Hi Frank,  for me, it's a matter of taste, but I use ball links at every end.  Considering the importance of smooth controls and the safety of the airplane, I consider the cost factor to be nil.  So, wasteful doesn't fit in my own thoughts on this subject.  You'll get to loving the smoothness of these rascals once you use them.
I do use a larger washer on the side opposite the horn to prevent the possibility of one popping off, just a precautionary thing for me.  I've never had it happen, but.......

If I used the standoff's I would use a washer there as well, but I never use them.

I'm over cautious I guess.

I don't know which ball links you have used, Robert, but I've never in over 20 years had one that was stiff.
Bob M. told me the ship got stiff in the controls because the wood swelled up, don't know, but that was a very scary sight to watch!  He did a marvelous job to save it after he fell!!

Our pushrods move both vertical and horizontal.  With music wire pushrods and brass tube bushings in both horns, the control system will bind as the bellcrank pulls the pushrod left/right while it moves up/down at the horn.  The only way to cure the bind is to introduce slop into the joints.  Ball joints are a more elegant solution, as they allow motion in all directions.

Re: Sticky Ball Links - I ordered 3 packs (30 total) of the Rocket City ball links from Nelson Hobbies a few years ago.  Of the 30 ball links, maybe half a dozen were/are free enough to be smooth.  The rest are too tight to use.  It's most noticeable when a linkage has 4 links, since the sticky joints are multiplied.  I really enjoyed cutting into a nearly finished model to replace the joints I could get at, the one at the bellcrank is still a sticky Rocket City link.  So far the large Dubro 4-40 links I've used are much more free.  

The Dubro heavy duty ball links are very good.

Is there a way to loosen them up?

http://www.clstunt.com/htdocs/dc/dcboard.php?az=show_topic&forum=103&topic_id=327637&mesg_id=327637&listing_type=search#327714

My experience was that everything I tried made the problem worse if any different at all.