As the NBA considers the prospect of expansion at some point in the future, the league must consider a pair of factors. How many markets can support NBA teams? And how many teams can the league support without harming the level of play? In an interview with ESPN's Darren Rovell earlier this week, new NBA commissioner Adam Silver expressed skepticism about the size of the player pool from which the league draws.
"I and the owners will look at not only dilution of economic opportunities with one more partner to divide national and international money but also dilution of talent," Silver said. "Right now [people] are already making comments about the [Eastern Conference], so is the ideal time to be adding another 15 or 30 players to the league?"
This is not the first time Silver has raised the issue of expansion weakening the league. He and former commissioner David Stern actually disagreed on the topic when they jointly addressed media before last year's NBA All-Star Game. After Silver questioned the number of potential NBA players, Stern countered, "I think it's an unlimited number."
As evidence, Stern pointed to the number of international players entering the league, specifically from Africa. Indeed, the NBA's growth in popularity internationally is a key reason the league could support not only one or two additional teams but potentially several more.
The growing talent pool
Consider a hypothetical question: How many people could potentially play in the NBA? In theory, that number is limited only by the population of humans between ages 20-40, roughly. In practice, it's not that large. While the league has gotten better at identifying and developing talent from every corner of the globe, the average African is far less likely to reach the NBA than the average American -- especially if he is not exceptionally tall. Of the 28 NBA players raised in Africa (raised instead of born so as to exclude Steve Nash and others who were born in Africa but learned the game elsewhere), 27 have been 6-foot-8 or taller. Shorter stars might choose soccer or other sports that are more popular locally.
There is a handful of other explanations, too, for why the NBA doesn't pull players as efficiently from foreign countries as the U.S. It's also more difficult to pick out and nurture potential NBA players without the benefit of height as a criterion. And the pool of guards in the U.S. is much larger, so there's less need to look overseas.
Still, it's obvious that there are more potential NBA players now than there were three decades ago. As recently as 1981, not a single player in the league grew up outside of the United States. This season, more than 17 percent of minutes have been played by players who would have had virtually no chance of reaching the NBA when Pau Gasol was born.
To estimate the size of the NBA pool, I theorized that the total player pool is proportionally as large compared to the population of the United States year-by-year as the relationship between minutes played by American players and those raised overseas. So while American population has grown steadily for several decades, the league's talent pool has grown much more quickly, from 217.2 million in 1976-77 when the NBA merged with the ABA to 380.8 million this season.
Now, these numbers shouldn't be taken literally, as they include people too young or too old to play in the NBA. The important factor is their relative size. Since the merger, the NBA's player pool has nearly doubled. Yet in that span the league has increased by barely more than a third, from 22 teams to 30. As a result, the ratio of potential NBA players to teams is now higher than it has been since 1969-70, when the league was in the midst of rapid expansion to counter the arrival of the ABA.
In the 1980s and 1990s, considered the NBA's golden age, there were around 10 million potential players per team, from a high of 10.7 in 1987-88 just before a four-team round of expansion to a low of 9.4 afterwards. Right now, the equivalent number is 12.7 million potential players per team. Adding two teams would still leave that figure at 11.9 million, far higher than any point in the '80s and '90s. To get the ratio down to 10 million players per team would require adding a staggering eight teams.
The lost NBA-caliber players
If there are many more players of NBA quality than a decade ago, but no more roster spots, where are these extra players now? My SCHOENE projection system is uniquely capable of answering this question using the translations I've devised that turn player stats accumulated in the D-League, the Euroleague and EuroCup into their NBA equivalents.
For players who saw action in those three leagues in 2012-13 and played a qualifying number of minutes, SCHOENE combines their translated performance over the previous three seasons (adjusted for aging) into a per-minute rating (Win%, which is equivalent to PER) that represents our best guess at how they would perform in the NBA. Here's how this looks graphically for players in all four professional leagues.
Within the NBA, talent mostly takes on the familiar bell-curve shape of the normal distribution, albeit skewed to the left. Because the average rating (.500, or 15 for PER) is actually the average minute rather than the average player, it actually peaks below average; the best players play more minutes, so there are far more below-average players than above-average ones.
But when the other leagues are layered on top, it becomes obvious that NBA talent isn't normally distributed. Instead, the league is pulling from the tail of a much larger bell curve that in theory would include the NBA ability of every able person in the world, if we could rate it. This concept was crucial to developing the concept of "replacement level" in baseball, because it means there is a huge supply of roughly equivalent players that can play competently at the highest level.
In the NBA, I've found that replacement level -- the performance we'd expect from a player signed for the minimum -- is a little better than 80 percent of average, or about a .415 winning percentage. The peak in the chart there is slightly artificial. In the absence of other evidence, SCHOENE assumes that a player contributes at replacement level. If we had more data, we could separate more of these players from replacement level either way. Nonetheless, it's clear that many players in other leagues and particularly the elite Euroleague are capable of contributing above-replacement performance in the NBA.
Let's put some hard numbers to this. Of the 478 players from the NBA sample, 322 rated better than replacement -- about 11 per team. Why not all of them? Some players' above-replacement contributions on defense or away from the ball can't be qualified from the box score. Others, simply, are in the NBA by mistake because of bad talent evaluation or guaranteed contracts. There are about 30 above-replacement players in both the D-League and the EuroCup, and a whopping 120-plus in the Euroleague, the highest level of competition outside the NBA.
That's a total of more than 500 players better than replacement level, and even if we required every roster spot to be filled by someone meeting this qualification -- not the case today -- still enough to comfortably add three more teams to the league.
Barring international expansion, the NBA will never capture 100 percent of the players capable of playing in the league. Some European players, like Rudy Fernandez, would simply rather play at home. But others, including Americans playing abroad, would likely make the leap to the NBA if there were more money or guaranteed roster spots available, which expansion would provide.
Expansion could raise issues with competitive balance. In the short term, naturally, expansion teams weaken the bottom of the league. It's no coincidence that the Chicago Bulls won their record 72 games in a double-expansion season. In a larger league, the best players would become even more valuable than they are today, making their desire to team up problematic. Still, the number of NBA-caliber players shouldn't be an argument against expansion. In the global NBA, there are more than enough to go around.