We reconsider calibration of statistical distance scales for planetary nebulae, examining precision and systematic error for various distance methods used as well as the scales themselves. A different calibration strategy, one based on precise trigonometric parallaxes by Harris et al. (2007; some improved by Benedict et al. 2009), is presented. Most statistical scales have an overall scale error; in addition, all four tested show dependence of distance ratio [scale/actual] on nebular radius. However, the H-alpha scale of Frew (2008; slightly revised in Frew et al. 2014) has only a small radius dependence at large R and overall is consistent with these parallaxes. Some methods used for calibration have systematic errors, and the extinction method is found to be mostly unreliable. Comparison of Hipparcos parallaxes for a few large planetaries with our "best estimate" distances for those confirms the overestimation of 2.5 found by Harris et al. for PHL 932. We suggest a possible connection with the smaller overestimation found by Melis et al. (2014) when testing Hipparcos parallaxes for some of the Pleiades.

29 pages, 25 figures, 20 tables, 1 appendix. Submitted to MNRAS; revised to reference Frew et al. MNRAS 2014 submission