Certainly, feel free to use any content here that you’d like. Good luck with your book.

The intended use for these charts would be for use in my forthcoming book (entitled “Get It: A System for Situational Genius”) to reinforce the point that our behavior should be informed by both the best analytical insights as well as world class intuition. I think your charts illustrate this point perfectly. The book is something of a “theory of everything meets best business practices”, with the target audience being young professionals. The book would be available for sale in both print and digital versions, first targeting the US/Canada market, but potentially going international from there.

Of course, I would fully credit you and TheSpax per standard citation practices (usually immediately below the image). Please let me know if you are able to grant permission to use this material. If you require any additional information or would request a licensing payment or have specific conditions for use, please do not hesitate to contact me. Thanks. Brett

The problem is with the headers. Every so often, there’s a change in the headers required to scrape from the NBA stats site.

You can replace the headers= ... line with headers = {'Host': 'stats.nba.com','Accept': 'application/json, text/plain, */*','x-nba-stats-token': 'true','User-Agent': 'Mozilla/5.0 (Linux; Android 6.0; Nexus 5 Build/MRA58N) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Mobile Safari/537.36','x-nba-stats-origin': 'stats','Origin': 'https://www.nba.com','Referer': 'https://www.nba.com/','Accept-Encoding': 'gzip, deflate, br','Accept-Language': 'en-US,en;q=0.9}'} and that should make the scrapes work.

There appears to be inconsistency, likely due to some restriction on scraping (i.e. you make too many calls in a short amount of time). Not sure if I ran into that problem in the past, and if so I’m not sure if I had a workaround for it as I haven’t messed with the API in a while. Sometimes this means that the first few calls to requests.get work, but the code breaks down somewhere in that loop on players and gets stuck forever. You can see this happening if you add prints to the code.

But I was able to run pass_web('GSW','2016-17','AST') at least once which created a passing network that looked like my examples in this article.

If not, how was this data collected? + Where can we get more..

It should be interpreted as an average over the entire 25 year timespan while accounting for the prior. Looking at smaller samples (like five year RAPM results) would be more appropriate for estimating a player’s peak value.

That is correct (the team plus-minus is adjusted for how many possessions were played in that stint, though – it’s per 100 possessions, to be exact). And yes, if two players are always on the court with one another, we would not be able to isolate each individual player’s impact.