Backpicks GOAT: #4 Bill Russell

Key Stats and Trends

  • Spearheaded the most dominant defensive dynasty in NBA history
  • Shut down the best centers of the time based on available box scores
  • Mediocre scorer but vision and rebounding made his offense passable

Scouting Report

In many ways, Bill Russell was the first athletic “freak” in NBA history. He was an Olympic-level leaper, standing 6-foot-11 in shoes and playing in the 230s.1 Yet he was fast and agile, able to lead the break as a center:

His defining athletic attribute was explosive leaping, reaching shot-blocking altitudes in the blink of an eye. Here’s a 35-year old Russell, in one of the last sequences of his career, inhaling an overmatched Mel Counts drive.

His defensive IQ was probably his most valuable asset. In his biography, he recalls practicing different pick-and-roll coverages with dance partner KC Jones, rehearsing the all-important pas de deux long before the league popularized coverages like “ice.” He also grasped efficiency decades before Dean Oliver formalized it, (correctly) claiming that his team would benefit mightily if Wilt scored a lot but was forced to eat up extra possessions to do so. Russell was aware of how disruptive he could be without even blocking a shot, describing his psychological warfare like this once:

“The thing I got to do is make my man think I’m gonna block every shot he takes…Say I block a shot on you. The next time you’re gonna shoot, I know I can’t block it, but I act exactly the same way as before, I make exactly the same moves. I’m confident. I’m not thinking any more but I got you thinking…I don’t even have to try to block it.”

In the clip below, Russell’s mere presence forces Oscar Robertson — one of the greatest offensive players in history — into an off-balance flip shot:

In the limited film we have of Russell, he demonstrated an intuitive feel for when to help, and how to roam and recover (something he dubbed his “horizontal game”):

These were modern, efficiency-crushing actions, similar to what Kevin Garnett excelled at decades later. However, without a 3-point line to worry about, Russell was content to let marginal shooters fire away from midrange. If he had a defensive blemish, it was resting off these players too frequently, waiting for them to make one or two jumpers before extending his coverage:

Conservation of energy was a strategic play then — Russ logged nearly 45 minutes a night — so these occasional hiccups traded quality for quantity. On most possessions, he engaged, and his timing made him nightmarishly effective. Here he is inviting a player into the lane so he can blast off at the last second for a block:

And his quick leaping regularly led to plays like the ones below:

These challenges were hyper-efficient, rarely drawing free throws or surrendering points. (Russell posted some of the lowest per-possession fouling rates in NBA history.) In perhaps the prototypical Russ play, he stays near his own man as long as need be before springing to help at the last second, controlling the block and starting the Celtics’ break:

Here’s what former teammate Bill Sharman said about him in the 1967 documentary “Year to Remember:”

“Russell, who is a little quicker than [Wilt Chamberlain or Nate Thurmond], will go to the corners, block a shot or get back underneath and get the big rebound or again pick up the cutter.”

He could cover just about any opponent in front of him; here he is switching on to Oscar and then West, completely disrupting both:

Oh, and in case you were wondering what his man defense was like on the block, there’s the occasional logic-defying play like this:

In the publicly available games, the number of blocked shots is staggering. Based on about 100 unofficial box scores (journalists would sometimes track them), Russell tallied around 8 blocks per game during his career, and was closer to 9 per contest during his peak seasons. Adjusting for pace, this yields about 5.5-6.0 blocks per 100 possessions, which would fall between the 30th and 60th-best seasons on record.2

Whatever the number was, it’s clear that his combination of awareness and athleticism was like a cheat code for the era, introducing verticality into a previously grounded game.3 His rebounding was also top-notch, and the best estimates of his glasswork put him on the edge of the top-100 seasons historically, with rebounding rates around 20 percent. He wasn’t into powerful box outs, instead, counting on timing and a nose for the ball to snatch it off the rim:

On the other end, Russell’s half-court attack was fairly straightforward, consisting of a little face-up shot from midrange, right and left-handed hooks and the occasional drive. Here’s a sample:

These shots weren’t too accurate, as evidenced by his slightly above-average field goal percentage — between 43 and 47 percent for the heart of his career.4 He often served as a passing hub, looking for cutters or open men to hit from the post. These would lead to a number of “Rondo Assists” — more vanilla passes that find basic openings — instead of warping the defense with his threat to score. The play below is a typical Russell post-up that ends in a marginal shot:

Sometimes, he would turn that into a hook or shot attempt, but he was often surveying for slashers and rarely ate up possessions trying to force his own scoring. Overall, he looks like a good, but not great passer, and could occasionally play-make for teammates. In the clip below, he senses a double-team and slips it out to the open shooter:

He also completed nifty passes off his own shot action — the dishes below hint at vision and awareness that were good enough to throw high-level dimes. In the second clip, despite briefly dropping his head to secure his handle, Russell is still able to map the court and find the backdoor.

