and mastering software. The L2 is a Loudness Maximizer, which lets engineers crank up the volume with a single level fader. Any idiot can use it, and therefore, many idiots do.
The problem with cranking it up is that music can only be so loud in digital format - it's called 0db. There is none louder. So instead, as you turn it up more and more, more and more of the recording has the same very loud volume. Loud parts, quiet parts, in between parts - it's all loud, which means it lacks depth and dynamics. It starts to distort and break up. It's just one ill-defined fatiguing mess.
So why do they do it? Because the last loud album did it, and they want to be louder than the last guy.
Stupid, isn't it? Instead of mixing so the listener turns up their volume (which anyone with a stereo can totally do), They think they need to turn it up in the mastering process, which results in what is known as "ear fatigue" where everything just starts sounding flat and the same and sonically uninteresting, if not outright annoying. Ear fatigue leads the listeners to either turn the music down, or off completely. So these mastering engineers are actually having the opposite effect. But it's not their fault entirely - there is some record exec or A&R guy (or maybe even the someone in the band - probably the drummer!) saying "yeah, make it one louder than that other guy!" with apparently no awareness or concern that they are destroying the integrity of the music.