APBRmetrics

[Eli W]

In case anyone hasn't seen these, I've assembled a play-by-play database with four seasons of shot location data that I can use to make some interesting charts. So far I've made ones for FGA, FG%, Ast/FGM, eFG%, Blk/FGA and ORB%.

http://www.countthebasket.com/blog/2008/03/27/where-players-take-and-make-shots/

http://www.countthebasket.com/blog/2008/03/27/more-shot-charts/

http://www.countthebasket.com/blog/2008/03/29/rebounding-by-shot-location/

If people have other ideas for charts I can try them out. One idea I have is basing player similarity scores on these kind of charts. I also may look at some individual player or player combination charts (e.g. where on the court does Nash get his assists, what does Marion's chart of unassisted made shots look like compared to his chart of assisted made shots, etc.)
_________________
Eli W. (formerly John Quincy)
CountTheBasket.com

[Mountain]

Player shot chart similarity sounds like an outstanding direction to pursue that I'd welcome seeing.

Your choice over time of course but did any of the 15 chart ideas I listed at your site strike you as something you want / intend to try?

[hoopseng]

I love the "shooting temperature" charts, congratulations for great work!

It would be exciting to have team by team allowed FGM locations if it is doable?
_________________
http://www.nbastuffer.com

[Eli W]

[Eli W]

[blb2397]

Any thoughts about a data only listing for each area or creating a chart with less grids as well. i.e 3 by 3 or 5 by 5 instead of one by one for a more general view.
Great work!
_________________
NBA Post Up with Player Report Cards

[Mountain]

Ok Eli. Just wanted some sense if the comments were useful to you. Thanks for sharing the research and continuing to dialogue and push further.

[Brian M]

What would be neatest of all would be if you could construct a flexible b-r.com like method for generating these charts based on general input parameters, if possible. This way interested folks can plot out anything they can dream up rather than having you labor through making each individual chart.

[Eli W]

[bballfan72031]

Eli, the work you've been doing is some of the most spectacular stuff I've ever seen.

The big thing I've been hoping would be done for a while now is an analysis of diminishing returns on shot locations.

There is an assumption that a team can't simply put a bunch of post scorers on the court at the same time and perform as well as the individuals would perform if they weren't getting in each others' way. There is a need for balance in basketball, and it seems to me that this need is huge.

I assume this could partially explain why Channing Frye has shot better his first and third seasons, playing primarily center, while he was considerably less efficient shooting his second season when he played primarily power forward alongside Eddy Curry.

I would guess that there would be a larger affect on diminishing returns the closer you get to the basket (it's likely that it is easier for multiple three point shooters to spread out and have their own space to operate than multiple inside scorers), yet it would still be optimal to have a balance of inside and outside scoring.

I think an analysis of diminishing returns on shot locations could be immensely helpful for coaches in analyzing who to play together and for general managers in analyzing how to put together teams. It would be another step toward statistically analyzing the importance of balance.

Anyway, I sound like an annoying fool saying "Someone should do this" and not doing it myself, but I honestly just don't have the background or have any clue how to go about doing this. And with the work you and others here have done, I thought it could be a helpful suggestion, if it hasn't been made already.