I don't think that "Not always" is a valid answer to "Haskell can be faster"; or rather, in the sense that it is a valid answer, I don't think your link proves it.
Of course. There are some benchmarks where Perl is faster than Java 6.
Why haven't you linked to them?
I want to discuss the nature of benchmarks, not which line of Java to change to make one program run one second faster.