ortho2003 makes a good point. You can't use the unmatched percentage as a reliable guide to how tough a given specialty is to get into compared to others. The fact that people apply to more than one specialty and match in backups makes that data useless. The real number would be the number of applicants whose first choice was ortho, plastics, ophthal, etc. vs. the number of spots. I do not know of any source where you can actually get that data because nobody knows how many people have what as a first choice. I guess NRMP could poll their computers to see how many people ranked an ortho program No. 1 vs. others and figure it out, but I doubt they ever will.