I'm trying to model real players using the Skill Ratings in the game. Can't quite get my head around FGI, FGJ and FJD. As I understand, these are three different shooting percentages for ranges inside the 3pt line. So I can ignore FT and FG3 from this discussion.
FGJ (Jumpshots) is Field Goal Percentage on 2pt Jump Shots. I'd assume it is used for shots in the 16-3PA and 10-16 distance ranges. It probably contributes to shots in the 3-10 range as well.
FGI (Inside) I assume is used for shots in the 3-10 range but probably in 0-3 also.
FGD. Seen this attribute related to both Driving and Dunks. I don't believe its related to Dunks. Could relate to shots attempted driving to the Basket and perhaps PGI relates more to shots from posting up. But the "D" could be misleading and means something else.
About 90% of players have a FGD higher than their FGI. That would suggest dunks and lay-ups come under that category. But I think there is more to it than that. Most players out perform their FGD in the 0-3 range, so I feel that FGD must incorporate a bit more than dunks/lay-ups.
Should add that the actual shots will be based on more than just the players shooting percentage at each range. Even outside their own skill attributes I suspect size/playing position will influence the shooting percentage close to the basket.
I'd be interested how a player like Matthew Marsh previously of Wake Forest would model. In 22/23 every shot (94 attempts) he took all season was close to the basket; hence his outlier 88% FG Shooting. Although he had a few offensive rebound putbacks, it was a small number (under 10%). The majority of his scores were assisted so I suspect he was receiving the ball in the (very) low post and scoring from close in. So easy to model where he shot from on court and that he didn't create his own shots, but based on that limited overview does FGD or FGI (or both) model the type of player he is?