Lowe 1

Jonathan Lowe

Professor Pompos

ENC 1102

29 March 2024

Advanced Statistics Interpretation

Introduction

To many people, basketball doesn’t have much interpretation besides the main goal for any

player or team; “Put the ball in the basket” or “Stop the other team from scoring.” As a student of

the game, I’ve heard these phrases countless times, and it never fails to irk me. One of the main

reasons I’ve come to love basketball because while the goal of the game is simple, the analysis of

it is as complex as it gets. With complexity comes interpretation, more specifically of the statistics

found within the game.

In basketball, specifically for the National Basketball Association (NBA), the interpretation

of statistics is regarded lowly when compared to the “eye-test” of a player’s observed skillset.

However, this interpretation is undeniable as it is the one thing that people cannot deny, numbers.

Numbers which directly portray a player, or teams totals and percentages. The community of

basketball analytics collectively utilizes these statistics in order to further their understanding of

the game. Analysts within the community utilize a variety of different statistical databases for their

own research as well. These databases include, Basketball Reference, the NBA’s statistical

database, Stat Muse, and other statistical repositories. All of these are online and are easily found

through a Google search making these statistics accessible to anyone within the community. Many
 Lowe 2

members of the basketball analytics community communicate their findings and opinions through

online forums such as Reddit and Quora and even social media spaces such as X (formerly known

as Twitter), Instagram, Discord, etc. Typically for seasoned analysts, they will communicate their

findings at basketball conferences where professional research over these statistics is conducted

and introduced to the community. In other cases, you might find an analyst talking on a personal

podcast or a broadcasting network that covers advanced basketball analytics. This discourse

community is not limited to fans of the game; professional basketball analysts and retired players

also use the same forms of communication and the same databases. Basketball is not just limited

to surface level analytics either (i.e. Field goal percentage (FG%), Three-point percentage (3PT%),

etc.), though, there are advanced statistics as well. These advanced statistics require an advanced

level of expertise in basketball analytics to properly interpret these numbers and translate them

into coherent ideas. Having a deep understanding of the surface statistics allows for proper

interpretation of the more advanced ones.

Now, you might be thinking “how are advanced statistics any different from surface

statistics?” Advanced statistics utilize long formulas that take many surface statistics into account.

An example would be Value Over Replacement Player (VORP), which is described as “A box

score estimate of the points per 100 team possessions that a player contributed above a

replacement-level (-2.0) player, translated to an average team and prorated to an 82-game season.”

(Basketball Reference) To put into better terms, this stat exhibits the amount of points a starter is

producing over a benched player, per 100 team possessions over the entirety of a season. This stat

uses points per game (PPG), percentage of minutes played (%MP), and points per possession of

the average of ANY replacement player on a team at that position. As you can see, the interpretation

of this is far from something as simple as FG% where it is the ratio of made shots to missed shots.
 Lowe 3

Throughout this paper, I will provide you with necessary information and definitions of advanced

statistics not covered yet as the guidance will be needed.

Literature Review

The current conversation is the same as it’s been since advanced statistics were introduced

into basketball, are they even worth discussing? Answering that question is nearly impossible as

there are many cases in which the interpretation of these statistics accurately depicts a team’s or

even just a player’s skillset. For example, University Wire interviewed James Whitford, a former

basketball coach for Ball State University and found that coaches believe advanced statistics to be

a means of improvement within a team. Stancin and Jovic have proven this as well through their

findings with various statistics derived from NBA teams with winning records and teams with

losing records. Players, however, tend to disagree and tend to dismiss these statistics as “useless”

or more accurately stated by Charles Barkley, an NBA Hall-of-fame player, “It’s just some crap

that people who were really smart made up to try to get in the game because they had no talent.”

