'A ball'? You mean 'The ball' It's a single ball sitting
under the spindle
and only provides vertical thrust.
Sideways movement is controlled by the steel spindle in a
bronze bush,
at least in the Garrard, and the Roksan diagram looks
similar.
Does that change your recommendation?

Yes, sorry, I've had a thing about ball-race bearings ever
since motorbikes changed to plain bearings and suitable oil
for older bikes with rollers and balls disappeared.

Phosphor-bronze relies on being porous, so oil for that part
shouldn't have additives that might clog the pores, although
there's so little loading there that it might not matter
much.

If the thrust bearing isn't immersed in oil, then you do
need an oil that stays in place and makes a strong film. I
assume a ball rather than a cone is used so that an oil with
strong capillary action will be drawn in. Thicker oils and
greases aren't any better at staying in place under that
kind of pressure if there's nothing to keep the lubricant in
circulation, as there is in a gearbox or a wheel bearing.

I can't remember the tech term for capillary
crawling...something to do with surface energy...but that's
what's needed. An oil that crawls and forms a strong film
that repairs itself quickly if it breaks. To some extent,
all oils are supposed to do that, but there will be an
optimum consistency, so best go with the recommended type,
unless someone really knows better. I don't.

Not the ideal kind of bearing, seems to me, but anything
better would be lots more expensive. I would have thought
there's a risk of the contact point being off-axis.

Why isn't the plate made of nylon or PTFE, I wonder? Or even
graphite. Hmm...could be a market for audiophile oil-free
inserts.

Ian