I watched the CDF presentation early yesterday morning at the April APS meeting. Nothing to indicate they were anything but meticulous in their data analysis and modeling. Keep in mind the main change in CDF II is a great reduction in uncertainty - the central value is not much changed from the original CDF results many years ago.

Tommaso Dorigo [1] covers a lot of the points where some of their assumptions/methods can be reasonably questioned but the effects would not push the result that much closer to the current SM theoretical value.

[1] https://www.science20.com/tommaso_dorigo/is_the_cdf_w_mass_m...