Whatever the stance may be, none can accurately answer the question. However, as Kerry Dirk

stated in Navigating Genres, “by studying the genres that we find familiar, we can start to see how

specific choices that writers make result in specific actions on the part of readers.” (Dirk 254) and

as Laura Bolin Carroll asserted in Backpacks vs. Briefcases: Steps Toward Rhetorical Analysis,

“Media is one of the most important places where [rhetorical analysis] needs to happen.” (Carroll

46). Through the analysis of the various genres and interpretations of advanced statistics, I plan to

provide insight into the rhetorical choices that members of the basketball analytics community

make based on the statistical data provided by online databases.

Members of the community like James Whitford and Charles Barkley are exactly the

reason why there’s discourse within basketball analytics. These opinions are just as important as
 Lowe 4

the statistics themselves. They represent the effects that advanced statistics have on an analyst’s

opinions and allows for a deeper understanding into the rhetoric placed with these viewpoints.

Why does Charles Barkley loathe advanced statistics as much as he does? As a former player,

maybe through the analysis of his own statistics, he was dissatisfied. Why does James Whitford

lean so deeply into these statistics when searching for methods of improvement in his team? Can

the improvement of his team be attributed to development from these stats? Is it valid to believe

that advanced statistics “reinforce what you see with your eyes as to where you need to increase

your energy and where you need to put time into” (University Wire)? Obtaining answers to these

questions is integral to the goals of the basketball analytics community. To summarize these

questions to encompass the common theme found within them would be as simple as this, how

does the interpretation of advanced statistics influence a member’s opinion on a player/team’s

observed skillset? Additionally, how do opinions derived from advanced statistics affect a

member’s feelings towards the principle of advanced statistics as a whole? To answer these

questions, I analyzed the statistical performances of NBA players over the 2022-2023 regular

season. To conduct this research, I utilized Basketball Reference as my main repository for these

statistics as it is the main database that I am familiar with and the ease of access to its statistics is

unparalleled. In my findings, I will show my own interpretation of these statistics as well as

interviews that I conducted that exhibit other members of the community’s findings based on these

same statistics.

Methods

For my research, I mainly focused on the ways in which one’s own interpretation of

advanced statistics differs from one to the other and the assumptions gathered of advanced

statistics. I have already established two examples of these differing viewpoints from coaches and
 Lowe 5

players; however, there I also spoke of online forums in which I had previously researched prior

to this paper. Though these are not peer reviewed journals or “scholarly” sources, they are members

of the community just as I am. Their interpretations of these statistics are necessary as they serve

as the viewpoint of a fan of basketball. Just as I did with my other sources, I observed the rhetorical

choices made when arguing a point and the initial stance they have on the topic of advanced

statistics. I also analyzed their interpretations of a player’s skillset based on the conclusions derived

from their own research of the genre. Combining these factors together, I was able to provide a

potential answer for my research question. For me to find these sources, I used Google and

searched using keywords and phrases such as “advanced statistics in basketball” and navigating to

the “Perspectives” tab on google which shows results from discussion forums such as Reddit or

Quora. As for the peer-reviewed sources (i.e. Whitford, Stancin and Jovic), I utilized the University

of Central Florida’s (UCF) library scholarly repository to find these articles. I used the same

keywords I used to Google the discussion forums; however, I also narrowed my search down to

just “NBA stats” and was presented with results relating to my research.

Additionally, I conducted research of my own in the form of interviews where I asked

questions relating to topics frequently discussed within the basketball analytics community. These

topics included the interpretation of the advanced statistics of a player considered to be a superstar

in comparison to a player considered to be a role player. I also asked questions to the participants

as to whether advanced statistics are a viable option when creating opinions on players. My focus,

however, was the rhetorical choices made in their interpretations of advanced statistics.

Understanding why the participant formulated the opinion they did with the given statistics and

whether the participant leaned more towards a visual interpretation of a player rather than a strictly

statistical angle. During interviews, I recorded answers through my cell phone’s “Notes” app as
 Lowe 6

these interviews were not performed in a necessarily professional manner (in my living room).

Through this process I have been provided by the subjects of my research both verbal and written

consent for the usage of any answers provided. Also, I plan for their anonymity to be upheld in

case of a breach of personal security. The participants that I have decided to include in my research

are active members of the basketball analytics community themselves both as fans, players, and

coaches of the game.

