We study the implications of interpreting the recent muon gµ −2 deviation from the Standard Model prediction as evidence for virtual superpartners, with very general calculations that include effects of phases and are consistent with all relevant constraints. Assuming the central value is confirmed with smaller errors, there are upper limits on masses: at least one superpartner mass is below about 350 GeV (for the theoretically preferred value of tan β = 35) and may be produced at the Fermilab Tevatron in the upcoming run, and there must be chargino, neutralino, and slepton masses below about 600 GeV. In addition, tan β must be larger than about 8.