Determining the NBA’s Greatest Ever – By Win Shares

Photo by Hunter Johnson on Unsplash

The debate on who is the greatest NBA player of all time has been settled for some time. That is to say, the folks at ESPN settled it in favor of Michael Jordan, and a generation of ESPN-fed sports junkies have agreed, largely because no one else dared to offer up much of an argument or analysis (pro or con) except, “Mike was 6 for 6 in championships, and no one else was. The end.” Okay, to be fair, many have offered up Jordan’s impressive stats, his invincible winning spirit, his solid defense, strong all-around play, and the fact that no one in the 90s or early 00s really matched Jordan either by the numbers or through the eye test. But most simply say he’s the best because … shoes.

The eye test–now, that’s an underrated thing. I never played NBA ball, never coached a team, never even tried out for organized team ball at any level. However, I know statistic, I know the game, and I have been a rabid fan for 50 years. I started watching and analyzing the NBA during the third era of NBA basketball, the Wilt Chamberlain Era. It had followed the First Era, George Mikan’s Era, when he and others really invented the pro game and set its standards. After Mikan came the Celtic Era, led by Red Auberbach’s cigars and Bill Russell’s 11 titles in 13 years of playing. I cheered for Wilt then, hated the Celtics because of the stories of racism I’d heard third-hand about its city’s citizens, and picked up the Lakers as “my” team when Wilt ended up there in the NBA’s third era, which, frankly, overlapped Boston’s. The Celtics owned almost all of the titles from the mid-50s through the 60s, but Wilt owned the individual game. His star began to fade in the early 70s, and with it, so did the fast-paced, if mediocre-caliber basketball we’d come to love.

I followed the game through the HORRIBLE, low-scoring 70s, when even playoff games were shown on a tape-delayed basis after the late news. We had to skip the sports segment of the news broadcast just to make the game worth watching. I was still a fan when Magic and Bird were drafted and watched them invigorate what would surely have been a dying league without them. When Jordan was drafted third by the Chicago Bulls in the 1984 NBA draft, I, like most people, never suspected he would one day set the game on his ear. We knew he was better than the overrated Sam Bowie, but no one saw the 90s Bulls coming. By the end of the 20th century, the NBA was on the global stage, and Mike was its first global star. He was the new icon, and if sports networks were to grow, they’d have to help the NBA expand its TV and online following. For that, they needed a roundball messiah, and number 23 was the natural choice. The 1990s were the Jordan Era, and basketball stars have been deemed to be subservient to Jordan ever since. The early 21st century saw the NBA’s obsession with athleticism finally start to wane after a ream of ballers with mad hops who couldn’t shoot or pass for shit. Thus, emerged the LeBron / Pace and Space Era, where the game has shifted from defensive prowess and hero ball to team play with high demand for multi-skilled talent and scorer-facilitators. It’s certainly, to my eyes, the best team basketball ever played, but with all that sharing going on, it sure is hard to tell who your all-time heroes might be, right?

I mean, no one owns the paint like Wilt or Shaq. No one dominates a game like Mike did. Hell, no one is even remotely like Jordan, even Kobe, who modeled his entire game after Mike.

But was Jordan the really the GOAT? Did he deserve the title and adoration?

Photo by Hunter Johnson on Unsplash

It’s hard to argue with 6 rings for 6 attempts, but how about 11 for 13? How about the 1980s and Magic Johnson’s Showtime Era? Those guys won 5 titles to Boston’s 3 and the Bulls won 6 in 8 years, numbers of titles unmatched until Kobe Bryant’s Lakers just missed repeating Jordan’s double three-peat (with the same coach) in the 2000s. Maybe it was Mike, or maybe individual players were overrated and we should have been talking about coaches all along. Perhaps, while we’ve argued over who’s responsible for teams’ wins, we missed the fact that maybe it’s coaching. The 50s had John Kundla. The Celtics had Auerbach. The 70s had guys like Bill Sharman, K.C. Jones, and Red Holzman, but no one really stood out and neither did the basketball (the Lakers, Celtics, and Knicks aside). Showtime had Pat Riley. The 90s and the 00s had Phil Jackson and basketball blossomed again. Since the late 90s we’ve had Gregg Popovich and others who’ve taught us how to pace and space (which is like the 1960s Celtics, except guys actually make their shots). Again, however, no coaches except Pat Riley, Jackson, and Popovich have won more than 2 titles since Red Auerbach did it 50 years ago!

