Tokenize and split to row -- I will remember that for the future. I brute-forced the replacement of each of the first word's characters with '' in the second word. Any characters left over in the second word meant it was not an anagram. Had I figured out the split-to-row, I would have used a different solution.