His assist rates were regularly at the top of the league among big men, and in his final five years he finished in the 60th percentile or better among all players in each season.

That generational athleticism also opened up transition opportunities and offensive rebounding chances. It made him a solid finisher as well; here he is with a modern-looking attack as the roll man in the pick-and-roll:

But pick-and-roll action like that wasn’t used as frequently in those days. Instead, Russell tallied a decent share of second-chance points, like this:

And he was extremely nimble running the floor. Here he is demonstrating that speed on the break:

In the latter stages of his career, he played more “point center” and looked for his own offense even less. He was a quick outlet passer, and, as shown above, could grab boards and even run the break himself. In his final few years, his assists increased to around 4 per 75 possessions while his scoring declined from a peak of 12 points per 75 at +3 percent efficiency — slightly below Draymond Green levels — to just under 10 per 75. This kind of low-usage approach won’t lift teams short on scoring, but it would complement perimeter-based offenses.

Russell was durable throughout his career, even gutting through a broken bone in his foot for the final two months of the 1966 season.5 He was slowed slightly by arthritic knees down the stretch of his career before retiring in 1969.

Impact Evaluation

There’s a lack of granular data on Russell, which makes it hard to ballpark his defensive impact. Was it worth 5 points a game (MVP-worthy) or something unheard of like 7 points (GOAT-season worthy)? It was clearly immense, and combined with his passable offense, left a considerable impact footprint.

He didn’t miss much time in his career, so WOWY numbers are hard to come by. Journalists and teammates always claimed that the Celtics fell apart without him; Boston was a 35-win team (-1.9 SRS) in 28 games he missed from 1958-69, and for the other 915 games of his career they played at a 59-win pace (6.4 SRS). This is a tiny piece of evidence – the years are spread out, teams change, and so on — but it echoes the same story as Russell’s other value signals.

For instance, when his teammates missed time, Boston rarely missed a beat. In 1958, Bob Cousy sat for seven games and the Celtics played far better without him. In ’59 and ’60, Sharman, Cousy and Tom Heinsohn missed a few games each, and the machine kept on ticking. In ’61, Sharman missed 18 games and the Celtics were (again) better without him. In ’62, Cousy missed five and, yes, the Celtics were better without him (portending his retirement years).6

But Russell missed four games in 1962 and Boston’s differential fell by 22 points. Four games is infinitesimally small, but all of these stories point in the same direction. It was only when Russell was hampered by injury (in the 1958 Finals) that the Celtics fell short of a title — the single time a Russell team failed to win in a 12-year span dating back to college.7

This trend would hold throughout most of Russell’s career. In ’66, Sam Jones missed eight games and Boston’s performance didn’t budge. Jones missed 11 more contests in ’69 and the team was about 2 points worse without him. All told, as the roster cycled around Russell, his impact seemed to remain. A more detailed calculation of his game-level value has Russell at the top of the impact-heap in his era, while similar studies have him behind only Jerry West and Oscar Robertson (who both had the fortune of playing on dominant teams during the most watered-down years in NBA history).

At the height of their dynasty, the Celtics were comically dominant. From 1962-65, their average margin-of-victory (MOV) was over 8 points per game. During the same time span, only two other teams even eclipsed 4 points per game – the ’64 Royals and the ’64 Warriors. And all of Boston’s separation was created by its historic defense, anchored by Russell:

Russell didn’t join the team until partway through his rookie season, and before hopping aboard, Boston looked like an improved club (playing at a 58-win pace for 19 full-strength games). Still, what transpired in the ensuing years cannot be attributed to teammates or a defense-first strategy.

