How will Rajon Rondo impact the Chicago Bulls' offense?

When the Chicago Bulls brought in Fred Hoiberg during the 2015 summer, many expected to see a lackluster Bulls offense under Tom Thibodeau transform into a more aggressive offensive attack. Well, stealing a line from Kurt Russell’s character, Jack Burton, in Big Trouble in Little China, 82 games later, the check is still in the mail for the Bulls. After averaging an offensive rating of 101.7 during the five season reign of Thibodeau with an average NBA rank of 17th, the Hoiberg Bulls topped out at a rating of 101.5 this past season, 22nd in the NBA or figuratively described as riding in the back of the bus, just close enough to the front where you can’t smell the fumes.

Much of the lack of success on offense can be attributed to injuries, but an underlying theme that has remained consistent since 2011 has been the instability at the primary ball-handling position. Since Derrick Rose went down with an ACL injury in 2012, the Bulls have now shuffled through 8 different point guards, none lasting longer than two seasons, other than Kirk Hinrich and the often injured Rose.

The Bulls spent this offseason overhauling a roster and brought in eight new faces, including veteran Dwyane Wade. But none may have more impact on their offense than the signing of Rajon Rondo as Rose’s replacement. While Rose offered more of a scoring threat, Rondo will be relied upon for his ability to create scoring opportunities for teammates. With the offense now flowing through Rondo, the biggest question that will determine the Bulls’ offensive success is whether Rondo’s playmaking abilities can translate into team offensive success and, if so, where can we expect it most?

The benefits of assists

More from Nylon Calculus

Before we can begin to measure this in relation to the Bulls, it is first important to consider why an assist should even be valued, as assists are the most tangible evidence of Rondo’s offensive impact. To illustrate this issue, imagine two players in a game both shoot 5/10 from the field and score 10 points. Player A, an iso-dominant offensive scorer (LeBron James), scores all five baskets without an assist while player B, a post player (DeAndre Jordan) scores all five of his baskets off assists. Whether a shot was assisted or not, the number of points did not change. Simply put, an assist describes just another way a team scores points.

As Seth Partnow has discussed in his previous work about playmaking, teams see a substantial boost in their shooting efficiency between assisted and unassisted shots. During the 2013-2014 season, Seth measured the league average improvement in shooting efficiency between assisted versus unassisted shooting at 15.5 percent. Using the same framework, that difference grew to 17.8 percent during this past season. So, when Rondo joined a Kings’ roster last season that previously had few playmakers and produced an assisted eFG% of 57.6 percent, it wasn’t surprising to see not just their AST% (percent of made shots that are assisted) increase by about 6 percent, but also their assisted eFG% increase by about 2 percent as well.

Measuring the likelihood of an assist by made shot

While we’re not reinventing the wheel here by asserting teams that share the ball tend to play better on offense, we will need to take a creative approach to understand the impact of an assist at the player level. Currently, there is little, if any, public information about the relationship between the passer and the shooter on a potential assist. However, we can begin to map this relationship by first looking at the likelihood of an assist for a made shot by location using a similar method to the one I developed to measure Russell Westbrook’s impact on the likelihood of an offensive rebound.

By breaking up the court into 12 distinct regions, we can measure the probability that a shot in one of these regions was assisted on for a given player while he is on the court. The resultant probability assumes the player is playing with average teammates and opponents, allowing for equal cross-comparison of players. Below is the league average this past season by shot region:

For our purposes, there are two key benefits to look at assists this way that helps us understand better a player like Rondo:

1) We can understand the location of shots most likely to be assisted by a player while he is on the court.

2) We can overly a teammate’s shot-chart to understand how a player’s influence on an assist corresponds to his teammate’s relative shooting tendencies.

