my shooting RAW instead of JPEG goes back very long. I used to only shoot JPEG, even back in my E1 days, and I must even acknowledge that the JPEG output of the E1 was (is) stunning. Having said that I then started to really play around with RAW. I did this first with C1 then with LR and finally Aperture. I almost never used the camera vendor sold RAW software. And I found over the years that - what most others find - RAW gives you much greater freedom compared to JPEG. So this is the reason why I am using RAW exclusively today. Whenever I try JPEG out of camera, I immediately see artefacts etc, which I almost never do when processing JPEGs from RAWs from one of the above mentioned programs.
Why do it in camera with rather limited processing capabilities if you can do it much more sophisticated in post processing with much more capabilities. I consider it as an advantage that some companies rely just (or almost heavily) in post processing products, as this gives them a lot of experience (obviously) compared to the vendors, who might have for sure more experience with their own camera/sensor setup, but actually I can only say this for Hasselblad and Phocus and actually in this case you have again only RAW files coming out of the Blad.
I agree that here are some camera vendors who are producing excellent JPEGs but I think you can count them on the fingers of one hand And the same limitations for changing something in your images after they are out of camera are still there. Creative filters? Well ok, they are nice and I like to play around with them, but then again I can achieve all this and much more in post processing in PS. So why bother and waste my time with camera built in JPEG engines?
But of course everybody has different needs and preferences. In my case (and I think also in yours) the preference seems to be RAW anyway.
The one thing which strikes me is if there are tons of comparisons from out of the camera JPEGs, as I said already above, this outcome is (has) to be based on very limited resources from in camera. What one gets there can be good in certain conditions, but in general you are limiting the possibilities of a sensor and lens and camera combo significantly. Plus you tweak what will be the result in terms of different camera built in optimizations. And then you start comparing? All my history as an engineer tells me that this is wrong, because one saying is still true, even in the most advanced and digital days - "who measures a lot, measures just nonsense" translated to JPEGs "who plays around with camera built in processing and changes the output and then compares compares just nonsense".
Sorry for this lengthy reply, but one cannot just forget reality.