review of first draft about basketball HOT 5 CLOSED

I think we've now addressed the majority of the issues you raise -- it's definitely cleaned up relative to the first draft. Let us know if anything still stands out to you.

from basketball.

First of all, I'm embarrassed that it has taken me this long to respond. Please accept my apologies!

The revision is certainly an improvement. In particular, I think the end is much stronger with the concluding paragraphs. I still think there is too much code, but I guess that decision should be made by @jennybc and @hadley.

A few general comments:

• I see that you have R-ified the code by replacing a few for loops with calls to apply(), which is good. I still see places where further clarity could be achieved. For example, in plot_distribution_of_movement_distances() couldn't the for loop be replaced by a call to lag()? Anyway, I realize it's ridiculous to expect a full code re-write, but others may spot similar things.
• The language in the Conclusion, especially in the On-ball Plots section, is not appropriate IMHO. While I share the authors' enthusiasm for NBA basketball, I don't think we as academics should be passing judgement on Steph Curry's body development. Phrases like "almost unstoppable" are similarly appropriate in the media but I don't think in a serious professional discussion. Do you think that the Cavs' sports analytics group---who may read this paper---would agree that Curry is "unstoppable"?

Some minor comments:

• "data" is plural
• pg. 2: put the variables in the same order that they appear in the R output. Also, clarify that a and h stand for away and home.
• I would use open and closed parentheses to denote function names (e.g. add_possession_columns_to_moments()
• pg. 4: It wasn't clear to me what $TEAM_ID means.
• pg. 8: Is $\delta$ in the units of feet per second? If so, state that.
• pg. 9: Spell out "Histogram" at the top of Figure 1.
• I still think you need to say something at the end of the section on Movement Simulation Functions. There is a long block of code, then a figure, and then another long block of code, and then a new section! Can you remind us why we are reading this and what the point of the section is?
• pg. 10: "one-hot"?? Is that a typo?
• pg. 13: Please refer to the Figures by number in the text. All of the Figures should be referenced in the text.
• pg. 14: "quickly" instead of "quick"
• pg. 14: "most unique": I believe that uniqueness cannot be qualified -- something either is or is not unique. Also, do you mean unique among the three players you have listed? Or among all players? In any case, I'm not sure that uniqueness is really what you are trying to highlight here.
• pg. 14: I don't really follow the argument about Steph Curry's off-ball movement. Can you clarify?

Thanks for contributing this paper -- it was my pleasure to read!

from basketball.

Thanks @beanumber for the continuing review!

from basketball.

from basketball.

Thanks for the detailed review! I'm working on addressing your comments right now.

RE: one-hot encoding, I've seen it used in the literature with a hyphen. Looking it up online I see some with it (like the Wiki page) and some without.

from basketball.