MIAMI – DECEMBER 25: (L-R) Head Coach of the Los Angeles Lakers Phil Jackson and Head Coach of the Miami Heat Pat Riley share a laugh before the game on December 25, 2005 at American Airlines Arena in Miami, Florida. Copyright NBAE 2005 (Photo by Victor Baldizon/NBAE via Getty Images)

What’s the conclusion there? Well, let’s look at the candidates for greatest ever that are widely touted: Michael Jordan, Wilt Chamberlain, Kareem Abdul-Jabbar, Kobe Bryant, Bill Russell, LeBron James, Magic Johnson, maybe Oscar Robertson, by a knowledgeable few. So, what, if anything, do they have in common? Jordan and Bryant won all of their titles with Phil Jackson (11 rings). Kareem and Magic won with Riley (5). Russell won with Auerbach (9 with Red and 2 with himself as coach). Only LeBron, the Big O, and Wilt won their titles without ever playing for one of the top-notch coaches (defined by more than 2 titles). So is it surprising that those latter 3 guys don’t match Jordan and the others for rings? Not really. How can we evaluate the feats of a guy who plays for a winner when the other guy plays for a mediocre chief play caller?

My hypothesis was, after 50 years of watching, that basketball is truly a team game. While the media obsesses over scoring and over high-value metrics, championships are won by matching capabilities with needs and then having offensive and defensive schemes that maximize your talent. Auerbach kept the Celtics running and shooting, knowing that no one could keep Bill Russell off the offensive glass but Chamberlain, and he didn’t have enough help to beat them. Phil knew his version of the triangle, with unstoppable penetrators (Jordan, Bryant, Shaq) kept defenders chasing the ball until they ended up in a one-on-one match-up they had no chance of winning. Riley kept his Lakers running, and no one could stop Magic on the fast break with the finishers he had to choose from. Scheme, talent, team cohesion, health. Championships are made from those lucky enough to have all four things going for them.

The All Day Sports Networks of the day are fed by ex-ballers who were raised to believe that only rings matter, so all talk of GOATs are tainted by ring counts (especially by the cats who have none). But as we’ve shown above, multiple rings have as much to do with coaches and teams than any individual performance. So, for this analysis, we will IGNORE CHAMPIONSHIP RINGS. First, that kind of analysis is dumb as hell, and we aren’t. Second, if we’re going to count rings, then Russell is the GOAT, and 7-8 of his teammates (plus Robert Horry) come ahead of Jordan.

Shut up, ESPN. All multiple titles mean is that you had a great coach within a great organization. If you want to know who was the greatest player, you have to examine which players–game in and game out–scored, passed, defended, rebounded, and created opportunities for himself and his teammates to win. To win a title, you have to win games, and for that, you need teammates, full stop. You can’t make an assist if no one makes the shot. One great scorer will impress the shit out of you, but you won’t win a title unless you have others who can put the ball in the hoop too.

So how do we evaluate who is truly the best? Simple. We need to understand what a player’s contribution is to overall winning. For that, we turn to the metric, Win Shares, originally developed by baseball’s Bill James. Basketball-Reference.com does a good job of explaining Win Shares, so I’ll refer you there instead of detailing it here, but in summary, the idea is this: When a team wins a game, each player gets a share of the win. Offensive points developed are allocated, and marginal offensive ratings evaluate how a player affects a team’s offense. Similarly, a defensive rating is calculated, and when combined, you have win shares. Players, like LeBron James, who figure into a large percentage of a team’s scoring, will have high offensive win shares. A Bill Russell type, cleaning the offensive and defensive boards, will also figure into the Win Shares appropriately. Sadly, there’s little addition for guys who’ll take a charge (unless he’s generating turnovers and keeping his team’s pace up) or dive after a loose ball, but overall, it’s a fair way to allocate the points-differential that leads to wins.

As Basketball-Reference points out, using their data, we have numbers for the NBA players with the most Win Shares (WS) of all time. The all-time leader is Kareem Abdul-Jabbar, unsurprisingly, who generated 273.4 Win Shares during his 20-year career. That’s an average of 17.66 WS per year that Kareem played, meaning his efforts were worth almost 18 of his team’s wins each year.

Now using career numbers has obvious problems. For one, it rewards guys like Kareem who hung around a long time, and punishes guys like George Mikan, who had a relatively short career. By this measure, Michael Jordan is 4th all time, behind Kareem, Wilt, and Karl Malone. As you can see from the orange line in the graph above (click the graph to enlarge) the number of games these men played to reach their career WS varies greatly. Jordan, LeBron and John Stockton are very close in WS, however Stockton took a lot longer to reach his numbers than the other two men.