Advanced Statistics at Ball State

University Wire, better known as “UWire”, is a college newspaper organization that utilizes

various college press releases to develop their seemingly endless online news databases.

Information found within this database ranges from sports to campus organizations to even the

“Top 10 Things to do Near UCF.”

In University Wire’s article, they interviewed James Whitford, the Ball State University’s

head coach from 2013-2022 with the intention of understanding the analysis that Whitford

undergoes after each game. According to the article, Whitford would be emailed the advanced

statistics for his team. These statistics include, True shooting percentage, defensive rating, player

efficiency rating, and many others. However, one statistic stood out for Whitford, defensive points

per possession (DPP). This metric is to be interpreted as it is read, how many points a team allows

per possession defended. Whitford believes that though this statistic is indisputable when

observing whether a team’s defense is good, it doesn’t exhibit where a team might need work.

Additionally, Whitford went on to say that advanced statistics. This claim places Whitford on the

side of the basketball analytics community that believes advanced statistics truly depict how a

player or how a team performs specifically on the defensive side of the game.
 Lowe 7

Whitford’s rhetoric when observing advanced statistics is centralized around both data

interpretation and visual opinion. This is mainly because this is asked of him as a coach for a

professional basketball team. However, Whitford describes advanced statistics as “part science,

part art. The art part is what the coaches’ eyes see and what they believe when they watch. The

science part is the statistics and numbers aspect.” (University Wire) Whitford’s stance serves as a

healthy example of the belief that using advanced statistics as your only mode of analysis will

stifle the true understanding of skillsets of players when viewed collectively as a team.

Reddit Joins the Conversation

In a reddit thread in which a member of the basketball analytics community argued that

win shares (WS) and win shares per forty-eight minutes (WS/48) are poor statistics that do not

accurately depict a player’s contribution to a win. To give some insight into what win shares are,

they are a metric that is calculated from a multitude of a player’s statistics, both defensive and

offensive, to determine how likely they contributed to a win. A positive win share is favorable as

it shows that a player contributes to wins rather than holding their team back from a win. The

writer utilized LeBron James’ and Dirk Nowitzki’s playoff performances from 2011 and James’

and Anthony Davis’s regular season performance from 2020 as examples supporting the times the

WS metric falls short. To summarize this, Davis and James had nearly identical statistics in 2020

yet despite James finishing higher in Most Valuable Player ladders, Davis had the higher WS. As

for 2011, it was the same argument as Nowitzki defeated the team in which LeBron was on, the

Miami Heat, in the NBA finals and yet LeBron still topped Nowitzki’s WS total. In the end, the

writer was brought to the conclusion that the WS were heavily impacted by defensive efforts as

Davis finished second in the voting for the league’s Defensive Player of the Year and James was

notoriously a better defender than Nowitzki despite losing to him in the NBA Finals.
 Lowe 8

At the end of the writer’s response, they asked the question, “Are there advanced metrics

that you think miss the mark in their attempts to further an understanding of the game”

(iCE_P0W3R)? Through this post, the writer was arguing that many advanced metrics fall short

of providing a deeper analysis of basketball and rather confuse members of the community. In the

replies of this post, a common theme was that individual advanced statistics are limited by the fact

that basketball is a team game, specifically in reference to the defensive win shares as not all

defensive statistics can be attributed to just one player. For example, a user by the name of

chickenstew907 stated that “Advanced stats tell a pretty good picture offensively but defensively

they are totally worthless.” Another example of this was from the user bananajunior3000 who

stated, “Measuring defensive impact using at-best imprecise tracking data and no knowledge of

team schemes/responsibilities seems like an impossible thing yet some people treat those as

argument enders… we’re still pretty far from reliable defensive metrics compared to where we are

with offensive stuff.” The same can be said for offensive win shares as team play designs are

created with the intention to allow for easy opportunities to score making isolation and “coast-to-

coast” fastbreak opportunities the only truly individual offensive statistics.

