“…1 we show E and M as a function of P, including both perturbative and numerical values and their difference. Since the leading order error of the perturbative results scales with 4 P we find remarkably good agreement even for moderately large values of the momentum P. For conformally flat Bowen-York based initial data, P=M 0:6 at the predicted location of the innermost stable circular orbit (ISCO, e.g. [3,4]), at which point the relative error of the perturbative values is approximately 2% for the ADM energy, and less than 1% for the irreducible mass.…”