Nylon Calculus: Who’s had the best single-game performance thus far?

Nov 9, 2016; Washington, DC, USA; Washington Wizards forward Otto Porter Jr. (22) drives against Boston Celtics guard Avery Bradley (0) during the first half at Verizon Center. Mandatory Credit: Brad Mills-USA TODAY Sports
Nov 9, 2016; Washington, DC, USA; Washington Wizards forward Otto Porter Jr. (22) drives against Boston Celtics guard Avery Bradley (0) during the first half at Verizon Center. Mandatory Credit: Brad Mills-USA TODAY Sports /
facebooktwitterreddit

Though it’s only a month old, the 2016-17 NBA season has already featured its share of memorable individual performances. From Anthony Davis’s 50-point debut to James Harden’s 30-point, 15-assist combinations to Russell Westbrook’s triple-doubles, fans have many single-game heroics to celebrate.

We can, of course, add some structure to this discussion by using Kevin Ferrigan’s Daily RAPM Estimate. The bulk production formula for DRE is designed to help “quickly analyze game to game performance” with basic box-score statistics, giving us a good place to start for our current purposes. Here are the top-10 performances through November 28:

screenshot-2016-11-29-at-5-47-07-am
screenshot-2016-11-29-at-5-47-07-am /

I imagine that Otto Porter’s placement atop the leaderboard is bound to raise some eyebrows, so we’ll go over his career game in due time. But, first, let’s set these numbers in context.

Read More: Shooting and shot creation, what’s more valuable?

From the start of the 2013-14 season through last night’s schedule, we’ve seen approximately 89,000 individual box-score lines — that is, player appearances in regular-season and playoff games. When we look at the DREs for each of these lines, they are distributed in the following way:

dredensity
dredensity /

The data are pretty consistent from year to year: a little over 27,500 lines annually, with the median and mean around 2.7 and 3.5 DRE, respectively.

During this time period, we’ve had 447 player performances of at least 20 DRE. That threshold puts us in truly elite territory, representing less than one percent of individual games. Even if we narrow the pool to games where players log 25 or more minutes (roughly 40,000), we’re still talking about a rather special occurrence. Indeed, a list of players with the most 20-DRE games produces an unsurprising constellation of NBA superstars:

screenshot-2016-11-29-at-5-47-40-am
screenshot-2016-11-29-at-5-47-40-am /

A deeper analysis might present the data here as a percentage of a player’s total games or make other appropriate adjustments. But the point remains that very few have the individual talents, team contexts, and game situations to reach 20 DRE.

Nonetheless, every once in a while, we do find an unlikely player who happens to put it all together on a given night. In March 2014, Jodie Meeks hit 6-of-11 3-pointers, made 14 of 14 free throws and tallied four steals (24.8 DRE). A month later, Randy Foye posted 30 points on a true shooting percentage of 80 with a 21 usage percentage, along with 15 assists (24.8 DRE).  Last November, Robert Covington had 28 points, 7 rebounds, 5 assists, and 8 steals (28.2 DRE).

Porter was a similar case when the Washington Wizards went up against the Boston Celtics a few weeks ago. He scored 34 points on 20 shooting possessions. He grabbed 14 rebounds, split evenly between the offensive and defensive ends. He added four assists, three steals, and three blocks — all without turning over the ball. He impacted the game in practically every conceivable way.

Yet, as noteworthy as Porter’s performance was, it’s intuitively difficult to comprehend how it could surpass Davis’s opening-night masterpiece. In that game, Davis not only scored 50 points on 60 percent true shooting, but he also compiled 15 rebounds, five assists, five steals and four blocks, while committing just three turnovers. How could his game possibly be outranked?

At this stage, I think it’s instructive to remember a couple of important points. First, as with other metrics of its kind, DRE is an “estimate” of overall player production, so we should be cautious about overemphasizing slight differences in results. Had Davis committed one fewer turnover, for instance, this discussion would be moot. Our evaluations shouldn’t be driven blindly by what raw numbers tell us.

Read More: Drives, shooters, and the players who can do both

More importantly, DRE is intended to be a simple and transparent starting point for game analysis, not a conclusive assessment of player performance. As Ferrigan stresses, it is limited in the way that it accounts for defense, spacing and creation. These limitations don’t discredit DRE by any means. We just have to put it in perspective — and that includes drawing upon other sources of information to get a better understanding of the circumstances.

Specifically, despite comparable counting stats in some areas, Davis carried a larger share of his team’s workload than Porter did. Their usage rates (45 percent versus 23 percent) and assist rates (32 percent versus 18 percent) certainly reflected their different levels of responsibility. The same applied to the types of offensive opportunities that they had. While Davis had more defenders around him and created more shots off the dribble, Porter benefited from a healthier dose of open catch-and-shoot options requiring limited time with the ball. Even their rebounds varied. Close to half of Davis’s boards were categorized as contested; Porter hovered around one-fifth.

These details aren’t meant to diminish Porter’s feat. He clearly had a career game, and if there’s anything that DRE and other such metrics help us do, it’s to spotlight accomplishments that might otherwise go unnoticed.

dreporter
dreporter /

At the same time, it’s important to dig deeper into contexts and nuances. All-encompassing metrics and rankings should serve as gateways to further exploration, not as destinations in and of themselves.