Within US professional sports, trades within one’s own division are
often perceived to be disadvantageous. We ask how common this practice
is. To examine this question, we construct a date-stamped network of all
trades in the NBA between June 1976 and May 2019. We then use yearly
weighted exponential random graph models to estimate the likelihood of
teams avoiding within-division trade partners, and whether that pattern
changes through time. In addition to the empirical question, this
analysis serves to demonstrate the necessity and difficulty of
constructing the proper baseline for statistical comparison. We find
limited-to-no support for the popular perception.
Draft under review.
A common refrain among sports commentators is that professional sports teams avoid trading players with other teams from within their own division (Wong 2017; Simmons 2020). Some offered conjectures on why this should be avoided include: not wanting to improve the competitiveness of a team’s direct rivals (Ley 2017), especially in leagues where games within the division are more common than other match-ups (Bates 2015), or wanting to avoid fans being reminded of players they gave up (Fox Sports 2015), especially if they turn out to play better for their new teams (Ley 2017). Despite the frequency of this speculation, attempts to quantify the actual frequency of such a prohibition has been rare (for one limited exception, see Ahr 2018).
Beyond the popular perception of the avoidance of intra-division trades, there is also ample scholarly literature that would lead us to the same expectation, particularly in professional sports (Stewart and Smith 1999). For example, within a field where the actors are (or are perceived to be) competitors with one another, they may avoid cooperating with one another because it is perceived to be a competitive disadvantage (Hoffmann et al. 2018), though that assumption has been questioned (Bengtsson and Kock 1999; Peng et al. 2012). Instead of cooperating directly with one’s competitors, actors may therefore seek out means of collaboration with those outside the competitive field (Soda and Furlotti 2017). For example, in the case of NBA player development, this could lead to a variety of strategic patterns. Teams may prioritize trading with teams they do not perceive to be direct competitors (Barman 2002), such as those outside their own division. Alternatively, general managers may develop strong stable relationships as resources (Elfenbein and Zenger 2014), for example with particular player agents to provide comparative advantages in access to players on the free agent market (thus reducing their reliance on trades). Teams also could carve out recruitment niches that avoid direct competition (Soltis et al. 2010; Barman 2002). For example, recent shifts in player “apprentice” opportunities through the developmental league and foreign partnerships has opened opportunities for teams to gain contractual rights to players outside of the trade system (Keiper and Barnes 2020). Each of these possibilities could bolster teams’ ability to avoid cooperating through trading with their direct competitors.
In sum, we therefore investigate how strongly division shapes trade partners among National Basketball Association (NBA) franchises, focused especially on whether teams tend to avoid trading with other teams within their own division.
Here, we draw on a database of all player transactions in the NBA from the beginning of the 1976 season—when the NBA merged with the ABA uniting major professional basketball in the US under one league—through the completion of the 2018-19 season (Richardson 2020). We compile this list of 1,977 trades into 43 annual trade networks, with the nodes representing teams, and each edge representing a unique trade between those teams. Each trade is assigned to the “trade season” which runs from the end of the previous season through the corresponding season’s trade deadline.1 The number of teams within each annual slice changes over time as the league gradually expanded from 22 to 30 teams.2. The number of trades observed within each slice varies from year to year (range 22-86 or 0.8-2.9 if expressed in per-team averages; see Figure 1), and teams exhibit differing rates of trading (see Figure 2).
Figure 1 presents the trend of number of trades per season. Since, the number of teams varies across the window, we standardize these values to represent the number of trades per team in each season. This pattern shows a roughly U-shaped trend, with the highest rates of trades (nearing 2 per team) occurring early and late in the period and the lowest rates occurring in the early 1990s. Notably, this low-point came after the introduction of unrestricted free-agency (in 1988), allowing players to sign with any team after their contract expires. As a corollary, Figure 2 presents the frequency distribution of trades per team across the full observed window. There are some teams (e.g., the Spurs) who consistently trade less frequently, and others who are much more trade active (e.g., the Mavericks).
To address our research question, we needed to supplement these trade data by compiling a list of each team’s division membership. From the 1976-77 season through 2003-04, there were four divisions (Atlantic, Central, Midwest, and Pacific). From the 2004-05 season onward, there were six divisions (Atlantic, Central, Southeast, Northwest, Pacific, and Southwest).3 While those division/conference names otherwise remained stable, teams changed divisions at various times, whether for geographic reasons or as new expansion teams were added to the league. Accordingly, each team’s division is assigned as current to a particular trade season (for example, the Buffalo Braves were in the Atlantic division in 1977-78, but became the San Diego Clippers and moved to the Pacific division for the 1978-79 season). Combined, these data allow us to investigate the tendency for teams to avoid trading with other teams in their own division.
Analytically, we proceed in three steps, which allow us to address two simultaneous aims. Primarily, these steps allow us to build an appropriate test of our research question. Secondarily, these steps also allow us to illustrate the proper way to statistically condition a question such as this, and to explain the need for doing so. This combination motivates our usage of a relatively recent development for statistically modeling weighted network data.