Interviews to Interpretation

I conducted a series of interviews with a few of the members of the basketball analytics

discourse community to find more information regarding my research question. Through these

interviews, I found a trend in the opinions that arise from the analysis of advanced statistics which

is that a lot of specialist’s skillsets are accurately depicted whereas superstars are skewed due to a

variety of factors. The notable factors included a player’s usage rate, their defensive rating, and

the players per 36 minutes statistics. A line that depicted this was when I was interviewing my final

interviewee and they said, “When looked at from a statistical standpoint, [Luke Kennard] would
 Lowe 9

never be described as an all-time great shooter despite being a frontrunner for three-point

efficiency while a guy like [Paul Pierce] is praised when shooting 37% on 4 threes [per game].”

When observing a specialist’s advanced statistic, one thing to take note of is that their overall

player efficiency rating (PER) is rather low as that is a stat that takes all “positive” statistics into

account (points, rebounds, assists, steals, blocks) while their offensive or defensive efficiency

ratings might be much higher as they might specialize in those statistics. An interviewee referenced

Michael Porter Jr. a small forward for the Denver Nuggets, who is known as a “3 and D” forward.

While Porter Jr. had only a 16.8 PER in the 2022-2023 NBA season (highest: Nikola Jokic 31.2),

he had the highest three-point efficiency and percentage for players shooting 6+ three-pointers per

game.

In another case, a participant who previously played Division 1 basketball completely shut

me down when I proposed the idea of advanced statistics. They stated that, “Advanced stats mean

absolutely nothing when it concerns a player’s offensive skillset. Look at Shai Gilgeous-Alexander

(SGA), he’s doesn’t have the highest true shooting [percentage] because he shoots mainly [mid-

range shots].” True shooting percentage (TS%) is described as “A measure of shooting efficiency

that takes into account 2-point field goals, 3-point field goals, and free throws” (Basketball

Reference). This statistic considers 3-point field goals as the most efficient shot that a player can

take as they yield the highest points per possession when consistently made. For a player like SGA,

a player looming around the top of the NBA’s most valuable player for this season, a low true

shooting percentage would typically indicate that he’s less efficient than other players who would

attempt more 3-point field goals. However, when observing SGA’s 2-point field goal percentage,

we find that he is shooting 63% which is 11% over the league’s average. Additionally, when used
 Lowe 10

as a subject to the “eye-test” any experienced member of the basketball community can recognize

his offensive prowess as a one-of-a-kind talent.

The Future of Advanced Statistics

The future of advanced statistics is a difficult thing to picture as the analysis of the game

can only go so far past its current state. The community is already in disagreement as to whether

advanced statistics are credible sources that depict an individual’s skillset in basketball. However,

research has shown that there are more accurate ways to take advanced statistics through the

introduction of new technology. Adria Sanguesa suggested a cost-effective alternative to our

current methods of tracking advanced statistics in basketball which could entirely change how they

are to be interpreted. In this article, Sanguesa and others analyzed notable NBA 3-point shooters

such as Stephen Curry, to find a more accurate way of discerning a player’s “openness” through

various cameras set around the court. Sanguesa suggests that “open/tough shots cannot be

generalized into a single model” (Sanguesa). As an alternative, these trackers will consider the

distance of the closest defender to the shooter creating a whole new statistic that considers three-

point shooters contested shooting rates versus their open shooting rates.

This innovation for advanced statistics could prove offensive advanced statistics to be the

most accurate data to a player’s shooting skill. Sanguesa’s idea, if put into motion, would be able

to discern the great shooters from the good shooters. In high-level basketball, as one would expect,

open jump shots have statistically the highest efficiency rating in modern analytics. However, no

one ever observes the contested jump shot as it is typically written off as a “poor shot.” When

presented with the ability to discern how well a player shoots open jump shots to how they shoot

contested jump shots and the volume that they shoot these shots with would make all the difference

when scouting and creating lineups for a team. Sanguesa’s research could potentially serve as the
 Lowe 11

catalyst for a new era of analytics. One where even a 40% 3-PT percentage can’t be taken at face

value.

