This isn't entirely accurate. Think of the exact same system performing the same intra-system permutations (changes) but in two different reference frames (one

This isn't entirely accurate. Think of the exact same system performing the same intra-system permutations (changes) but in two different reference frames (one relativistic and the other not). The changes are identical but the time required by each will be different due to their different reference frames. If time was simply a measure of change than you would expect the time taken by these two systems to be identical. Time then takes on a quality (or rather perhaps has a dependency) that exists beyond simply a "measure of change".