No matter how strong the pull of the moment is on our attention, the history of the game calls to us. The urge to crown a king and place him in succession to the kings of the past is an echo of the call. Ranking the top 10 or top 20 or top 50 or top 100 players of all time is another echo. The call rings out clearest of all, however, in the eternally debated question “Who is the greatest player of all time?” The GOAT debate is nothing more and nothing less than our predisposition to measure the immediacy of current players against the historical stature of legends from the past.
No matter how one describes greatness, or measures it, the desire is the same. We all want to know who is and was the absolute best. Ring-counters want to know the truth of the matter. So do points-praisers, eye-testers, and analytics afficionados. Coaches want to know who is the best, and so do fans. Everyone involved in the game has the same need to compare, to evaluate players in relation to one another. It’s a competition, after all! If we want to understand what makes teams win, it is crucial to be able to determine which players have a stronger or weaker influence on winning. The goal is not to field an objectively talented team. The goal is to field a team which can defeat opponents as frequently as possible.
The Eye Test
Through whatever lens we view the game, we cannot view it all in the same moment. Even someone who watched the NBA in the 1950’s and is still watching it today is not able to view today’s game with the same “eyes” with which s/he saw the league in their youth. The peculiarities of the present impress themselves upon us, but are in part washed away by the failings of memory. The statistical record can compensate for this loss of information in many ways, but there are still problems. The statistical record also experiences data loss. Even more importantly, statistics present events from widely varying contexts as though they were identical in nature.
Thus, the eye test fails and statistics fail to reveal to us the GOAT. Valuing only points – one component of the multi-faceted statistical archive – is necessarily limited as well. Ranking players by points (one statistic) can by definition be no more accurate than ranking players by multiple stats; less information leads to lower resolution, and thus lower accuracy. Even counting rings proves insufficient at some point. Eventually, we need to compare players who won championships in a league with 10 teams against a player who won championships in a league with 30 teams.
Measuring Greatness is not as Simple as it Seems
Each measuring stick we try to use breaks down sooner or later when we try to answer the one question we most want to know the answer to: Who is the GOAT? In order to accurately analyze and compare players across the history of the NBA, we will have to go deeper than any of the current methods. To answer the question “Who is the GOAT?”, we must first grapple with the question: How do you measure greatness?
Fans and sportswriters up through the mid-1970’s were able to address the question with a “flat” method. They were able to view each season on the same level playing field. As such, viewers could analyze a player’s contributions across each of those more or less similar campaigns. Things don’t stay the same forever, though. The NBA merged with the ABA, and then other expansion teams joined the league. Pace, or the speed at which teams play, began to change significantly as well. In short order, the differences between seasons and eras became to great to ignore.
The Game Has Changed
(Yes, that’s right. The entire purpose of this subheading was a Kirk Goldsberry joke. You’re welcome)
In point of fact, the differences were significant before that point as well. The NBA of the 1950’s was a bizarre landscape of variable (and shorter) schedules, folding franchises, 90% white players, and low or intermittent pay for players. The truth is that this game was different from the game at the end of the 50’s, when Bill Russell and Wilt Chamberlain and Oscar Robertson burst onto the scene. The 60’s were again substantially different from the game in the early 70’s when the ABA siphoned talent off of the NBA’s pipeline.
Then the NBA added the three-point line … then moved it, then moved it back. Later, teams started shooting three-pointers at much higher rates than in the past. In between, there were changes in rules regarding illegal defense, hand checking on defense, landing space for shooters, and numerous others. These changes shaped the style of the game and the performances of players great and small. To measure greatness, then, is to express a player’s value relative to his own context, but to express that value in a way that does not depend on his context.
Playing in a low-pace environment (like the 1990’s) means that a player’s statistical profile will be artificially smaller, relative to high-pace eras. The average player will have fewer points, fewer rebounds, fewer assists, fewer blocks, steals, free throws, field goals, turnovers, fouls and everything else. We know, however, that a player isn’t more valuable simply because he played during a certain time. Scoring was higher in the 60’s, 70’s, and 80’s than it has been since then, but it does not follow that all the best scorers in NBA history played before 1990.
Evaluating Players Accurately
The definition above has two parts: 1) expressing the player’s value relative to his context, and 2) expressing the player’s value in a manner that does not depend on his context. To accomplish the first part of the task, we must well and truly comprehend a player’s performance within the context of the season in which he recorded that performance. In order to know how good LeBron James was in 2011 or how good Michael Jordan was in 1991, we have to understand how well Michael Jordan performed relative to the NBA in 1991. Flights of fancy transporting 1991 Michael Jordan into today’s NBA, or the NBA of 1971, or any other time are entertaining but subjective. To truly measure a player, we must carefully measure his performance within his own context.
For 1991, this means evaluating a player in a league that was slowing down and relying more on perimeter players other than the point guard than previous eras had. This furthermore means comparing Jordan’s performance with the performances of his peers in that setting and only that setting – a league with 24 teams where player and team salaries were less tightly regulated than they are today, and where the three-point shot was much less common. It is necessary to rigorously analyze and measure his performance relative to the 1991 NBA before comparing him with player-seasons from a 12-team league or a 30-team league, or campaigns where every team in the league depends on post scoring, or campaigns where every lead guard in the league shoots five 3’s a game. These other comparisons are useless if we have failed to quantify the player’s value in his own setting.
The GOAT Ladder
In order to make my method of evaluating players as clear as possible, I have chosen to present the introduction in a series of short posts. The next post will outline my determination of offensive greatness, then the post following it will explain my process for measuring defensive greatness. The take a look at … well, the different ways to look at the stats I’m presenting. After that will come a brief explanation of the format in which I will be presenting each player’s case. After that, we’ll start the countdown to number 1 – the greatest of all time. To navigate these sections of the intro, there will be blue navigation buttons at the bottom of each post to let you move forward and backward. The entire intro series is published now, so don’t wait – you can move forward and read the other sections now!