Three point shots and winning percentage
Reading some of my favorite basketball blogs this morning I came across this little snippet in Ballerblogger where the author references John Hollingers per diem piece from Friday. It’s a great read, I highly recommend it. I am not a big Hollinger devotee, but he is great at noticing statistical trends. In the Friday Per diem article, Hollinger makes the connection between teams that attempt a lot of three pointers, and teams that don’t, and how they respectively shake out in wins and losses.
“…the reason teams shoot more often from out there (and perhaps employ players who do it better) is a simple one: It works. In fact, few stats correlate better with winning than 3-point attempts. If you tell me only how many 3-pointers a team has chucked up this season and provide no other information, I can tell you whether it is a winning team and be right eight times out of 10.”
“Check this out: The teams in the top 10 in 3-point attempts per field goal attempt have a combined winning percentage of .593 … and those in the bottom 10 have a combined winning percentage of .400.
That’s no accident. Three-point attempts have correlated highly with winning for the past several years.
Nine of the top 10 teams in 3-point attempts per field goal attempt also are above the league average in offensive efficiency, the lone exception being 20th-place Indiana. Similarly, only two teams , Utah and Golden State, have had below-average rates of 3-point attempts and still rank in the league’s upper half in offensive efficiency.”
Hollinger has a point there. Being as how the Thunder are dead last in three point attempts per game in the NBA, and also pretty close to the bottom of the win/loss column as well I decided to put Hollinger’s hypothesis to the test.
Thunder field goal shooting:
fga/gm=81.8
fgm/gm=36.7 (combined 2fgm+3fgm)
Points per shot (PPS) from fg=.947 (combined 2fg+3fg)
NBA league average Points per shot=.998
The Thunder attempt 81.8 field goals per game, which is the combined two pointers and three pointers, less free throws. They make 36.7 of the field goal attempts (44.8%), and score .947 of a point per each shot, which is pretty significantly below the NBA league average. We get fewer total points per shot attempt than the average NBA team.
If we break it down a little further by two point shots and three point shots it gets a little clearer:
2fga/gm=70.2
2fgm/gm=32.6
PPG fm 2fg=65.2
2fg %=46.4%
PP2ps=.928
Above is the two point shots alone. We attempt 70.2 and make 32.6. For each two point shot attempt we score .928 of a point.
3fga/gm=11.7
3fgm/gm=4.1
PPG fm 3fg=12.3
PP3ps=1.05
3fg%=35%
Above are the three point attempts apart from all other shots. We take 11.7 of them (again, the leagues fewest by far) and we make 4.1. For each three point shot attempt we score 1.05 points. We shoot them at a lower percentage than two point shots, but the reward for making them is 50% more points.
We get an extra 12.2% (or 1/8th) of a point from each three point attempt more than we do from each 2 point attempt. That means as regards efficiency, the three point shot for us is a very efficient shot. You would have to be an incredibly terrible three point shooting team to not have an increase in efficiency from the three ball.
Hypothetically, if the Thunder played an imaginary game where we took all of our usual 81.8 field goal attempted as three point attempts (instead of a mixture of twos and threes) we would score 85.89 points off of our shots (excluding free throws of course), instead of our usual 75.9 points from our field goals. Our scoring in this hypothetical game would go up by something close to 10 points per game without being able to shoot them any better than we do currently.
In the real world of course, this can’t happen, but it shines the light on what is a big gaping hole in the Thunder’s offensive makeup: we are dead last in three point attempts per game, and 25th in 3fg%. We don’t take many, and we don’t have a makeup of players that are good at it outside of Durant, Green and Weaver.
If we took more three pointers, our points per shot, points scored per game and our offensive rating would all go up. And since we are near the bottom of the league in offensive efficiency, points per game and the win column, how can that be a bad thing.
Hollinger rightly points out that it is 3 point attempts that are the key factor in the increased scoring (and thus winning percentage), not necessarily 3 point makes. I hope Presti reads Hollinger or this post.
