I've had the opportunity to address some of the substantive objections to including general issues affecting the editor base in advocacy goals off-list, and I would like to propose something different as an alternative.
I propose that we establish quantative measures of potential advocacy actions which score the extent to which they would advance the mission. In practice, this would mean using aggregate rubric scores for each clause of the mission, asking a randomly selected subset of the community how much success in a specific advocacy goal would advance that clause, and then using the median (not mean, to prevent outlier effects) scores to rank the different potential advocacy actions.
Here is some background information on organizations which have done similar things:
http://www.researchgate.net/profile/John_Crotts/publication/270959033_Aligni...
http://www.emeraldinsight.com/doi/abs/10.1108/AAAJ-10-2013-1488