During development of the original ANSI standard for C, the Cray rep and others wanted to add features that would enable C to kill Fortran. This meant adding complex numbers (which happened in C99 and was made optional in C11), and doing something about unrestricted pointers. The latter lead to the noalias fiasco, but ultimately we got strict, which accomplishes the same goal.
Actually, specific pieces of code get optimized, and compilers that can "undo" optimizations put in to the source code by hand are so rare I daresay they don't exist.
Of course compilers are relevant, but Fortran has long presented optimization possibilities that aren't typically available in other languages, like not passing by reference.
Luckily there are things like Julia now that may eventually change the landscape.
Is this blind speculation?
However, programming in Fortran is uncomfortable. Just like programming in C. The advantage of Fortran over C in High-Performance Computing is that it's less uncomfortable to write maths and things with arrays in Fortran.
If you ever need to handle enormous arrays without breaking a sweat or solve systems of equations with hundreds of thousands or millions of variables, you can give Fortran a try. The syntax has come a long way since FORTRAN 77.
Fortran's strength has always been speed. In the early days most Fortran programs would run faster than C programs. The trick is to intentionally leave out features that bloat the language such that all there's left in the language can easily be optimized by the compiler. Only after the C99 standard (I think), C caught up to Fortran in most applications due to stricter aliasing rules.