I wonder what effect +OptimiseStringConcat has on those benchmarks. It's supposed to collapse a lost of .append calls and tries to use the minimum amount of StringBuilder objects
However with a normal JVM the only way to know if its happening (AFAIK) is to look at the compilation logs or JIT'd assembly, bytecode isn't enough
In any case, I re-ran the benchmarks with -appendJvmArgs -XX:+OptimizeStringConcat and saw no significant difference in the timing on OSX. You can of course run the gist linked at the bottom and see for yourself whether there is an impact for your OS but there is JEP280 which is looking at generating invoke dynamic for Java 9 which will change the measurements again and make the -XX:+OptimizeStringConcat flag obsolete.