We investigate star formation rate (SFR) calibrations in light of recent developments in the modeling of stellar rotation. Using new published non-rotating and rotating stellar tracks, we study the integrated properties of synthetic stellar populations and find that the UV to SFR calibration for the rotating stellar population is 30% smaller than for the non-rotating stellar population, and 40% smaller for the Hα to SFR calibration. These reductions translate to smaller SFR estimates made from observed UV and Hα luminosities. Using the UV and Hα fluxes of a sample of ∼ 300 local galaxies, we derive a total (i.e., sky-coverage corrected) SFR within 11 Mpc of 120-170 M ⊙ yr −1 and 80-130 M ⊙ yr −1 for the non-rotating and rotating estimators, respectively. Independently, the number of core-collapse supernovae discovered in the same volume requires a total SFR of 270
+110−80 M ⊙ yr −1 , suggesting a tension with the SFR estimates made with rotating calibrations. More generally, when compared with the directly estimated SFR, the local supernova discoveries strongly constrain any physical effects that might increase the energy output of massive stars, including, but not limited to, stellar rotation. The cosmic SFR and cosmic supernova rate data on the other hand show the opposite trend, with the cosmic SFR higher than that inferred from cosmic supernovae, constraining a significant decrease in the energy output of massive stars. Together, these lines of evidence suggest that the true SFR calibration factors cannot be too far from their canonical values.