In our experience, linear metrics are by far the

We thank the reviewers for careful examination of our paper. For the rebuttal, we use references from the main paper plus some references added here. Note that most existing linear-fractional metrics satisfy these conditions [7, 11, 12]. We conjecture that our bounds are tight (section 7), but we leave a proof for future work. We thank the reviewer for pointing this out and will clarify in the final version.