There has been considerable debate about changes in the temperature of the troposphere measured using the Microwave Sounding Unit (MSU) instrument or radiosondes. Fu et al. linearly combine time series from two MSU channels to estimate vertically integrated 850-300-hPa temperatures and claim consistency between surface and free-troposphere warming for one MSU record. We believe that their approach overfits the data, produces trends that overestimate warming and gives overly optimistic uncertainty estimates. There still remain large differences between observed tropospheric temperature trends and those simulated by a climate model.