The value of the second point is that it helps us begin to visualize how a player’s relative playmaking skills mesh with his teammates shooting tendencies. Ideally, an assist doesn’t just create an open shot for a player, but creates it in his proverbial sweet spot. To put this in context with the model, say, for instance, player A is a prolific shooter from only the left-wing three-point region and his teammate B has an increased probability that an assist occurs from this region. If these two players share the court and A makes a shot from this region, there is a greater chance it was assisted, what we’ve established suggests that a more valuable shot was created in A’s most frequent shooting location. Something a basketball chef would call a recipe for success.

Visualizing Rondo’s impact on the Bulls

The biggest concern coming out of the Bulls’ offseason roster shuffle was the lack of stretch players that could provide a three-point threat and help spread the court for Butler’s and Wade’s drives to the basket. As a result, much of their outside shooting will rest on the shoulders of Doug McDermott and Nikola Mirotic. So, what will this look like with respect to his new Bulls’ teammates?

We can visualize Rondo’s relationship to these shooters by overlaying McDermott and Mirotic’s shot charts with a color gradient that describes Rondo’s impact that shots made from these regions are assisted compared to the league average. The shot chart identifies where McDermott and Mirotic shoot on the court and highlights the frequency of the shots by the size of the dot. For McDermott, this is primarily from the right-wing and straight-away while Mioritic is best from the left-corner and straight-away. Given the boost of an assisted shot over an unassisted shot, we are most concerned about the frequency of their shots, not necessarily the efficiency compared to Rondo’s impact on assists in these regions.

With visual context, it’s clear that although Rondo is an assist machine, his impact doesn’t necessarily match up perfectly with McDermott and Mirotic, which is specifically noticeable in either corner three; two regions where McDermott and Mirotic shoot often. This is to say that McDermott and Mirotic made shots from these regions will not have an increased likelihood of being assisted — corresponding to a more effective shot — while Rondo is on the court, something our basketball chef from earlier doesn’t enjoy.

However, Rondo’s impact will be greatest from straight-away three-point shots as he provides a 4.07 percent increase above the league average that made shots that are assisted from this region, which matches well with one of McDermott’s more frequented and successful shot regions. In addition, Rondo also adds a 1.25 percent boost to the likelihood that shots made from the left-wing three-point are assisted, which also matches well with one of Mirotic’s more frequented and better shooting regions. In both cases, Rondo creates about a 1.70 percent and 0.50 percent increase that shots from straight-away and the left-wing respectively are assisted, compared to Rose. The relationship between Rondo and other Bulls’ players can be further explored in my R-Shiny application of the model that yields similar results.

Much of Rondo’s increased impact on the likelihood of an assist has to do with how he spends his time with the ball during each possession. When looking at the amount of time Rondo controls the ball per possession in comparison to the total (non-shot region specific) likelihood of an assist, this characteristic is certainly noticeable and many will point out his eagerness to seek out assists falls on the side of greediness and is done at the expense of his team.

However, assists provide a substantial boost in shooting efficiency and Rondo’s roughly 6 percent increase over Rose should be a breath of fresh air for the Bulls who watched Rose spend a comparable amount of time with the ball, but resulted in one of the lowest probabilities of an assist among all points guards in the league. With their previous roster compositions, the Bulls couldn’t afford this from Rose as a primary ball-handler and could partially explain the Bulls’ offensive struggles and potentially why McDermott specifically has struggled.

Heading into the 2016-2017 season, the Bulls built their offensive focal points around two players in Butler and Wade who excel in slashing towards the basket, but failed to complement them with an abundance of perimeter players that could provide spacing and prevent defenses collapsing on those drives. Adding Rondo provides one way to counter defenses from cheating on these drives because of his ability to create shots for his teammates.

While Rondo won’t provide a complete improvement in every key shooting region for the Bulls’ primary perimeter shooters, he should provide an immediate boost over Rose and increase the likelihood of assisting McDermott and Mirotic in their three-points shots from some key regions while they share the court, increasing those shots’ efficiency. That should provide enough impact for McDermott and Mirotic to at least make defenses think twice before collapsing on interior drives by Butler or Wade.