While technically true, you can still improve performance of existing Elisp code by redefining certain commonly-used Elisp functions perform better in Deno.

Sure the Elisp bytecode itself doesn't evaluate faster, but that distinction isn't really important here. The end-to-end perf of certain functions can be improved in a backwards-compatible way.

If we're entertaining writing elisp methods in a faster language to improve overall performance, then it's already been done and being done via writing methods in C.

I don't see the point of polluting Emacs with Javascript.

Instructing users who want high-performance code in Emacs to write their code in C and rebuild their Emacs binary, or figure out how to make it a dynamic module, is about the most user-hostile way I can imagine to improve Emacs performance.