This paper by Murphy et al. used cephalopods as a case study to demonstrate the necessity of better biological realism when making inferences with size-based food web models. They showed that failing to incorporate feeding and growth characteristics for important functional groups in the modeling process led to biased estimates of ecosystem functions such as biomass, production, stability, and turnover time. This study offers some useful insights, and the message has broad relevance. The comments I have are mostly about framing.

I don’t think the title correctly reflected the purpose and the conclusions of this study. To me, this study was really a cautionary tale about improving biological realism in models than an in silico test on what happens when a food web has a more diverse predator-prey size relationship. I was expecting a very different paper when reading the title, and the paper was not set up in the way that the title suggests.

I also thought the paper could perhaps have struck a better balance between highlighting the idiosyncrasies of cephalopods and the broader significance of the findings with them as the focal group. I have a couple of thoughts on this. First, I think calling cephalopods “rule breakers” might be an overstatement. This is not to take away from the fact that cephalopods do have some unique characteristics, but the same thing can be said for any functional group. The two main cephalopod features highlighted in this paper concerning the allometry of PPMR and growth, though perhaps different from fish, are still well within the known principles of the predator-prey size relationships and life-history strategies.

Going further along the same vein, I am not convinced that cephalopods are truly that different from fish. The uniqueness of cephalopods may have more to do with the sometimes-justifiable practice to treat all fish as one single functional group, described by the same feeding and growth characteristics. Within fish, there are species with live-fast, die-young life histories, and there are planktivorous species whose PPMR would most likely scale with body size the same way as low-activity cephalopods. So in the context of this paper, the fact that cephalopods appear to have “broken the rules” reflected more the (over)simplification on the part of fish. To me, it is perhaps more appropriate to say that cephalopods (or more precisely the modeling treatment the authors gave them) breaks the “norm” instead of any underlying rules governing the size-scaling of PPMR and growth.

Following from my thoughts above, I thought the Discussions were too cephalopod-centric. This paper makes a very convincing argument for applying the same approach to other functional groups, especially fish, when constructing size-based food web models. I feel that the authors did not emphasize this enough. I would therefore suggest rephrasing lines 233-234 about the generality of this study. I agree that the particular parameter numbers may not be directly applicable to fish or other functional groups, but the main messages and lessons from the analyses have broad applicability. Given the rather tight word limit, I would also suggest the authors to condense the first two paragraphs and make room for expanding the last paragraph of the Discussions to place their findings in a broader context. In essence, this paper would be better framed to focus more on advocating better encapsulation of feeding and growth dynamics in models than on cephalopods themselves.

Minor comments:

This paper was clearly written and easy to read. I am only pointing out a few places where more information would be helpful.

1. It is worth mentioning in the supplementary material how the modified PPMR equation differed from the conventional equation and the rationale behind this particular manner of modification.

2. lines 133-135: I have a hard time understanding what was done here, and the supplementary material didn’t really offer more information. Some clarifications would be helpful.

3. line 154: It would be helpful to clarify what the data points were when calculating CV. I understand this is a very common analysis but this added information would help readers that are less familiar.

4. line 165: Spell out RMSE the first time it was mentioned.

5. How turnover time was estimated should be mentioned in the main text.

The paper is now significantly improved, and I appreciate the effort the authors put into the revision. One last minor comment: the Introduction (and corresponding section in the Abstract) now focused exclusively on the predator-prey size relationships; the life-history aspect was sidelined, only to feature significantly in the analyses, which was a little jarring. The authors could potentially start the Introduction by stating that predator-prey size relationships and life-history traits are two important aspects of size-based feed web models but are often oversimplified (or something more elaborate to this effect), and go from there.