The 3-Point Shot Epidemic

[Vermont Green]

As a follow up to this great, detailed analysis, I tried to see if there was any correlation between 3PA rank and opponent Fast Break Points (OFBP) rank.  After a bunch of copying and pasting stats into a spreadsheet, and sorting data, I found that there is no correlation at all.  The top 10 teams for most 3PA have an average ranking of 12.9 for OFBP, so actually better than average.  The middle 10 average rank is 17.8, the middle give up the most OFBP, and the lower third rank is average 15.8, so pretty close to average.

I know this is probably hard to follow but there is no correlation.  I thought with more missed shots (you are going to have more missed shots with more 3PA) and longer rebounds, that you would get more "back at 'cha" fast breaks.  But maybe this is counteracted by having more players out from the basket and it is easier for them to get back on D.

I also checked the correlation for Opponents points off turnovers vs. 3PA, again, nothing.  I thought maybe teams taking more 3s would give up fewer points off turnovers, but no, this is not the case.  There is no correlation at all, in either direction.

[Roy H.]

The one, fairly obvious, correlation with shooting a lot of threes:

Top 10 teams in 3PAs/100 possessions.  League ranks for FTAs/100 possessions in parentheses.

1. Celtics (15)
2. Hornets (28)
3. Bulls (30)
4. Warriors (23)
5.  Twolves (26)
6. Nets (16)
7. Cavaliers (24)
8. Spurs (12)
9. Heat (6)
10. Thunder (29)

It's obviously not a perfect correlation.  The Heat, for instance, are the rare team that gets a ton of FTAs and also shoots a lot of threes.  But, six of the eight worst teams at FTAs are in the top-10 in 3PAs.

[Kernewek]

The raw numbers are important here, too, for context: the Celtics shoot a league-leading 50 threes and shoot from the line 21 times per game, while Denver shoots a league-leading 25 free throws per game while attempting 30 threes per game.

Another wrinkle is that the Grizzlies (who have overtaken Miami in 3PA per game as of this morning) shoot effectively the same number of FTs as Denver (24.6 free throws per game) and shoot nearly 39 attempts from deep.

OR:

Rounding for simplicity, the range across the whole NBA is 19-25 FTAs per game. So the worst team in the league at FTAs (Chicago) is leaving 8 points on the table, if you like (since they make just about 80% of their FTs)

Chicago is currently shooting 44 threes per game at a 37% clip, the league average is 37 threes per game - if we assume no one ever gets fouled on three point shots (for simplicity) and they drive inside on seven more possessions (they're already one of the top 2PT% shooting teams in the league, which will drop as they shoot more from two) trying to draw more fouls, they're very likely going to be scoring fewer points than if they continued to shoot threes.

[Redz]

I get it.  The math works.  Especially on open 3?s.  I just don?t think it?s as enjoyable a product when the 3 point shot isn?t any special.  Do benches even stand up in anticipation when a guy throws up a 3 any more?  They?d be exhausted if they do. 

Same thing with baseball metrics dictating that swinging for the fences and striking out a ton is worth the outcome of a more frequent home run.

The math might work, but the game suffers. 

I?ll get excited for a Celtics game still every time they play and hope they win the way they play just as much as any other era.  The objective hasn?t changed: outscore the opponent. 

[Kernewek]

Yeah this is the root problem - basketball is 'solved' to such a degree at the pro level that you're always going to see unflattering ways of getting the best shots, whether that's the elimination of the mid-range game or Mark Jackson backing someone down for 13 seconds.

It's incumbent on the NBA to change the rules if they want to change the way the game is played, especially now that teams are universally so much more advanced at finding the optimal way to play than in years prior.

[Neurotic Guy]

Yet with all the threes in last night?s win v Minnesota, I enjoyed it.  One of the better games to watch so far this year.  Aside from the first quarter, both teams played defense, there were outstanding performances, teams shot well, it came down to a tense final shot, and of course, the Cs won which always makes a game more entertaining in retrospect.

[Vermont Green]

The one, fairly obvious, correlation with shooting a lot of threes:

Top 10 teams in 3PAs/100 possessions.  League ranks for FTAs/100 possessions in parentheses.

1. Celtics (15)
2. Hornets (28)
3. Bulls (30)
4. Warriors (23)
5.  Twolves (26)
6. Nets (16)
7. Cavaliers (24)
8. Spurs (12)
9. Heat (6)
10. Thunder (29)

It's obviously not a perfect correlation.  The Heat, for instance, are the rare team that gets a ton of FTAs and also shoots a lot of threes.  But, six of the eight worst teams at FTAs are in the top-10 in 3PAs.

This is great, and makes sense.  Of course you get fouled more when you  take it to the hoop over jacking up 3s.  Below is the top 10 for points per FGA.  This is a simplified efficiency that takes into account 2P%, 3P%, and fouls.  I didn't check the correlation to 3PA but what you see is that these are in general all good teams.  There are different ways to get a high pts/FGA value, more 3s would help, more FTA would help, just good shooting of course helps.

