A week ago in this space, I used a metric called Advanced Statistical Plus/Minus (ASPM) to rank the talent on every Olympic team the United States has fielded since 1992, the first year that professionals were allowed to participate in the Games under FIBA's bylaws. The team on top of that list wasn't altogether shocking -- the original 1992 Dream Team, probably the greatest basketball team ever assembled, took the crown in the end -- but that's not to say the results had no intriguing aspects.
For starters, the difference in talent between that fabled '92 team and some of Team USA's later incarnations wasn't as huge as you might think. Against an average team of their NBA peers in 1992-93, the Dream Team would be expected to rack up a margin of plus-23.1 points per 100 possessions. That is a huge number (by comparison, the 1996 Bulls' best-ever single-season mark was merely plus-13.4), but not one very far removed from the U.S.'s expected plus-20.5 margin in these 2012 Olympics -- to say nothing of America's plus-22.5 and plus-22.4 talent levels in 1996 and 2008, respectively.
Hypothetically, this means that Kobe Bryant's best squad, 2008's Redeem Team, would have about a 44 percent chance of toppling Michael Jordan and the legends of 1992 in a best-of-seven series on a neutral court. And, as I wrote last week, that number could actually be given a boost by the supposition that the average player in 2012 is better than his 1992-93 counterpart. After all, today's players are certainly stronger, faster and more athletic than players were 20 years ago.
But as it turns out, that assumption is faulty. From a pure basketball production standpoint, it appears the typical NBA player is actually no better in 2012, in an absolute sense, than he was in 1992.
How did I determine this scientifically, you ask? Borrowing a methodology from Baseball Prospectus' "Baseball Between the Numbers", I looked at all players since 1992 who played at least 1,500 minutes (per 82 scheduled games) in back-to-back seasons, then tracked how their ASPM scores were impacted by moving from one league year to the next. (In case you were wondering, I also controlled for the year-to-year effects of aging.)
The logic here is simple: If a league is more difficult one year than it was the year before, the same group of players would be expected to put up inferior ASPM numbers than those they posted the previous season.
As it turns out, when you run the data since 1992, the NBA is essentially operating at the same talent level as it was when the original Dream Team played. Although the league's year-to-year talent level has changed over time -- it declined quite a bit from 1992 until the 2003-05 period, for instance -- it has bounced back in recent seasons to reach essentially the same point at which it had been 20 years before. (Incidentally, this seems to vindicate those who would characterize the early post-Jordan years as the NBA's dark ages, a period of transition where overall talent hit a lull.)
The shortcomings of the weak 2000 and 2004 Olympic teams reveal this lull in talent level.
The team the U.S. sent to Sydney in 2000, despite going 8-0 and winning gold, had even less raw ability than the version that lost in Athens four years later. Many of 2000's top-tier stars would have been mere supporting players on the 1992, 1996, 2008 or 2012 teams, while it's likely the entire lower half of the roster wouldn't have survived the cuts for any other Olympic team aside from 2004. The poster child for that 2000 team's lack of talent was Vin Baker, who played in the Olympics just months after putting the finishing touches on a below-average 14.0 PER season (with horrendous defense) for the Sonics. At minus-2.1, Baker had the lowest true talent estimate of any pro player to suit up for Team USA in the Olympic Games.
The 2004 team's stunning failure was well-documented, to the point that it has now passed into cliché that they lost because they weren't a cohesive unit with enough role players. LeBron James, Dwyane Wade, Lamar Odom and Amare Stoudemire were nowhere near the players they would eventually become; Stephon Marbury, Carlos Boozer and Richard Jefferson were never all that great in the first place; and the team employed not one but two players with below-average true talent ratings -- a 20-year-old Carmelo Anthony, fresh off a usage-driven 17.6 PER rookie season in which he played exceedingly poor defense, and Emeka Okafor before he had logged a single NBA minute.
It's a testament to the improvements made by the rest of the world that the United States' equally unimpressive teams of 2000 and 2004 met with such different results. In fact, all of these true talent ratings provide a telling gauge of how much the field has improved since the first Dream Team took the court in Barcelona. To wit: The 1992, 1996 and 2008 U.S. teams were roughly equal in terms of aggregate NBA talent, but the '92 team crushed opponents by 43.8 ppg; the '96 squad won by 31.7 on average; and the '08 version beat their opponents by only 27.9 ppg.
Regardless of international competition improving, the 1992 Dream Team is still the United States' best-ever Olympic roster, especially when factoring in that the NBA talent level in '92 was essentially the same as it is today.