NavList:

Well, I'm still not absolutely sure what's going on - but I have updated my data to the FK5 values and everything seems to be working much better. Attached is a spreadsheet with the old values I was using and the FK5 values converted to the same format for comparison. I analyzed the differences and found some surprising errors. (Hint: σOct. [facepalm]) The coefficient of variance for the difference in PM in RA is 90,559% (yes, ninety thousand percent). But the fact that all of this slipped by me in the first place just goes to show what a small difference it makes most of the time.

Frank, you mentioned trying a Google search with quotes. I thought about that, but I can't find the raw data I used and I'm certain I reformatted it for my spreadsheet.

I'm going to do some more testing and hopefully post an updated version of my almanac soon. Thank you to everyone who replied. Every post is getting me closer to understanding what's wrong. I'm still confused, but I think I'm on the right track.

Thanks!

Sean C.