Interesting, but why compare only three photos? Can hardly get any type of conclusive results with only three photos.
There can be biases in the type of testing, and not necessarily with the user.
The high end phones' cameras (and processing SW) typically excel at a particular setting/type of photo or video (studio lights, low light, no light, night time, indoors, outdoors, portrait, video, zoom, etc.), but typically not one one excels at them all.
In Marques Brownlee's test, they only did three setting types, all were photos, and none were video.
Maybe the iPhone just doesn't do as well in those types of settings.
Check out some of MaxTech blind unbiased camera tests, and spoiler alert, the iPhone doesn't always win them. What I like about Maxtech's camera tests is that they do a few photos or videos for many different settings, and not just three different photos like in Marques Brownlee's test.
Here is there latest one: