Above-the-break 3: 13 of 33, 39%, league 35% 13/33 39% Left corner 3: 9 of 20, 45%, league 39% 9/20 45% Right corner 3: 6 of 20, 30%, league 39% 6/20 30% Midrange: 5 of 12, 42%, league 42% 5/12 42% Paint: 15 of 27, 56%, league 44% 15/27 56% Restricted area: 26 of 40, 65%, league 66% 26/40 65% Want the video? Switch to Dots and tap any shot.
Color vs league average in that zone: 6+ pts below 2-6 below about even 2-6 above 6+ above
Shots updated Sat, Sep 26. 1 end-of-quarter heave not counted, same as the box score. Regular season only. Shot locations come from ESPN play-by-play. How zones work
Who shoots like Jamal Cain? See his shot twins
Shot profile Shot diet At the rim 26% lg 30% Paint 18% lg 19% Midrange 8% lg 10% Corner 3 26% lg 11% Above-the-break 3 22% lg 31% Bar = share of his shots. = league share.
Best zone Paint 56% vs 44% league (+11)
Worst zone At the rim 65% vs 66% league (-1)
How to read this view Each zone is colored by his FG% there vs the league in the same zone: blue is better, red is worse. Labels show makes of attempts. Glossary
Shot quality Shot selection +0.01 expected points per shot vs league
How good his shots are before anyone shoots: the points an average shooter would score from the same spots and shot types.
Shot making +5.3 points per 100 shots above expected
How much better or worse he finishes those shots than an average shooter would.
2025-26 regular season, 152 shots: 1.16 points per shot vs 1.11 expected. Under 200 shots, so no rank yet. Whole season, filters not applied.
Shots in view (152): 1.16 points per shot vs 1.11 expected (+5.3 per 100).
Shot quality leaders and how it works