Computational chemistry tends to be done in collaboratiin between specialists, especially after grad school. You also have to confirm things - you can compute whatever you like, but there's no way to know if your simulation reflects reality without confirmation.
And in what way had physics has beaten chemistry in the case of LK99? They did a post-facto DFT calculation supporting the apparent results, which may or may not be real.
It seems like an illustration of why chemistry is as robust as ever.