In order to normalize this, we have to adjust the WS to account for playing time. Here’s where I differ from the creators of the measure. They tout using WS per 48 minutes, similar to how other measures work, in order to see “how many WS would each player create were they playing a full 48-minute game.” Using this measure, Jordan is the GOAT. There’s a HUGE problem with this adjustment, however, in my opinion. Some players actually DID play 48 minutes–at least, Wilt did. In 1960-61 and 1962-63, Wilt averaged seconds under 48 minutes per game for the entire season (79 and 80 games). In 1961-62, he averaged OVER 48 minutes per game over 80 games, never coming out and playing a few games into overtime.

Looking at WS per 48 minutes says, “Let’s pretend that Jordan would play at the same level over 48 minutes as he did over the 40 per game he averaged during his prime.” Maybe he would have or maybe he’d have faded, but Wilt actually did it, and endurance and injury-avoidance have to be one of the criteria you evaluate when looking for the NBA GOAT. So we’ll discard WS per 48 minutes and likewise ignore WS per season, since that ignores the fact that games per season vary with injuries and some players ball for 19, 20 years, and their productivity fades. We need to account for the fact that the GOAT is largely evaluated for his best body of work, not his season coming out of the showers at age 40 in hopes of one last fling.

So, next, I looked at WS per Game. This is per games played, whether the player was on-court for 48 minutes a game or 48 seconds. It normalizes the metric, only looking at a per-game output, but unlike WS per 48, it expects those who play more minutes to earn more WS. Guys you have to drag off the court — Wilt, MJ, LeBron — deserve not to be compared to guys whose little feeties hurt after 30 minutes.

Using this metric, we get a different ranking than the career numbers. First of all, George Mikan is off the charts, but instead of claiming he’s the GOAT, we’ll give him an asterisk. He played only 7 years. He played in a different era, with slower pace, fewer players, and fewer teams. Many rules were different, because Wilt hadn’t come along yet and forced the NBA to rewrite a bunch of them. All we can say for certain is that Mikan was easily the best player of his day, but we can’t compare his to any other era.

Now, before we proceed further, a HUGE caveat here. This analysis DOES NOT account for statistical significance. In layman’s terms, I can’t tell you whether Chamberlain’s 0.237 WS per game is statistically higher than Jordan’s 0.200 or not. All I can tell you is that it’s nominally higher. But since all basketball statistics ignore the fact that a player’s lower scoring average could be within the statistical margin of error of the scoring champ, I’m simply playing by the rules already around. In order to tell you if one number is statistically higher than the other, I’d have to know sample sizes, and I don’t have that data. My gut feel, however, is that Mikan-Chamberlain form one tier, numbers two through five another tier, and so on. But who cares? Where’s the fun in that?

With all of this in mind, we have the GOAT Ranking as 1 – Wilt Chamberlain, 2 – Michael Jordan, 3- LeBron James, and 4 – Chris Paul. Wait, Chris Paul?? That can’t be right. I mean, CP3 hasn’t really had time to play past his peak yet, and besides, he’s got ZERO RINGS! Remember, however, our GOAT status isn’t about rings. LET’S BE CLEAR: MJ is the greatest playoff player of all time, or at least he passes my 50-year eye test (and I never saw Russell play until 1967). But we want to know who was the best guy for an 82-game season, trusting that the best team would win the title even if the best guy wasn’t on it. Still, I worry that CP3 is being evaluated in year 13, while he’s still marginally in his best years, while we’re looking at Kareem well into his 40s. The orange line in the graph above shows the WS per Game for players’ first 13 years, and you can see some players were much better in their prime than later years, Kareem being one of them.

So, after a bit of thought, I came up with my definitive measure, which is still WS per Game, but which only evaluates players in their first 13 years. Why 13? Well, for the most part, I noticed a pattern wherein most Hall of Famers seemed to average around 14 years or so, fading dramatically after their 13th year. Past ballers came out of college at 22, and by 35, were pretty much done. This measure eliminates Jordan’s last 2 years, when he came back for the Wizards after 3 years off. It also doesn’t penalize Kareem and Karl Malone for staying healthy long past their best years. Doing so¬† allows us to fairly compare the old guys to Chris Paul on as apples-to-apples a basis as we can get.

So, what does the data say?

