Maybe I'm biased because I'm a no-name physician at a community hospital, but the "best doctors" in my eyes are the ones with the best outcomes, lowest utilization, highest efficiency and highest patient satisfaction. These measures don't correlate well with the physician's credentials. I learned this in residency, which I did just down the street from a "top" university hospital. We out-performed that place on every level except our doctors went to "lesser" medical schools. They simply spent more money (on wild things like ECMO, which makes everyone feel they're making more of a difference even though the outcomes are the same without it).
Our profession is moving toward identifying these doctors and promoting their behavior to the entire field.
There's a study that showed lower heart attack mortality when all the "top" cardiologists were out of the hospital attending a cardiology meeting. Too many confounders to make any real conclusions from that, but it lines up with my point.
Of course, if your child has a serious life-threatening and/or rare disease, then you want her to be treated at a vanguard institution like the one you took yours to. No dispute from me on that one.
But in the system as a whole, we make the most difference reducing complications of cardiovascular disease, sepsis, pneumonia, COPD, and kidney disease (which are orders of magnitude more common). And we're starting to find out how to identify doctors that do it better, with fewer complications, and with much higher efficiency than others. I think those are the best doctors and it's what I'm striving to be.