Component timing over 12 real transcript tails showed the escape-stripping regex costing 49.9 ms against 10.4 ms for case folding, while 238 of 240 fragments contained no ESC byte at all. A pattern anchored on ESC cannot match a string without one, so proving absence with a byte scan (11.5 ms) short-circuits the regex entirely on almost every fragment. Unlike the ASCII fast path this applies to all input, and it is the larger of the two wins: original 101.6 ms/round 1.00x + escape-absence guard 64.1 ms/round 1.58x + ASCII fast path 57.9 ms/round 1.75x The normalize tests now compare both shipped paths against a pristine copy of the original formulation rather than against each other, so neither optimization can drift from the semantics the matcher was built on.