USA Today Sports
I know saying one conference is distinctly better than the other may be considered oversimplification, but the Western Conference, when compared to the Eastern Conference, is just better. There have been a ton of words written about this disparity around the interwebs this season, so take to Google if you have any confusion about this basic premise.
Recently, Kirk Goldsberry made a postseason bracket (pictured below) that featured what the NBA playoffs would look like if the seeds were determined by record, without divisional/conference based restrictions. Goldsberry’s method is simple; the best record, seed #1, would play the worst record, seed #16. Ethan Sherwood Strauss, also, recently suggested the idea of abolishing conferences. Had the Eastern Conference not been as bad as it’s been, outside of the Miami Heat, Indiana Pacers — [We think], Toronto Raptors, and Chicago Bulls [Stretching] — it’s unlikely that such a vast proliferation of reform suggestions even surfaces in the first place.
Out West; the Dallas Mavericks, Memphis Grizzlies, and Phoenix Suns are competing for two postseason positions, ultimately, one of them will not qualify. Moreover, moving East; at least one, if not two, sub .500 teams will qualify for the postseason. This is not an ideal balance. Goldsberry’s “Sweet 16” includes the Timberwolves, Suns, Grizzlies, and Mavericks but disqualifies the .500 Charlotte Bobcats, as well as the sub .500 Atlanta Hawks.
The New York Knicks, who are fighting for their playoff lives, stand no chance using the Goldsberry method. I particularly enjoyed this bit within Strauss’ column, an essential call-to-arms demanding the league address the issue at once.
“Regardless of tactics, the NBA should at least address this in some capacity. The East/West divide takes a regular season that’s already assailed as “meaningless” and adds some absurdity on top of it. Though the reasons for the status quo are understandable, the status quo makes for bad entertainment. It’s best to try and fix that.”
Enter myself and Matt D’Anna, also known as @Hoop_Nerd. We’ve come together to try and do the legwork that Goldsberry, Strauss and many, many others have suggested before us. D’Anna and I aspire to present a logical solution to the NBA’s issue involving balance between the conferences. However logical, altering the landscape of the league requires significant changes that may seem irrational or illogical without context. Some may consider the entire premise of this proposal to be a knee-jerk reaction because of the unique instance occurring this season. Ultimately, think of it as potentially laying a foundation for changes the NBA may wish to implement sometime in the future.
Back in November, Zach Lowe published Abolish (Useless) NBA Divisions: Step One of a Radical Plan. Lowe states that divisions exist, in theory, to invigorate rivalries and ease travel. He’s right. The clear-cut reason that divisions exist are for creating a set-amount of regional matchups (four times a year). This allows more head-to-head instances between proverbial bitter foes, each year. The Los Angeles Clippers-Lakers and the Brooklyn Nets-New York Knicks are two that come to mind off-the-top of my head. Rivalries exist predominantly because of history and also because of location, but, they aren’t dependent on divisions; postseason meetings and individual player/coach/staff narratives also factor into rivalries.
Why are divisions bad?
Divisions are hurting the NBA’s product, particularly this season. The geographic sorting that determines each division has added salt in the league’s wound that is the current conference structure. Conversely, remember, the format in place is also considerate–in compliance with an 82 game schedule–of travel. This is how each team’s season is divided.
Four games against each of the other divisional opponents, [4×4=16 games]
Four games against six, out-of-division, conference opponents, [4×6=24 games] Ex: The Minnesota Timberwolves, in the grouped with the Oklahoma City Thunder, Portland Trail Blazers, Denver Nuggets, and Utah Jazz, play 24 games against teams in the Western Conference, but outside of the Northwest Division.
Three games against the remaining four conference teams, [3×4=12 games] Above, we’ve scheduled a total of 40 games against 10 opponents [four division opponents, and six conference — but not divisional — opponents]
Two games against teams in the opposing conference. [2×15=30 games] Each Western Conference team will play each Eastern Conference team, twice, per season.
- A five year rotation determines which out-of-division conference teams are played only three times.
