Hi, I think there a few issues with your system & test protocol.
System Issues. I was surprised to see your 9700K reach about 97 degC during the Cinebench loops while the CPU Package Power was only peaking at about 148W. The 360mm AIO cooler should have cooled it more efficiently that that. I have a delidded 4 core Intel CPU, i7-6700K and a Noctua NH-D14 air tower cooler and it keeps the CPU in the high 60's & low 70's at a 140W load (this is at 1.4V and 4.7Ghz) - so I think there's some kind of an issue with your 360mm AIO cooler or perhaps a problem with the soldering on your 9700K. That might warrant further investigation, especially if you're gonna be using that same 360mm AIO cooler in future reviews.
Test Protocol Issues. You say you don't notice much difference between 9700K and 9900K, but a lot of your testing with the Vega64 GPU is GPU bound rather than CPU bound, so that is contributing to the lack of difference seen. During your 1080p testing the Vega64 is mostly GPU bound. The 720p testing is a lot more valid & showing greater differences between CPUs, so I think you should have just used that data to derive your performance scores, if you've included the 1080p results in your final calculations then you've muddied the water.
On a general note, 9700K is a good gaming CPU, that's for sure, so is the 9900K, ha & likewise the 8700K.