Yeah, as Reelya has said a while back, the IQ tests are really just tests on how well you do tests.
Doing good on the IQ tests also happens to correlate well with a range of cognitive abilities associated with what people commonly think high intelligence means. So if somebody scores significantly higher than you, then it's a good bet you're less of a bright spark in comparison.
Dismissing those tests as worthless is unwarranted, no matter how soothing to one's soul.
Like I said above, though, they aren't calibrated to accurately measure differences in high scorers. Their value lies in detecting and measuring cognitive impairment, not in giving you an ego boost because you test well. "What people commonly think high intelligence means" is inherently circular. When people hold up their high IQs to describe how smart they are, of course a lot of people are going to think their traits embody high intelligence.
Put bluntly, the type of people with a burning need to prove how smart they are generally aren't all that smart, and having a number that's bigger than the other guy's is an appealingly simple way to turn relative intelligence into a genital-measuring contest. The same is true of MENSA and other people-who-need-to-prove-their-high-IQ societies: they tend to be short on scientists and engineers, but absolutely stuffed to the gills with various kinds of professional wingnut who need to sound impressive to morons and folks you'd never think were smart if they didn't go out of their way to "prove" it. Doctors don't join MENSA. Homeopaths do.
Reelya has done a fine job of finding lay-accessible articles on how poorly high IQ correlates to actual achievement, which is probably the most objective measure of capability we can agree on. It can measure how dumb you are, but not how smart.
Of course, when Trump scores an 80 I'm sure we'll hear all these arguments from Kellyanne Conway.