Hey guys, i've recently been looking at Basketball statistics and I came across this very interesting blog which has recently studied the case of 3FG% on severall players such as Trevor Ariza. I thought I would share this with you guys since his FG% has been discussed throughout the BBS. Here's a snippet of the article, too understand it i'd advise to check out the rest of the post. Happy new year! link
It looks like Ariza should keep the 3-ball in his arsenal, but he should attack the basket more and come off screens and jump shoot more.
This model sounds pretty worthless to me. The 95% confidence level ranges from 27% to 43%? Wow, that sounds like quite a prediction.
Adelman should buy a whip and whip Ariza every time he takes a contested three. Good god, this graph sucks!
I don't get Nash's graph. His percentage goes up as he ages, not like the "round mound" curve the author is projecting. Also, I'd like to see Jason Kidd's graph.
Just wanted to comment and help the interpretation of this. The benefit of this model is that is shrinks our uncertainty about Ariza's ability based on his performance. This is done by using our knowledge of how other players in the NBA perform. The confidence interval Steve_Francis_rules quoted was for his actual performance this year. This model shrinks that uncertainty from 16% to 8%. The points on the graph represent the percentage the player posted in each year, and it is a function of the player's role in the offense. The curve represents the best estimate of the player's ability. We should not expect a player's actual performance from each season to follow this smooth curve. That's why the model is useful, as it allows us to use the seasonal information to help us figure out the player's true ability. Here is Jason Kidd's graph. Note that since the move to Dallas he's decreased his offensive load (represented by the area between the blue and black lines). This increases our expectation of his observed 3FG% performance.
Question: Is the assumption that increased usage leads to decreased efficiency? That's the way I interpreted the positioning of the blue lines and red lines based on their associated usage rates. Edit: Nevermind, I think I found the answer. I am usually hesitant to read statistical stuff because it stresses me out. "USG%: ...for each additional 1% in an individual’s USG% the odds the individual makes a 3FG attempt are decreased by 0.6%. As we would expect, this suggests that a player that increases their usage from 20% to 21% would expect to see their odds of making a 3pt FG attempt decrease by 0.6%"
Nash's graph is probably skewed in the same way his career has been. His peak has really been the last five years, and over that same time period his efficiency has been at it's highest. I guess going on that trend, it would seem like his percentages increase with age.
Yes, I'm seeing an attempt at analysis in this thread. Thanks for joining in basketballgeek - I've been reading your blog and using your data a lot lately. It's been added to my "daily routine". If this is Ryan, your appearance is ironic, since I emailed you a few days ago about the row terminator (\n) issue.
