I develop an improved Hamiltonian for classical, Minkowski Yang-Mills
theory, which evolves infrared fields with tree level corrections from
lattice spacing a beginning at O(a(4)). I use it to investigate the r
esponse of Chem-Simons number to a chemical potential, and to compute
the maximal Lyapunov exponent. The Lyapunov exponent has a small a lim
it, and the Chern-Simons number response appears to be approaching one
at the finest lattices considered. In both cases the limit is within
10% of the limit found using the unimproved (Kogut-Susskind) Hamiltoni
an. For the maximal Lyapunov exponent the limits differ between Hamilt
onians by about 5%, significant at about 5 sigma, indicating that whil
e a small a limit exists, its value depends on the specifics of the la
ttice cutoff. For Chern-Simons number the difference between Hamiltoni
ans is within statistical errors of about 10%, which constitutes an up
per bound on the lattice regulation dependence.