Top 10
Cleveland Cavaliers      1.37
Denver Nuggets           1.35
New York Knicks           1.35
Dallas Mavericks           1.33
Memphis Grizzlies         1.32
Phoenix Suns               1.31
Indiana Pacers             1.31
Boston Celtics              1.31
Milwaukee Bucks          1.31
Los Angeles Lakers       1.30

Bottom 10
Philadelphia 76ers        1.26
OKC Thunder               1.25
Orlando Magic              1.24
Toronto Raptors            1.23
GS Warriors                 1.22
Portland Trail Blazers    1.21
Houston Rockets          1.20
Washington Wizards     1.20
New Orleans Pelicans    1.19
Charlotte Hornets         1.18

A couple of surprises in the bottom 10 (OKC, HOU) but a pretty solid trend.  I am curious now to see how things line up in terms of opponents pts/FGA.

[Csfan1984]

I'm for eliminating the whole flat side line three/corner three. If the three was the same distance every where you can shoot it from, that might actually improve three points shooting fg%. Make the semi circle stay all the way to out of bounds. At same time it will make the three easier to defend and force more inside out play making to get open threes. That's a win.

[bdm860]

There was a similar thread a couple of weeks ago where I left similar thoughts as I'm about to leave here , that being said:

I think the blame on the "3-point" shot is way overblown (echoing jpotter's ratings scapegoat comment last page). If we look at the league historically, FG% is similar to where it's always been (the last few years have actually been the most accurate of the last 30). So from a fan enjoyment perspective, does it really matter where shots are taken from if they're still being hit at the same general rate and taken with the same volume? Teams are attempting 88-90 shots per game, and hitting 41-42 of them. About the same or better than any time in the last 40 years.

Here's a graph showing FG% and FGA by year:


If you listen to the complaints, you'd think it's all guys running down and taking above the break 3's with no offensive set up, but overall APG has been trending towards the highest point ever (and turnovers trending to the lowest point ever since it's been tracked):


If it's not 3's, the other boogeyman is Fouls/Free throws (players flopping, hunting for free throws, refs calling too many fouls) but those are also trending to the lowest point ever:


The rise in 3 point shooting has also opened the floor up to more dunks (that's good for entertainment/viewership, right?). Basketball-reference has dunk stats back to '97 if you dig for it. My math has teams averaging 3.5 dunks per game from '97 to '17, which has now increased to 4.7 dunks per game currently.

So teams are shooting just as accurately if not better then as they have in the past while taking the same amount of shots, but scoring more points, dunking more, assisting more, turning the ball over less, while also fouling less and shooting fewer free throws. And this is a problem?

Now I'd agree, more variety in style of play would be nice, but at the same time there's still a lot of variety. Look at the best players in the NBA: Jokic, Embiid, Giannis, Tatum, SGA, Doncic, Brunson, Ant, etc. That's a lot of variety in position/styles. For a lack of variety, I think of years like '94 and '95, when 4 of the top 5 MVP candidates were Centers (Hakeem, Ewing, Robinson, Shaq), and the other 2 were PFs (Barkley, Malone).

Anecdotally, highlights seem to be at peak level too, the dunks, the passes, the buzzer beaters (all in HD, online seconds after they happen) are as good or better than the highlights from any other era.

I don't think the rise in 3 point shooting is a problem at all. It's the low hanging fruit when looking for a scapegoat on "ratings".

[mobilija]

There was a similar thread a couple of weeks ago where I left similar thoughts as I'm about to leave here , that being said:

I think the blame on the "3-point" shot is way overblown (echoing jpotter's ratings scapegoat comment last page). If we look at the league historically, FG% is similar to where it's always been (the last few years have actually been the most accurate of the last 30). So from a fan enjoyment perspective, does it really matter where shots are taken from if they're still being hit at the same general rate and taken with the same volume? Teams are attempting 88-90 shots per game, and hitting 41-42 of them. About the same or better than any time in the last 40 years.

Here's a graph showing FG% and FGA by year:


If you listen to the complaints, you'd think it's all guys running down and taking above the break 3's with no offensive set up, but overall APG has been trending towards the highest point ever (and turnovers trending to the lowest point ever since it's been tracked):


If it's not 3's, the other boogeyman is Fouls/Free throws (players flopping, hunting for free throws, refs calling too many fouls) but those are also trending to the lowest point ever:


The rise in 3 point shooting has also opened the floor up to more dunks (that's good for entertainment/viewership, right?). Basketball-reference has dunk stats back to '97 if you dig for it. My math has teams averaging 3.5 dunks per game from '97 to '17, which has now increased to 4.7 dunks per game currently.

So teams are shooting just as accurately if not better then as they have in the past while taking the same amount of shots, but scoring more points, dunking more, assisting more, turning the ball over less, while also fouling less and shooting fewer free throws. And this is a problem?

Now I'd agree, more variety in style of play would be nice, but at the same time there's still a lot of variety. Look at the best players in the NBA: Jokic, Embiid, Giannis, Tatum, SGA, Doncic, Brunson, Ant, etc. That's a lot of variety in position/styles. For a lack of variety, I think of years like '94 and '95, when 4 of the top 5 MVP candidates were Centers (Hakeem, Ewing, Robinson, Shaq), and the other 2 were PFs (Barkley, Malone).

Anecdotally, highlights seem to be at peak level too, the dunks, the passes, the buzzer beaters (all in HD, online seconds after they happen) are as good or better than the highlights from any other era.

I don't think the rise in 3 point shooting is a problem at all. It's the low hanging fruit when looking for a scapegoat on "ratings".