To be fair analyzing charts (WiWavelength) is not the end all be all.
It would be like analyzing a say 350hp car horsepower rating reported to the government. But then people drive the car and realize its a heck of a lot faster than the horsepower numbers suggest in real life.
Real life is what matters, numbers are for bean counters.
If this was as wide spread terrible in the real world it would be a huge story; especially people in the midwest where coverage can be spotty. But we're just not hearing about it from more than a handful of people.
It makes you think this is something software not holding onto certain towers/bands which is why you see it in rural where towers are spread out much farther; and some are seeing it in the big cities too where its very congested and lots of towers to switch between.
I agree with you that real life reviews are the most important, and we're seeing some of those reviews in this thread. It's hard to judge how widespread this is. If you're in a strong coverage area, you probably won't see the problem. And online commenters are a tiny fraction of overall users, who probably never heard of Macrumors, Ars Technica, etc. You can't dismiss the FCC data, though. If it were so inconsequential, the FCC wouldn't require it at all. In addition, it allows you to compare RF performance data between two phones in a controlled environment. Also, antennas are designed using software simulation, which is theoretical, but usually bears closely to real-world performance. I've mentioned a few times that we really don't know why some users are experiencing poor LTE performance. The only data we have is Apple's FCC filing as interpreted by some random guy who seems to have some knowledge of RF performance.