This format may be simplistic, and not necessarily unruly, because most are accustomed to the process — it hasn’t always been so detrimental toward one side, or the other. Looking at the standings as of April 7th, 2014, we can see the disparity throughout each division.
The Central and Atlantic divisions, within the Eastern Conference, host only two teams with records above .500, while the Southeast contains three because the Charlotte Bobcats won their most recent game. Out West, there are three teams above .500 in the Southwest and Pacific divisions, and two in the Northwest. The Wolves are 38-38 and, had they found a way to defeat the Orlando Magic this past Saturday, would be the ninth team in the Western conference with more wins than losses. The East has only seven teams above .500, but it could fall to six before the season ends.
Reverting back to the NBA’s scheduling formula, we know that teams in the East are playing inferior, divisional as well as in conference, opponents more often that teams in the West. The schedule favors teams in the East, because Western Conference teams are forced to face a higher level of competition, more frequently.
Collectively, the West’s record against the East is 281-164. Matt Femrite keeps East vs. West records, updated weekly, at his website; Chicken Noodle Hoop. To the left you’ll see how the West has performed against the East, and table to the right is E/W point-differential by month.
Clearly, one conference is better than the other. That’s why we’re here.
The Issue of Travel
Matt Winick, Vice President of Scheduling and Game Operations for the NBA, is responsible for all regular-season and playoff scheduling. He also oversees the administration of the NBA Pre-Draft Camp. Winick states: The NBA sets the league schedule to accomplish both competitive balance and a reduction of costs. The goal of the NBA schedule, as it is constructed, is to be “efficient from a competitive standpoint with an indirect consideration of travel costs.
The competitive balance has left a lot to be desired, this season. Winick and the NBA’s goal, in that regard, has failed, but what about “the indirect consideration of travel costs?”
How much does travel cost to begin with? This study states that ground transportation, for most teams, involves two charter buses and one medium sized truck to carry people, equipment and luggage. Ground transportation costs average $3,000 per day, narrowed down to the rate of $85 per person, per day, which makes the typical average cost per person in the traveling party total $432 per day (including hotel and per diem). A ‘traveling party’ may include, but is not required/limited to, 35 people. Players, coaches, trainers, administrative staff, and media personnel [that of which is owned by the franchise] are all members of a team’s traveling party. From 1999-2000 to 2001-2002 (three complete seasons) there were 1,867 trips taken by all teams. This includes ground transportation as well as travel by airplane.
Obviously, airfare for the NBA isn’t sold at the average rate. Most teams are part of the NBA charter program. The program allows individual franchises to contractually rent passenger planes, avoiding maintenance and upkeep of owning their own jet. The service involves transporting a team to a city and then flying to another city to transport a different team. “This lowers the average fixed cost by increasing the quantity of trips the plane makes and not having to sit idly on the ground while a different team uses a separate plane,” according to Randy Pfund, former General Manager of the Miami Heat. [cited within aforementioned study].
Winick and the NBA’s method that aims to achieve a ‘competitive balance’ has proven to be flawed. Without having the leagues budget/expense numbers, I’m unable to determine the NBA’s annual spending towards travel, per season. However, there is data that reflects that the geographical clustering of divisions, conferences, paired with the current scheduling format is unbalanced. The league has opportunity to be more efficient with each team’s annual travel expenses, in theory.
A Structure That Needs More Balance
Here is a map of the NBA’s current landscape, and how ‘geographically clustered’ each division is.
Note: The more black space between colored dots, the less clustered that division is.
D’Anna also constructed a chart that reflects each team’s Estimated Season Distance, or, ESD.
ESD is the Arena-to-Arena distance for every team in the league. It is not the actual distance travelled for each team in a season; rather, it is a metric representing the distance a team travels from game to game, relative to the rest of the league. ESD assumes travel to each away game begins at the team’s home arena.
