Every comparison is computed from the same dataset as the True-Cost Index — the data declares the winners it can, and the rest is honest fit guidance.