Pelton mail: Why so many triple-doubles in today's NBA?

ByKevin Pelton ESPN logo
Sunday, November 19, 2017

This week's mailbag features your questions on the chances of a 200-point game, rookies leading their teams in field goal attempts, and more.

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

Yes, they are. Last season, there were an average of 3.9 triple-doubles per team, far and away the highest average since the ABA-NBA merger. (There were an incredible 7.2 per team in 1961-62, when Oscar Robertson averaged a triple-double in a nine-team league.) The 2015-16 average of 2.5 triple-doubles per team ranked fourth since the merger, and the current average of 2.3 per team (through Thursday) would rank sixth.

As I noted in the spring of 2016 when triple-doubles first started to surge, a faster pace should lead to more triple-doubles. (By the way, just ignore the part of that article where I suggest Giannis Antetokounmpo is more likely to average a triple-double than Russell Westbrook.) Here's an updated version of the chart in that piece comparing triple-doubles per 82 games and pace by year since 1983-84.

The relationship between pace of play and triple-doubles is evident here. At the same time, we've still seen more triple-doubles than you'd expect given the current league pace, particularly in 2016-17. So something more is at play -- especially when you factor in that players don't typically log as heavy minutes now as they did in the 1980s and '90s.

To quantify it, let's consider the rebound and assist components of the versatility index I used to discuss Ben Simmons' versatility earlier this week. (We can exclude points because double-figure points typically follow when a player has double-figure rebounds and assists.) A geometric mean of eight rebounds and assists per 100 plays is a pretty good proxy for players capable of recording triple-doubles on a regular basis. And there have been far more such players the past few years than at any point since the NBA added the 3-point line in 1979-80.

In fact, the 22 players who had a geometric mean of at least eight rebounds and assists per 100 plays in 2016-17 were more than from 1982-83 through 1986-87 combined (20). The number is down a little this season, with 19 players qualifying in at least 100 minutes (I used 500 as a cutoff for full seasons), but an influx of skilled big men and athletic guards seem as responsible for the increase in triple-doubles as faster pace.

No, I don't think that passes my "any chance" rule. After all, the Houston Rockets scored 90 in a half, the most in a first half in more than two decades -- and still didn't even get three-quarters of the way to 200. Part of the problem, of course, is that you'd have to find an opponent capable of scoring enough to keep the game close because teams tend to slow their scoring pace considerably in garbage time. Even with that, you're probably going to need multiple overtimes.

From a mathematical perspective, if we take a look at the Golden State Warriors, 200 points is still 5.5 standard deviations away from their league-leading average of 117.5 points per game. For a normally distributed variable -- and despite the possibility of overtime, NBA scores closely follow the normal distribution -- we'd expect to see a performance 5.5 standard deviations better than average about once in every 5.8 million games or so, or about once in every 700,000 82-game seasons. So a 200-point game remains a huge long shot.

The rookie Donovan Mitchell is leading the Utah Jazz in shot attempts per game. Mitchell does not seem like an answer for Utah's offensive issues. How have teams with a rookie leading them in shot attempts performed in the past?

-- Jorge Cantú

I don't know if there's a better answer for the Jazz offensively, which is sort of the point. Teams that are forced to give a rookie such a large role are probably doomed whether they do so or not. Let's see whether that's actually the case.

Since the ABA-NBA merger, 41 rookies have led their teams in field goal attempts per game (minimum 20 games). It has happened just three times since 2009-10: Emmanuel Mudiay and Jahlil Okafor in 2015-16, and Joel Embiid last season.

Predictably, these teams have not been very good. Just six of the 41 have gone .500 or better; nearly half (18) won fewer than 30 percent of their games, and they averaged a .337 winning percentage -- equivalent to 28 wins over a full season. Specifically, they've been dreadful offensively, averaging 3.3 points per 100 possessions worse than league average. But there have been a few exceptions. Larry Bird turned the Boston Celtics into immediate contenders with a quality offense in 1979-80, and Tim Duncan led a deep San Antonio Spurs team as the No. 1 pick in 1997-98.

Perhaps the best comparison for this season's Jazz were the 2009-10 Milwaukee Bucks, whose leader in shot attempts per game was rookie point guard Brandon Jennings. Those Bucks ranked 23rd in the league in offensive rating, but rode the No. 3 defense to 46 wins and a playoff spot. Utah would love to do the same.

Through Wednesday night, per play-by-play data on Basketball-Reference.com, players had made about 80 percent of their free throws (80.3 percent, to be exact). That's better than the league average of 76.9 percent (which is slightly down from last season's 77.2 percent mark, the best free throw percentage in a season the NBA has ever seen), but that's not the right comparison since teams get to choose who shoots their technical fouls.

Weighted by attempts, the players who have attempted technical free throws so far this season have made 84.7 percent of their other foul shots. Some of this can be explained by the fact that players typically shoot better on their second attempt in a given trip to the free throw line, by a difference of about 4 to 5 percentage points, according to Matt Femrite's research on Nylon Calculus. Based on that, we'd expect players to shoot about 2 to 2.5 percentage points worse on technical fouls. (The difference should be about half the gap from the first attempt to the second, since about half of all free throws are the second or third attempts.)

The greater difference we've seen for technical foul shooters this season (4.4 percentage points) is probably partially noise, but we also saw a larger gap than expected in a study of technical free throws by Roland Beech (3.1 percentage points), so I think it may have something to do with the unusual nature of shooting a free throw without any other players on the line.

Fouls drawn are available on NBA.com/Stats, and it turns out that those drawing the most fouls in the league right now on a per-game basis are not generally guards and wings but big men, led by New Orleans Pelicans teammates DeMarcus Cousins and Anthony Davis. In fact, James Harden and John Wall are the lone guards in the top 10, though several more rank just outside it.

If we break down all players by position, centers are also first in fouls drawn per 100 plays:

This breakdown is similar to what Femrite found when he studied this topic too on Nylon Calculus during the 2015-16 season.

Copyright © 2024 ESPN Internet Ventures. All rights reserved.