Drag racers rebuild their engines after each run.
You really think cost is preventing them from running a skinny tire if they could gain a tenth?
I have corresponded with the author of the article you link and while he might understand physics, I'm not convinced he understands motorcycle dynamics in a practical sense.
The truth lies somewhere between the two. I have yet to find the answer. But the closest reasoning I've read that the coefficient of friction rather than being a constant can vary with temperature. Large contact patch might resist temperature change hence resist changes in coefficient of friction. Additionally, coefficient of friction does not accurately describe a rolling and cornering tire that operate with some slip angle. The tire is not stationary but not sliding either.
No, it is too simplistic to say contact patch doesn't matter.
From a practical motorcycling sense, much of Code's teaching talks about contact patch and friction. Based on all that practical experience of thousands of riders, there is some truth to the statement.
Just my opinion.