The 1980s and early ’90s are referred to as the NBA’s “Golden Age,” with stars like Magic Johnson, Larry Bird and the G.O.A.T. himself, Michael Jordan, patrolling the court. But are we entering a new golden age of star power?
Russell Westbrook is challenging Oscar Robertson’s record for a triple-double season, and doing so with a fraction of the possessions on the court. James Harden isn’t lagging far behind, but is posting far more efficient numbers. Stephen Curry just became the first unanimous MVP ever after leading the league in true shooting percentage and scoring in the same season in unprecedented fashion. Anthony Davis is posting 30 and 12 seemingly every night. Jimmy Butler is making it cool again to be a Chicago Bulls shooting guard (seriously, I can’t find a Chicago Bulls argyle sock monkey anywhere).
Superstars are forming constellations in Golden State and Cleveland.
This level of star power is unprecedented.
I hear the cries of “Apostasy!” drowning out my typing. There’s a knee-jerk reaction sometimes to hold on to the past. If we declare something today to be “equal” or even “better,” there will be some who reflexively protect what was before. But consider this: Carl Lewis won the gold medal in the 100 meters in the 1984 Olympics with a time of 10.32. This year, that time wouldn’t even qualify him.
Athletes have gotten better across the board. They run faster and jump further and higher. They’re better conditioned and have better diets, with doctors and trainers monitoring them more than ever before.
The study of the game has grown at a profoundly exponential level through analytics.
The “farm” for players — now groomed from grade school — has grown into the entire world.
Why wouldn’t the same thing that applies in other sports apply in basketball? Why wouldn’t that make for better basketball players?
Yet, some people bristle at the notion of comparing Russell Westbrook’s triple-double season with Oscar Robertson’s, as though it’s an affront to Robertson.
There’s an obvious caveat here: Every player needs to be judged based on his era. Who knows what Jordan or Robertson would be like today with the advantages in film study and the like? Yes, today’s players might be “better” athletes, but yesterday’s players would be more athletic if they played today and had access to the same training advantages.The argument works both ways.
That’s why players need to be evaluated by the age they played in. But that’s also the genius behind John Hollinger’s player efficiency rating, because he adjusts everything for league averages by season. By looking at PER, you see how a player did against his competition that year. A player with a 30.0 PER in 1960 has about the same level of dominance as a player who has one today, even if his stats are much bigger. PER neatly handles the faster pace, etc.
MVP-caliber players based on PER
While it’s not a perfect metric (and nothing is), it’s about the best thing we have and is effective.
According to Hollinger’s book, Pro Basketball Forecast, a PER of 15.0 is an average player, 25.0 is a borderline MVP, 27.5 is a typical MVP and 30.0 is a “season for the ages.” Based on that criteria, I broke down every player with a 25.0 in history, by season. Here’s how this year stacks up against the past:
The colors in the chart denote which “level” of MVP performance players are having. Red means they’re closer to the borderline level. Yellow is around that 27.5 mark. Green means they’re near or over the 30.0 demarcation.
The most noticeable thing here is that the number of MVP-caliber players this year is at a level that we’ve never seen. There are not only more players operating on an MVP level than ever before, but more players than in the two previous combined seasons.
While many seasons see three or few such performances, this year we’re seeing 16 of them (and there are another four between 24 and 25).
There are six players who are flirting with a 30.0 PER: Anthony Davis (32.6), Kevin Durant (30.7), Russell Westbrook (30.6), James Harden (29.5), Jimmy Butler (29.4) and DeMarcus Cousins (28.5). The most players to ever top 28.5 in a season is three in 2008-09. There were only 14 total before Charles Barkley and Michael Jordan accomplished it in 1990-91.
MVP-caliber players based on Game Score
The obvious disclaimer here is that we’re only a quarter of the way through the season. Everyone else did it for a full year; these guys have done it for 15-20 games. And that’s a perfectly fair and reasonable argument.
While we don’t have a perfect apples-to-apples way of looking at just the first quarter of the season, we can look at the average Game Score — Hollinger’s sort of “single-game equivalent” of PER. Using 20 as a bottom-end average, there are 150 players since 1983-84 who qualify:
Once again, this year is at the top, but it’s much closer, and it’s only tied with 1990-91 for that. So at first glance, it seems like there’s justification to the “it’s still too early” qualifier. However, if we pry a little deeper, that caveat needs a caveat.
You’ll notice that the columns are color-coded by average minutes per game, with red meaning fewer minutes and green meaning more. Unlike PER, Game Score is not a pace or minute-adjusted stat; it’s cumulative. So, players who are on the court more have an inherent advantage. Ergo, fewer minutes means it’s harder to get to 20.
If we look at the same chart where the color coding indicates “Game Score per minute (GS/MP),” it jumps out how much more economically the players of today are accruing their totals. That, in turn, reveals that the PER drop-off by some is inevitable, and it’s not likely to be as whole-scale as you might think:
The six players who are within range of 30 all have a GS/MP of at least .6137. Of the 10 qualifiers in 1990-91, only three — Jordan, Barkley and David Robinson — are on that same level per minute. And if we look at the aggregate, the 10 players collectively average .6325, which ranks the third-highest overall (regardless of sample size), and 10th-most where the group collectively had 100-plus games.
In other words, the caveat to the caveat here is that on a per-minute basis, today’s stars still shine brighter. Even when you account for the fact that we’re just a quarter of the way through the season, we’ve never seen this many players start out the season this well — or at least since 1983-84. And judging by the fact such high-level performances were limited to one or two players per year prior to that, it’s unlikely that there was anyone before.
Not just an anomaly
So, this raises the question: Is this an anomaly, or are there credible reasons to account for the sudden level of cream? There are a few things which indicate that it’s the latter.
First, there’s the fact that the number of players who are at such a level has been climbing steadily over the last few seasons. Last year was the most we’d ever seen before at nine. In 2014-15, the record was tied at seven. And the year before that, qualified players had six.
Over four seasons, that’s a total of 38 player seasons with a 25 PER or higher, about 19 percent of history. Furthermore, since 2004-05 when the last of the rule changes regarding hand checks and zones went into effect, there are 77 such seasons, which accounts for 38 percent of all players who have done it.
By comparison, during the entire stretch from Magic Johnson’s rookie year in 1980-81 to Michael Jordan’s retirement from the Bulls (spanning the entire “Golden Era”), there were only 58. That’s 19 fewer qualifying players in six more seasons.
It’s evident that there were two big things that happened that have propelled star power, and the rule changes are the first. But what prompted the second step up in the last few seasons?
This answer is a little more difficult to substantiate, but the corresponding rise in emphasis on shooting efficiency, player tracking, the increase in the use of video scouting services like Synergy and Vantage Sports, and the more economic focus on minutes all seem to have coalesced into the best players playing a better brand of basketball.
In sum: analytics work.
How about we stop bemoaning the “loss” of things like the “art of the midrange” or the “post-up?” I was there. I watched. It wasn’t that great.
There’s nothing innately exciting about watching a big man beat his defender back in painfully slow fashion just to get off a shot that has less of a chance of going in.
When people say “analytics don’t work,” the numbers make it pretty obvious they’re wrong, and it’s most apparent in the star power it generates. Never before have we seen so many guys playing at such an elite level. Let’s appreciate that.