Amateur Project to Practice

[bake502]

Hey guys I'm kind of scrambling for help and didn't know if anyone could help with a quick fix somehow. Right now I am attempting to use 538 CARMELO projections of WARP to create easier to read and decipher per game statistics projections. I have already converted the 538 WARP stat to the Basketball Reference VORP statistic in a set of data that tested significant in its relationship. I'm currently at a wall after converting the VORP stat to Box Plus Minus using the formula provided by Basketball Reference here: https://www.basketball-reference.com/about/bpm.html I now have projected Box Plus Minus for players and I am wondering if there is any way, outside of guessing, to start and fill in the advanced metrics that make up the Box Plus Minus equation. I understand there are a lot of variables involved in the equation and was just looking for any suggestions in this amateur run of things and admire almost everything posted here and aspire to get to that point one day. I'm running all my code on R currently and scraping off of those two sites mentioned. Thank you all and any advice would be much appreciated.

[DSMok1]

In general, I think it makes the most sense to directly project the constituent stats individually, from prior year stats and possibly aging. Most of the statistics are fairly independent of the other statistics, with the exception of volume of shots vs. efficiency on the shots.

I don't think there's any way to use the composite stats like WARP and BPM to back-calculate the constituent statistics.