That's true. I guess what makes Haskell stand out to me is that at this point it's fairly clear that the assertion is mostly wrong (i.e. idiomatic Haskell is not faster than idiomatic C at most algorithmic tasks, and often significantly slower) and instead of doing the reasonable thing and saying "additional safety and expressiveness is more important than raw performance for most applications" (something I'd completely agree with personally) they double-down with these meaningless, unfair benchmarks.

Meanwhile Rust is genuinely competitive with C performance-wise and thanks to generics and things like more aggressive inlining can actually equal or even beat C without requiring esoteric micro-optimizations. Of course the drawback is that writing Rust code is significantly more complicated and more restrictive.

And the really funny part for us C practitioners is, we know we'll rarely be as fast as machine code generated from Fortran.