Boston platooned different players around Russell while he anchored the greatest defensive dynasty in NBA history. At its height (1960-1966), Russell played 43 to 45 minutes per game while only Sam Jones topped 35 per game (once, in 1965). During the 1963 season, no other Celtic played over 31 minutes per contest. To put Boston’s defensive dominance into perspective, let’s zoom out and revisit the above graph, but this time using all defensive seasons since 1955:

As defensive stoppers ramped up their minutes in the ’60s and Russell evolved, Boston lapped the league. 1964 and ’65 were the two best defensive teams ever by this measure. (Amazingly, in Second Wind, Russell calls out the 1964 team as the best defensive team of his time without knowing any of the efficiency metrics.) He captained four of the top-five and five of the top-10 relative defensive seasons in history.

Despite a smattering of famous names, the offenses were never anything to write home about. In ’55 and ’56, Cousy, the sharp-shooting Sharman and the hyper-efficient Easy Ed Macauley powered attacks that were 2 to 3 points ahead of the league. With Russell in for Macauley in ’57, Boston’s offense dipped to around average, where it would hover until 1960.8 In ’61, Sharman trailed off in his final season, Cousy slowed further and defensive notables like KC Jones and Tom Sanders saw more time. During the heart of the ’60s, the Celtics finished about 3 points worse than the league in offensive efficiency based on our best estimations.

Yet Boston was viewed as a squad of offensive stars because they played at a high pace and scored a lot of points at a time when raw scoring was emphasized. Russell was certainly flanked by some offensive talent in the early years (Cousy and Sharman) and in the later years (Sam Jones and John Havlicek). Even his own offense was relevant; for instance, in 1962, he led the team in postseason scoring and efficiency and finished second in assists. But Boston wasn’t out of the ordinary for having a few good attackers, nor were they winning with their offense.9

Tom Sanders, KC Jones and John Havlicek made up an excellent supporting cast of defenders, although Boston lacked a second big man to play next to Russell. When he retired in 1969, along with Sam Jones — who was down to 26 minutes per game by then — the Celtics dropped a whopping 8 points in SRS (from a 59-win full-strength pace to a 36-win one) despite returning the rest of their eight-man rotation.10 So while Boston fielded a strong team around Big Bill, there’s nothing indicating that they could sniff the same heights without him.

In the postseason, the defensive domination rolled on. Below, I’ve compared Russell’s playoff defenses to those of other all-time big men. The gray bubbles in Russell’s column are the Celtics individual performance in each year. Note that Boston never had a subpar defensive postseason with Russell, and that its worst playoff runs were clustered at the end of his career as a he slowed down:

When pundits wax about old-timers dominating, these are the kinds of outlying result we should see from an ahead-of-his-time star. Boston’s best five-year defensive rating (1961-65) was 9.2 points better than league average. No other team in history has been 7 points better than the league over a five-year stretch. Even this understates the Celtics’ defensive dominance because the league average was pulled down by Boston’s presence.11

Finally, there’s Russell’s total lockdown of other All-Star centers. He was so complete as a defender that he was likely the best team and man defender of his era. Here’s what he did to All-Star pivots during his career:12

All of these players declined in efficiency, and only Thurmond improved his scoring (although on dreadful accuracy). Willis Reed and Zelmo Beaty were vaporized by Russell. The numbers are particularly compelling because, unlike today, Russell played most of the game and Boston did not double-team frequently.

Of course, his most notable conquest was Wilt. While one might think that Chamberlain’s line against Russell was exceptional — he averaged 33 point per game on +4 percent efficiency — adjusting for pace yields a scoring profile comparable to ’89 Roy Tarpley or ’10 David Lee. The sample above comes from Wilt’s volume scoring years only, but even in 1967, arguably Chamberlain’s most revered season, Russell slowed him down significantly. Against the league, Wilt averaged 24.6 points per 36 minutes on 64.9 percent true shooting (TS). But in nine games against Russell, his scoring dropped 4.3 points and his efficiency plummeted 10.8 percentage points.

In Wilt’s 1962 50-point season, he faced Russell 17 times and the rest of the league 75 times (he played Boston twice without Russell). Chamberlain averaged 50.9 points and 53.6 percent true shooting against the league, but 37.2 points on 50.1 percent true shooting against Russell (with 4.3 fewer free throw attempts per game). Russell shaved 14 points per game off Wilt’s average and his drop in efficiency — from 1.07 points per attempt to 1 point per attempt — is the full-game equivalent of a GOAT-level offense regressing to average.

