This is a very nicely written paper that generalises the results of the authors’ previous study, showing that the most extreme windstorms in all of the storm track regions have a downstream pre-existing cyclone. The paper also demonstrates that a low-resolution idealised model can capture the hemispheric asymmetry in the extreme winds (where the NH storm tracks have stronger extreme winds than the SH). Experiments with this model indicate that the asymmetries are related to the Eady growth rate, and the topography.

 I have a few comments and suggestions on the manuscript.

1. Every instance of “northern hemisphere” and southern hemisphere should be Northern Hemisphere and Southern Hemisphere.
2. Lines 54-55: I think the use of this intermediate complexity climate model is a key point of novelty in this study, therefore it should be given more attention in the introduction. I suggest including a paragraph on the use of such models and what previous studies have found when doing experiments such as the ones you have done.
3. Line 142: I’m not sure about the benefit of saying that you ran 11 different experiments and will only discuss 3. It might be better to fully justify the experiments that you are including – why choose these set-ups? What were the hypotheses you were thinking about when running the experiments?
4. Line 153: I’m not sure what you mean when you say the “prescribed SSTs are found”. Surely if you prescribe them you do not need to find them. Perhaps this just needs rewording.
5. Line 157: Can you say how you know the surface winds are at a higher level in the model?
6. Section 3.1: I think it would be really beneficial to include composites of the winds that you are trying to explain. This would help to see if there are structural differences between the NH and SH too.
7. Line 221: typo in “central”.
8. Line 284: Can you refer to a figure here?
9. Line 287: It would be good to make it clear here that this result generalises the findings of your previous study (rather than being a brand new result). This is also the case for lines 300-303.
10. Line 315: I think rather than saying when the jet core “weakens” it should say when the jet core “is weaker”.
11. Line 339-340: This statement that it is unclear if the model can represent the storms is a key point. I think it would be good to either do the tracking and look at the individual storms in the model, or explain fully why this is not possible. The model may be quite low resolution, but some tracking algorithms (e.g. the Hodges method) track on T42 vorticity and so it should be possible. You find that the model can represent aspects of the extreme wind climatology, therefore we hope that this is due to the storms that are present in the model. It would be beneficial to show this somehow.