Advanced Stats for Basketball

March 27, 2008

More Shot Charts

Posted by Eli in Raw Stats

Here are a few more shot charts along the lines of those from my previous post.

First, a chart of effective field-goal percentage by shot location, which some commenters requested. eFG% is basically a measure of points per shot (excluding points from free throws) that is different from FG% in that it gives extra credit for made threes. The formula is 0.5*(2*2PM + 3*3PM)/(2PA + 3PA). I didn’t smooth out the data like I did in the FG% chart, so each one-foot by one-foot square contains the eFG% on shots from that spot. The color scale runs from blue (low eFG%) to yellow (average eFG%) to red (high eFG%).

Where Players Score Points - eFG% by Location:

As expected, three point shots get a big boost by looking at eFG% rather than just FG%. Shots right around the basket (dunks and layups mainly) also rate well, as do elbow jumpers, anything down the center line of the court, and small areas on the left and right blocks. The red squares just inside the three-point line are artifacts of three-point makes that were mistakenly coded as being taken from inside the three-point line.

The other new chart shows in which locations shots are most likely to get blocked. Each one-foot by one-foot square represents the percent of field goal attempts from that spot that were blocked. The color scale runs from blue (few blocks/FGA) to yellow (some blocks/FGA) to red (many blocks/FGA). The results are basically as expected - most blocks occur close to the hoop.

Where Shots Get Blocked - Blocks/FGA by Location:

If anyone has any other suggestions let me know.


  1. Shot charts for PGs, wings and bigs?

    Shot charts for each group subdivided into top quartile, middle and bottom quartile by eFG% and a top vs bottom differential to show more clearly what shots the best take and hit more often compared to the worst?

    With the original all player charts playoffs vs not and East vs west or high pace vs slow pace might also be interesting to check.

    Comment by Mountain — March 27, 2008

  2. More split ideas:

    Shot distribution and eFG% for other players on court with a post player of a certain level (defined by some combo of inside shot % and inside shot FG%) vs without th ebenefit of such.

    Same for teams with a PG above a certain assist per 48 level (8 0r 10)?

    Players with x years experience vs not

    Subgroups of Players by position with exceptional height or a deficit (to see how much height leads to different shot choices and results)

    League average and above 3 pt FG% shooters vs those who take a qualifying level of 3s vs those who don’t.

    Players high and low on turnover rate

    Players high and low on assist rate (who does stong passing change shot charts?

    Players who meet criteria as an inside / outside threat vs just one vs neither

    Top 3 shooter on team vs rest

    By part of shot clock

    I am sure there is more possibilities if the dataset is easily segmentable

    Comment by Mountain — March 27, 2008

  3. What are you using to draw those plots?

    Comment by edk — March 27, 2008

  4. I actually did almost all of it in Excel 2007 (after getting the data out of a MySQL database). I set the column widths equal to the row heights to get a square grid with each cell representing one square foot, used the drawing tools to create the court, and then used some lookups to put the numbers directly into the cells on the court (e.g. the FGA from a specific location). Then I set a conditional formatting rule that automated the color coding, turned off the gridlines, and formatted the cells so that the text in them wouldn’t be displayed. The last step was to take a screenshot, crop it, and save it as a PNG file (I did that in GIMP but it could have been done in MS Paint or whatever).

    Comment by EliMarch 27, 2008

  5. Do you know of a way to download nba games play by play data and then set it up in excel to however one would want? I would love to know if you have insight.

    Comment by Greg — March 28, 2008

  6. Greg: do you know about

    Comment by edk — March 28, 2008

  7. Who is making those baseline jumpers? They are unassisted, mostly unblocked, and carry no benefit from the backboard. They look rare, too, which means they might be easy to lookup.

    Love yer grafx.

    Comment by Steve — March 28, 2008

  8. Offensive rebound percentage from different shots on the floor.

    Comment by Eric — March 28, 2008

  9. “(excluding points from free throws)”

    reduces a truly promising post to just silly

    Comment by j1mmy — March 28, 2008

  10. Thanks for the suggestions, Mountain and Eric. I just put up a chart for offensive rebounding percentage by shot location in a new post.

    As for points from free throws, unfortunately the play-by-play data doesn’t give shot locations for missed shots where players are fouled (presumably because these are not counted as official field-goal attempts). So a true shooting percentage (or points per shot) chart isn’t possible.

    I think the extreme baseline jumpers are probably just charting errors on the part of the scorekeepers, but I will try to take a closer look.

    Comment by EliMarch 29, 2008

  11. That was great i loved it. But i think that you should make a chart of lefty shooters vs. righty shooters. See ya

    Comment by G-DogMarch 29, 2008

  12. I agree you really should make a chart of lefty vs. righty shooters.

    Comment by BasketballMarch 29, 2008

  13. On J1mmy’s comment #9, eFG% isn’t silly but TS% would be worthwhile, better for some things, and offensive rating by spot would be even better from an overall efficiency standpoint.

    Comment by Mountain — March 30, 2008

  14. Well I should have thought about that a little further. You can’t assign turnovers to shots that didnt happen and there are none when the shot did get off. Maybe from video you could in a general way though to 3 pt, midrange and inside takes.

    Comment by Mountain — March 30, 2008

  15. Lefties vs. righties would be interesting to look at. I’ll look into it.

    Comment by EliMarch 30, 2008

  16. The shots by efg is exactly what I anticipated — red around the hoop (for dunks & layups), and red at the 3pt line. Thanks for posting this. Very interesting stuff.

    Comment by kjbMarch 31, 2008

  17. Given a shot chart and using a scale of feet left-right (assigned negative-positive) and away from basket you could calculate the central point of all of a player’s or team’s shots. That would give a first basis for comparing charts but maybe there is a middle ground between that and the original shot chart. Maybe you could find the average location of a shot in each of the 4 quartiles of the graph surrounding that central point and make simple visual comparisons with that. Or maybe go back up one step in detail and show the 3 average shot locations (using least distance from a point to actual shots or simply highest frequency) in each quartile of the chart to give the 12 most common shots of a player and display that as moderately detailed but more digestible representation of their pattern. Perhaps a higher level method could retain all the detail for the player similarity study but a visual would still have some appeal.

    Comment by Mountain — March 31, 2008

  18. Trichomonas vaginalis however is definitely an evil beast; you could not count on anything much less of the creature that lives out its complete life inside your genito-urinary tract. It triggers a terrible little illness from the similar name, but you should get fairly up close and private to catch it. For those who genuinely want to read about all the disgusting signs and symptoms, then the BBC Wellness website has some thing bordering on ‘too significantly information’..
    air jordan 6 rings

    Comment by air jordan 6 ringsJanuary 15, 2014

RSS feed for comments on this post.

Leave a comment