So, we can safely crown Russell as defensive royalty. His offense pales in comparison to other greats, but he was not a poor offensive player – in many ways, he was above average for his day, although it’s unlikely his contributions moved the needle much. Between 1959 and 1965, he finished in the top half of centers in points per 36 twice (’60 and ’62), while falling between the 61st and 71st percentile in true shooting (efficiency) in both years. His scoring regularly improved in the playoffs before trailing off in his last four seasons.

The impact studies we have for that era suggest he’s, at worst, a player with MVP-level lift, and at best, view him as one of the most valuable players of all time. The noisiness of that data and Russell’s outlying status as a defender make it difficult to confidently pin down his value. But, the restrictive dribbling rules, poor spacing and sheer volume of possessions played make it likely that his pre-3 point impact was well ahead of today’s best defensive scores.13

His portability was superb – any team at the time would have exploded defensively by adding him, and his passing and finishing would provide bonus value for any competent offense. He has excellent era longevity, and I consider his peak among the better ones in NBA history. Shaving my valuation of his defense by 5 percent per season — a plausible but conservative estimation — would drop him a spot or two in these rankings while keeping his per possession impact in line with the modern defensive juggernauts.

On the other hand, there’s a viable argument that he was even better than I give him credit for. Like Jordan and LeBron today, his prime was an onslaught of MVP-level seasons and, relative to his era, he might have been the most valuable player ever. Yet for this exercise, his ambiguity leaves enough doubt that he lands at No. 4, narrowly edged out by the man in front of him.

  1. He qualified for the 1956 Olympics in the high jump before surrendering his spot. According to his own claims in Second Wind, he “could get his eyes above the rim.”
  2. Wilt’s estimates are similar. Box scores courtesy of nbastats.net.
  3. When he unleashed this style in the ’50s, it carried the San Francisco Dons to a 55-game win streak, then steered the Olympic team to a comical 55-point differential in 1956, about a dozen points better than any other American team in history.
  4. League-wide field goal shooting settled around 43 to 44 percent for the bulk of the ’60s.
  5. I first saw this injury pointed out by a realgm poster named ThaRegul8r, who has an encyclopedic collection of NBA newspaper archives.
  6. In ’58, Boston went from a 5.1 SRS with Cousy to a 9.5 SRS without him. In ’61, from 6.4 to 8.7 without Sharman and in ’62, from 9.1 to 13.3 without Cousy.
  7. Russell injured his ankle in Game 3 of the Finals that year, missed the next two games and hobbled around for 20 minutes during Bob Pettit’s legendary 50-point Game 6.
  8. It’s possible a style shift slanted the numbers towards the defense, although even that would barely explain such outlying results.
  9. Having a few All-Star teammates was normal then. Russell suited up with six different All-Stars totaling 27 selections (in 13 years). West played with eight different players totaling 22 selections (in 14 years). Chamberlain played with 10 different players totaling 24 selections (in 14 years, including two teams he played on in 1965.)
  10. Four other teams in history have posted a 5 SRS then lost at least 8 points in the ensuing season: the ’90 Cavs (who lost Ron Harper, Brad Daugherty for half the year and Larry Nance for a third of it), the ’97 Spurs, (who lost David Robinson and Sean Elliot for half the year), the ’99 Bulls (who cleaned house) and the ’11 Cavs (who lost LeBron James).
  11. Using all the teams in the league, the 1993 Knicks had the best z-score in NBA history (2.9). But if we calculate the league average without the top team, the best z-score in history is the 1962 Celtics at 4.2 (!), and only eight teams have a score above 3.0, four of which were Russell’s Celtics.
  12. This data is only available from select games in which field goal attempts were recorded. Beaty’s sample was only seven games against Russell because of this. The others played around 20 games against Boston. Full logs are available for Chamberlain, so his sample against Russell was 101 games.
  13. Based on 20-years of RAPM data from the 3-point era, I credit him with about 6 points per game of impact on defense at his best. On a per possession basis, this is comparable to Hakeem Olajuwon.

Comments are closed.