Open cpsievert opened 9 years ago
From a really cursory look, the data is indexed by the gameID and event number (which are readily available from, say, play-by-play). So obtaining the data in a systematic fashion is feasible. Judging from the corresponding videos, however, I think there'll be a big overlap between consecutive files:
http://stats.nba.com/cvp.html?GameID=0021400877&GameEventID=1 http://stats.nba.com/cvp.html?GameID=0021400877&GameEventID=2 http://stats.nba.com/cvp.html?GameID=0021400877&GameEventID=3
Haven't thought about it in a lot of detail, but one way to get rid of the repeated location data will be with the timestamp (and possibility duplicate rows in some data frame storing player locations).
Jsonlite or tidyjson works wonders with the secret NBA api
Alex Bresler abresler@asbcllc.com www.asbcllc.com 917-455-0239 (cell) On Sep 1, 2015 4:16 PM, "skdeshpande91" notifications@github.com wrote:
From a really cursory look, the data is indexed by the gameID and event number (which are readily available from, say, play-by-play). So obtaining the data in a systematic fashion is feasible. Just by looking at the corresponding videos, however, I think there'll be a big overlap between consecutive files:
http://stats.nba.com/cvp.html?GameID=0021400877&GameEventID=1 http://stats.nba.com/cvp.html?GameID=0021400877&GameEventID=2
— Reply to this email directly or view it on GitHub https://github.com/cpsievert/bbscrapeR/issues/8#issuecomment-136847688.
Mine this post for ideas -> http://savvastjortjoglou.com/nba-play-by-play-movements.html