Thanks for your reply; I am disappointed that it did not work the way
I thought it would ...
I suppose that you did VACUUM ANALYZE your table recently, right? If
not, the stale table statistics may be playing a role too: do it, and
try again.
With this I have reached the outer boundaries of my "knowledge"; if it
doesn't work out, I hope somebody else may be able to help. In any
case, I would appreciate knowing more about your tests and the
eventual outcome ...
Miguel