New final results from the CMD-2 and SND e+e- annihilation experiments, together with radiative return measurements from BaBar, lead to recent improvements in the standard model prediction for the muon anomaly. The uncertainty at 0.48 ppm--a largely data-driven result--is now slightly below the experimental uncertainty of 0.54 ppm. The difference, a_mu(expt)- a_mu(SM) = (27.6 +/- 8.4) x 10^-10, represents a 3.3 standard deviation effect. At this level, it is one of the most compelling indicators of physics beyond the standard model and, at the very least, a major constraint for speculative new theories such as SUSY or extra dimensions. Others at this Workshop detailed further planned standard model theory improvements to a_mu. Here I outline how BNL E969 will achieve a factor of 2 or more reduction in the experimental uncertainty. The new experiment is based on a proven technique and track record. I argue that this work must be started now to have maximal impact on the interpretation of the new physics anticipated to be unearthed at the LHC.