Sloan'25

[schtevie]

Mike, I see what you're driving at (stupid joke intended) but i don't think it's exculpatory. The change in shot selection over time lets one infer the quality of those close-in shots that didn't go quite as planned. And they were crap. (That's exactly why they were replaced over time by three-point attempts.)

And they were still crap making any reasonable assumption about how often they would have led to a shooting foul. (Not a newsflash: NBA players don't tend to intentionally foul crap shot attempts.) Perhaps (well) higher than average ORs followed such shots; perhaps not (does anyone have any memory of any such story ever having been told?)

The NBA average is inclusive of the most dangerous scorers and everyone else, and we're talking the NBA average here.

[Mike G]

Perhaps (well) higher than average ORs followed such shots; perhaps not (does anyone have any memory of any such story ever having been told?)

Well, Dominique Wilkins always led the Hawks in FGA and near the bottom FG% and also (before Moses and Willis) led them in O-Reb. The team regularly had one of the best OReb% in the league.

I think of a "scoring attempt" as one that may be a good shot (esp. if it goes in) or a worse one, or free throws. A pass to a guy near the basket is more likely to be a TO than one to a guy further out. That's the main reason Jarrett Allen doesn't take the majority of shots for the Cavs.

You don't foul a guy who is about to take a crap shot. You make it a worse shot by getting in his way. Sometimes you get away with hacking him. Then you can call it a crap shot. Late in the shot clock, any shot is better than none.

[Crow]

82games study indicated low paint turnover rate was 2nd lowest location, only worse than corner 3 / catch n shoot. Half of high paint TO rate, with more help available and temptation to dribble. 40% less than baseline 2 with similar thoughts.

[Mike G]

League per game averages this year:

Code: Select all

3fg   3fga  3fg%   2fg   2fga  2fg%    FT    FTA   FT%    TO
13.5  37.5  .360   28.1  51.6  .545   17.0  21.7  .783   14.3

2fg% is 1.5* 3fg%

Now we may guess, for this era or another era, what fraction of FT and TO ensue from shots inside and outside the arc; and how this disparity may result in relative efficiencies from each range.
Here are some samples; they assume 2-pt shots involve a majority of both FTA and TO; and in each of 2 mini-tables, the top and bottom listings aren't intended to be 'reasonable'; perhaps plausible.

Code: Select all

% via 2fga     pts/att.       % via 2fga     pts/att.
FTA   TO     2fga   3fga      FTA    TO    2fga   3fga
1.00 1.00     .97   1.08     1.00  1.00     .97   1.08
.90   .90     .98   1.06      .95   .90     .98   1.05
.80   .80     .99   1.04      .90   .80    1.00   1.02
.70   .70    1.00   1.02      .85   .70    1.01   1.00
.60   .60    1.01   1.01      .80   .60    1.03    .97
.50   .50    1.02    .99      .75   .50    1.05    .95
.40   .40    1.03    .98      .70   .40    1.06    .93

In each batch there is a point where 2's and 3's are equally profitable.

[Crow]

BRef replied but it was unseen in my spam filter til now.

"The distances are generated from the shot coordinates input by official scorers, and 'lay-up' or 'dunk' are shot types inputted by official scorers. so fluctuations could be both real or the result in different types of decisions being made by scorers (or even a change in official scorers from year-to-year and people who may treat shots slightly differently). We are not measuring shots or assigning descriptors to shots ourselves and so cannot reliably explain the cause of any sort of variation across seasons."