 

Mikan is still out there somewhere, but Wilt is still not far behind. Wilt retired after his 14th year and hadn’t faded much then, scoring less but still rebounding like a maniac. This states that on a per-game, day-in and day-out basis, your GOAT is Wilt Chamberlain, who added more to offensive and defensive Win Shares than anyone else. More importantly, to me, it passes my 50-year Eye Test, as Wilt was the most unstoppable human on a court I’ve ever seen, bar none. Next to him on my measure, and in my mind, was Kareem. His sky hook was unguardable and his defense consistently strong. Yes, Jordan was better one-on-one, but basketball is played 5 on 5. You could do nothing with Kareem. Even Wilt couldn’t stop him, having encountered Kareem when he was past his peak. To my surprise, MJ is 3rd all-time on this measure, with LeBron close behind in 4th place. Given LeBron’s age, and despite the fact that he’s gone up so far in 2017 and not down, it’s very unlikely he’ll ever catch MJ on my WS per Game metric, even though he’ll pass him on the career WS later this year. MJ was a prolific scorer, a good passer, and a staunch defender. He became more of a jump shooter later in his career, but, like Kobe, was never really the pure shooter people think him to be.

Rounding out the top 5 is the Big O, Oscar Robertson, who I’ve always had in my top 5 based on players I’ve seen. I was frankly shocked to see Shaquille O’Neal as number 8, since I always thought him to be a bit over-hyped, but numbers don’t lie. The Admiral, David Robinson, comes in at 6th. My favorite Laker of all time, Magic Johnson, is at 10, just ahead of my 2nd favorite, Jerry West, the Logo. Who’s the best Laker ever? Not Kobe. He didn’t make the Top 25.

There are some current players, besides LeBron, who make the Top 25 WS per Game All-Time list. CP3 is in 7th; he is the Point God, bar only one, the Big O. Kevin Durant shows up in 13th place in his 11th year, and James Harden (22nd) and Steph Curry (24th) also show up in their 9th years of play. However, each of those guys have a few more years of productivity to maintain to stay in their spots (or rise), so it’ll be interesting to see how they do the rest of the way.

One little side-effect of this measure is my All They-Shoulda-Quit-Sooner Team. I looked at the dropoff of WS per Game in years 1-13 versus the end years of careers, and some numbers are startling. LeBron has only fallen off by 0.4% in his 1.4 seasons since his 13th year ended, but some guys fell off fast. Jordan only fell by 5.5% despite 4 years off, but Kareem fell by 18% in his waning years, Vince Carter by 19% so far, Shaq by 15%, Kevin Garnett by 13.6%, and Celtics teammate Paul Pierce and the great Moses Malone by 13% each. Clearly, these players chose the “wily old veteran” role long after their elite production had ended. But keeping in mind how high they once were, they actually only fell to normal good-player status. At millions per year, who could blame them?

A final thought piece, although I don’t know what you could do with it. To my knowledge, there’s no measure for showing which players make other players better, although maybe team WS while that player is on the floor vs. off does it to an extent (or, like, teammates’ WS before and with a LeBron maybe). In any case, I don’t think you can directly measure how much better a guy makes teammates when he’s unselfish. So, while an MJ was perhaps dominant and Wilt was unstoppable, there were huge stretches where they played the hero while teammates just sort of watched. Perhaps if we could factor in the guys who seemed to make other guys better–LeBron, Magic, CP3–maybe this WS ranking would change a bit. It does in my mind, anyway.

So, there you have it. By my analysis, using the Win Shares per Game metric and as a long-term game-watcher, Wilt is your GOAT; Kareem is #2; and MJ is close behind as #3 and the best non-big of all time and the winningest alpha dog of all time. If you are like ESPN analysts and care about 1) rings and 2) whoever was playing when they were kids in the fucking 90s, then feel free to stick with MJ (or Bill Russell if you only care about rings). LeBron has surpassed Wilt as my favorite player of all time, but I think he’s just a hair behind Jordan. If only he’d had better coaching earlier and they could have gotten him to post-up more, I wonder which slot he’d be in now. I guess we’ll just have to speculate.

GOAT

 

Bill Out, for I Never Played the Game Productions.

Note: Rude comments or those left with no underpinning analysis will be ignored and/or deleted. It’s that kind of blog. The metrics presented here are facts. Who is or isn’t the GOAT is solely my opinion. You are welcome to hold a different opinion than mine. I’ll be glad to read yours if it’s based on analysis. Otherwise, meh. If you want sports ranting or name calling, call into talk radio.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s