If you fold a puzzle for a while you eventually reach a dead end, but see lots of players ranked above you, almost mockingly.
It's something like this: http://cnx.org/content/m11467/latest/funnel.jpg You reach a peak/pit at the bottom of the fold, but are obviously not near the "native fold", or at least not the most thermodynamically-favored fold.
You try some iterations of various fold, shake, etc... You take a good look at the secondary structure to see if loops are so obviously hydrogen bound o each other that they most be helices or sheets. You try to to shift the sheets a little for better alignment. You try to rebuild the backbone if it is bright red. You resort to scripts like Fuse/Fuze. You try some of the scripts that have a scientific bent to them (push hydrophillics out of the protein, pull hydrophobics to the center). You try a compress script.
But you're still locked at the same score, without even a decimal change of improvement.
WHAT DO YOU NEXT?
A. Quit the Puzzle?
B. Do something drastic and illogical to the protein, then refold (Exploration)?
C. Something else to the protein (please state in the thread)?