Conclusion

Overall, the rhetorical interpretation of advanced statistics differs across various members

of the community however, the one common goal of reaching a better understanding of these

metrics remains. This research serves as integral part of this advancement towards the community’s

analysis of basketball through the understanding of various viewpoint’s interpretations of genre

analysis conducted on their own. A few things to take away from this research would be that

opinions created from these metrics will always be different. In one way or another, they will

always meet the criteria as a “fair statistic” or as a “poor statistic.” Moreover, advanced statistics

are and always will be a highly debated topic as they aren’t always a valid way of interpretation.

However, this is not to say that it cannot be used at all. As shown in many cases throughout my

research, the analysis of advanced statistics and the rhetoric we can make of it is important as the

add to the discussion of the discourse community.

 Lowe 12

Works Cited

A.A. Sangüesa, T. B. Moeslund, C. H. Bahnsen and R. B. Iglesias, "Identifying Basketball Plays

from Sensor Data; Towards a Low-Cost Automatic Extraction of Advanced Statistics,"

2017 IEEE International Conference on Data Mining Workshops (ICDMW), New Orleans,

LA, USA, 2017, pp. 894-901, doi: 10.1109/ICDMW.2017.123.

Carroll, Laura Bolin. “Backpacks vs. Briefcases: Steps Towards Rhetorical Analysis.” Writing

Spaces: Readings on Writing, vol. 1, 2010, pp. 45–58

https://wac.colostate.edu/books/writingspaces1/carroll--backpacks-vs-briefcases.pdf

Dirk, Kerry. “Navigating Genres.” Writing Spaces: Readings on Writing, vol. 1, 2010, pp. 249–

262. https://wac.colostate.edu/books/writingspaces1/dirk--navigating-genres.pdf

iCE_P0W3R. “Is There an Advanced Stat That You Just Don’t Think Is Very Good?” Reddit,

Nov. 2023,

https://www.reddit.com/r/nba/comments/17eq7w6/is_there_an_advanced_stat_that_you_j

ust_dont/. Accessed 12 Mar. 2024.

I. Stancin and A. Jovic, "Analyzing the influence of player tracking statistics on winning basketball

teams," 2018 41st International Convention on Information and Communication

Technology, Electronics and Microelectronics (MIPRO), Opatija, Croatia, 2018, pp. 1533-

1538, doi: 10.23919/MIPRO.2018.8400276.

 Lowe 13

Whitford speaks on advanced statistics in basketball. (2015, Apr 14). University

Wire https://www.proquest.com/wire-feeds/whitford-speaks-on-advanced-statistics-

basketball/docview/1672971716/se-2

“2022-23 NBA Player Stats: Advanced.” Basketball, www.basketball-

reference.com/leagues/NBA_2023_advanced.html. Accessed 26 Mar. 2024.

 Lowe 14

Appendix A:

 Do you believe a superstar’s advanced statistics weigh comparatively to a role player?

 Do you believe advanced statistics are a valid metric to create opinions from?

 How do advanced metrics affect your viewpoint on players in the NBA?

 Do you see advanced metrics being remade in the future?

 What statistics did writer read?

 Is there any mention of favoritism in initial post, username, or replies?

 Is there any mention of distaste in initial post, username, or replies?

 How credible is the author of this post or claim?

 What combination of statistics/eye-test led the author to this conclusion?

 Basketball Advanced Statistics from 2022-2023 NBA Season

 Lowe 15

Appendix B:

 Field Goal percentage is described as the percentage of made shots out of the total shots

taken by a player.

 A field goal is a shot attempt.

 A free throw is a free shot attempt after a violation has been committed.

 A free throw is not a field goal attempt however it is calculated with true shooting

percentage.

 Reddit is an online discussion forum.

 James Whitford is no longer the coach of the Ball State University’s basketball team. His

record during his tenure was 131-148.