His estimates must be taken seriously, here’s why: Andrew Nutting, currently a Visiting Professor of Economics at Hamilton College in New York, calculated the average miles traveled by NBA Teams in a Season from 1990/1991-2006/2007, omitting the strike-shortened season of 98-99. The larger sample size shows that D’Anna’s estimates aren’t egregious in any way.
Both collections of data show NBA teams on the West Coast and in the center of the country do the most traveling. Nutting’s study concludes there is little evidence that traveling further distances reduces win production. The purpose of displaying each study is not to imply that teams in the Western Conference are at a disadvantage, the intent is to show how many miles teams are traveling.
The Fun Stuff
Let’s try to restructure the NBA. Remember, we’re trying to find a state of equilibrium between two conferences whilst indirectly considering ways teams can travel, efficiently.
Proposal #1 was conjured with the intent to keep divisions. Because it’s the NBA we’re dealing with, the goal was to change as little as possible in order to appease the formula currently in place. Sure, if it’s not broken don’t fix it, just try altering it a little. The new divisions consist of six, five-team divisions in same East/West conferences that are designed by spatial clustering. Proposal #1 uses the same 82 game schedule. The table below displays the new divisions, the chart shows each team’s ESD.
Not only does the ESD average per team increase by 329 miles, Proposal #1 doesn’t repair the broken competitive balance.
Proposal #2 features no divisions. It uses the same conferences, and adds four games to the schedule. Under this proposal, teams play an 86 game schedule [43 home/43 away]. Each team plays opponents within their conference four times during the season, making 56 of the 86 games against inner-conference foes. The remaining 30 games are fulfilled by playing each team in the opposite conference, twice.
You may be thinking; MORE GAMES?!?!?! Adding more games eliminates the unfair rotations currently in place. The other benefits Proposal #2 would provide the NBA are as follows.
2013-2014 ESD: Miles per game is less on average, per team (509 MPG in current format, 506 in this format)
The 4 additional games add approximately 440 MPG per team; 69 miles less than the 509 MPG average in the current 82-game schedule
- More games at a lower per game cost!
Proposal #2 is feasible, but, considering that the current 82 game schedule is already widely considered to be a little too lengthy, it’s doubtful the league would consider this format.
Proposal #3 is where we decided to not only abolish divisions, but conferences as well. It would appease the Goldsberry Method by having the top 16 teams qualify for the postseason, each year. There are two options when it comes time for scheduling.
Play each team three times (87 game season). This option proves to be problematic. Teams would need to rotate every other season (2 home, 1 away vs opponent x in Year 1; 1 home, 2 away vs opponent x in Year 2)
- Play each team twice per season (58 game season). Less games is a good thing! The current schedule is 82 games in approximately 165 days (1 game per 2.01 days).
- This option would have the season start on Christmas Day, teams would play 58 games in 112 days and the playoff start date is preserved. (1 game per 1.93 days).
- More even distribution of travel across the league; Midwest/East coast is reduced significantly.
- Teams travel less often, while also covering less distance. Theoretically, because we don’t know how much it costs for the league to charter planes, this saves the NBA a significant amount in travel costs each year.
Proposal #3 is undoubtedly the most logical option of these trials, if someone in the league office is watching — that’s what we’re suggesting as of now. What we do know; the current system is flawed and possesses deficiencies.
Considering the state of the Eastern Conference, in comparison to the West, the NBA’s goal of scheduling to ensure a competitive balance has failed in the instance of this season. While ESD is not the primary focus of the current formula, it’s something the league took into consideration when implementing the scheduling method. While having subservient divisions may not be detrimental to individual and team performance, it still costs the league an unknown amount of dollars in travel expenses.
Enacting Proposal #3 would cut each team’s ESD by at least 10,000 miles per year, shorten the season from 82 to 56 games, and level the travel distance bias. This proposal would use Goldsberry’s “Sweet 16” playoff method, ensuring that the best teams compete in the postseason. Because there are less games, each win becomes more valuable whilst every defeat is more significant — giving the NBA’s regular season a greater, as Strauss defines it, “meaning.”
Data Visualizations by Matt D’Anna unless noted otherwise.