In CASP X, they told us which loops are bad. In CASP XI, we just have a GDT, which we have no way of measuring. Lua solutions exist but they are undesirable for various reasons.

For CASP refinement, we would like something that would communicate how far we are from the start. We've seen things like this in certain constrained puzzles.

In general, we would like some sort of delta (GDT, RMSD) vs. guide for any puzzle.