Repeated measures vs. repeated plantings in perennial forage grass trials: an empirical analysis of precision and accuracy