Pelton mail: Do NBA players shoot 3s better now, or not?

This week’s mailbag features your questions on the home teams in Southern California, length of games, and more.

You can tweet your questions using the hashtag #peltonmailbag or email them to peltonmailbag@gmail.com.


“Are NBA players not getting better at 3-point shooting despite the emphasis on shooting 3s?”

– Kevin Pelton

Time for the occasional question that pops into my own head, in this case while I was reading Michael Grange’s fascinating piece on the Noalytics software the Toronto Raptors are using to track — and hopefully improve — players’ accuracy during shooting sessions at the team’s practice facility.

Grange wonders how much all the shooting players do without feedback is really helping, noting that the league’s 3-point percentage hasn’t gone up much since the league moved the line back to its current distance in 1997-98. (I’ve noted that other basketball leagues have also settled in at similar percentages from 3-point range.)

I think the issue here is, as teams hunt for more 3s, that has inevitably meant attempting more difficult ones. For example, over the four full seasons for which we have tracking data, the percentage of 3s taken off the dribble (pull-ups, in the parlance of NBA.com/Stats) has increased from 23.5 percent of attempts to 25.0 percent. Since pull-up 3s are much lower percentage (32.3 percent last season) than catch-and-shoots (36.9 percent), that will serve to decrease accuracy relative to actual improvement.

Consider also how many more players have the green light to shoot 3s now. In 1997-98, 109 NBA players attempted at least 100 3-pointers. Last season, that number was 232 — nearly half the players who appeared in a game (486). If we somehow had the ability to give players from different eras the exact same quality of shots in terms of defensive coverage, catch-and-shoot vs. off-the-dribble and the ability of the shooter, I suspect we’d see a dramatically greater improvement in 3-point percentage that is apparent looking only at league average.


This is a really fascinating question because it gets at the thorny one of what creates home-court (or field, depending on the sport) advantage. Among the major American sports, the LA Clippers and Los Angeles Lakers are the lone two teams who both share a home field and play each other on a regular basis. (The NFL’s New York Giants and New York Jets, like all teams in opposite conferences, play each other once every four years in the regular season.)

Since the two teams moved into Staples Center in 1999, they’ve met 72 times. When the Lakers are the home team, they’ve gone 23-13 (.639) with a plus-2.6 point differential. When the Clippers are the home team, they’ve gone 19-17 (.528) with a plus-1.3 point differential. With travel, sightlines and other factors constant, that implies the home crowd (and I suppose the home jerseys and the actual court) are worth almost four points per game — about two-thirds of the league-wide home-court advantage (6.3 points per game) over that span.

While that could be attributed to the motivation provided by home fans, it’s also consistent with what Tobias Moskowitz and Jon Wilner argued in their book “Scorecasting,” that home advantage is mostly a product of how referees react to home crowds.


“The NBA made it very clear over the summer that they wanted to increase the flow of the game and make it go faster to get fans attention and tune into the games, rather than tune out. I’ve noticed a lot of games will end early, whether it’s a blowout or close game Like if a game starts at 8 p.m. and it’s a close one down the stretch, it ends right on time (10:30 p.m. or earlier). Can you compare time of game from past seasons compared to this season? Is there a major difference?”

— Deven Parikh

The website Inpredictable.com tracks game time, and creator Michael Beuoy put together a chart showing the average length of games by year excluding overtime games and other outliers:

While shorter games weren’t necessarily the NBA’s intent (their explanation, as Deven notes, focused more on the flow of games than their overall length), I too have noticed watching games that they seem to be ending earlier. So I’m a little surprised the decline in game time from last season isn’t steeper — less than half a minute, on average, though the average time is way down from the mid-2000s.

Beuoy further notes that second and fourth quarters have been faster — which makes sense given the removal of a mandatory timeout in those periods — but that the first and third quarters have been slower on average. It will be worth keeping an eye on the average length of games as the season goes on and the sample gets more robust.


I suspect the most important impact would be what you mention: Teams won’t want to trade unprotected first-round picks because of the increased possibility they could win the lottery without being one of the league’s worst teams. No scenario comes back to haunt a front office more than giving up what turns out to be a top-three (top-four going forward) pick via trade.

Of course, the example of the multiple lottery picks the Boston Celtics got from the Brooklyn Nets has already caused teams to move toward trading their own first-round picks, though with protection. As far as I can tell, there has been only one trade in all of 2017 in which a team gave up its own first-round pick completely unprotected (so not counting the Celtics trading the Nets’ pick in the Kyrie Irving deal), and that was by the Houston Rockets, who already knew their 2017 first-round pick wouldn’t end up in the lottery.

So I’m not sure we’ll see a dramatic difference in trade value because of lottery reform.


No, even though relegation and promotion have been quite effective in European sports, I dislike the way the threat of relegation forces bad teams to adopt a short-term mindset that leads to frequent midseason coaching changes and younger players getting benched in favor of more reliable veterans.

Moreover, hypothetically, moving toward promotion and relegation would force the NBA to completely rethink player development after making progress in turning the G League into a true minor league. So I think it would severely hamper the NBA’s ability to develop young players, and I don’t think the increased competitiveness of games late in the regular season is enough to offset that.

Leave a Response