Measuring Player Importance
Points, rebounds, assists. A player is generally evaluated based on these three stats. And this is really not a bad method for being able to tell how good a player is — guys like LeBron James and Russell Westbrook, who consistently fill up all three categories on the stat sheet, are two of the best players in the game.
However, the player with the most points, assists, or rebounds is not necessarily the most important player to his team. Look no further than the playoffs of the 2014–15 season. Stephen Curry led the Golden State Warriors in both points (28.3) and assists (6.4) en route to an NBA title. Yet, it was Draymond Green, not Curry, who was the Warriors’ most “important” player during that year’s playoffs.
Referencing the Warriors’ plus-minus ratings when Green and Curry were either on or off the court can help explain why Green was more “important” than Curry in the 2015 playoffs. Per minute, the Warriors outscored opponents by 0.25 points whenever Green was on the court, and 0.20 points whenever Curry was on the court. Conversely, the Warriors got outscored by 0.12 points per minute when Green was off the floor; whereas with Curry out, the Warriors were able to outscore opponents by a very small margin (0.01 points per minute).
So, Green had a higher on court plus-minus than Curry and a lower off court plus-minus, which indicates he was the more “important” player. Simply subtracting the two numbers — on court plus-minus and off court plus-minus — gives a measure for a player’s overall impact on a per minute basis. For Green, this comes out to 0.25 - (-0.12) = 0.37, and 0.20 - 0.01 = 0.19 for Curry. This measure is often referred to as Roland Score. To account for differences in time spent on the court, the player’s Roland Score is multiplied by the number of minutes he averages per game, resulting in the final “net value” he contributes to his team.
The 20 players who were most important to their team during the 2015 playoffs are listed in the table below.
Again, it’s important to reiterate that a higher Roland Score or net value does not mean one player is better than the other. What it means is that a particular player was more vital to his team’s success. While Curry’s ability to space the floor drives the Warriors system, it can be inferred from the net values that it was Green’s unique capabilities of serving as a floor-general and guarding multiple positions which more significantly contributed to the team’s success.
Use Cases
There are several applications for this method; two examples are provided.
James Struggles in 2011
LeBron James had one of the biggest low points of his career in the 2011 NBA Finals, as he had reached the Finals in just his first year with the Miami Heat but failed to make much of an impression in the critical moments. His net value for the 2011 playoffs was slightly above -11, indicating that the team performed better when he was off the court. That is quite a revelation, considering he is the best player in the world.
However, there was a reason for James’ ineffectiveness. In Miami, he now had another “first option” in Dwyane Wade with whom he could share scoring and playmaking duties. This basically threw James off his game, as he was accustomed to having the offense flow through him in Cleveland. The net values indicate as such, and could have helped the Heat to understand that the offense needed to run through James for the team to be more successful. Not surprisingly, this is what the Heat ended up doing, which led them to win two consecutive NBA titles in 2012 and 2013.
Warriors Shifting System
Kevin Durant has been with the Golden State Warriors for only two years, and they have already won two NBA titles. In the 2017 playoffs, the Warriors demolished their opponents, finishing their postseason run with a 16 -1 record. However, the 2018 playoffs was a different story. For the first time since Durant joined the team, they seemed vulnerable, as the Houston Rockets were able to push them to seven games in the Western Conference Finals. The Warriors triumphed in the end, but if not for Chris Paul’s injury late into the series, who knows what the result could have been. Interestingly enough, the net values of the Warriors’ players in the playoffs changed significantly from 2017 to 2018.
Immediately from the two tables, what sticks out is that the net values of some of the Warriors’ key players were a lot lower in the 2018 playoffs than in the 2017 playoffs, which is to be expected as the Warriors were much more dominant in the playoffs in 2017. Stephen Curry, Draymond Green, and Klay Thompson, who led the Warriors in net value in 2017, all saw major drops in their net value in 2018. It was Kevin Durant who had the highest net value in 2018, almost doubling his value from the previous year. This could possibly be due to the Warriors’ offense revolving more around Durant in 2018. In fact, many were quick to point out that Durant seemed to be letting the ball stick in his hands and playing too much “iso-ball” which contrast with the Warriors’ typical strategies of ball movement and player movement. This style of play potentially may have hindered the Warriors’ effectiveness as seen by the heavy drops in Green, Curry, and Thompson’s net values and the team’s struggles against the Rockets. For the Warriors to return to their 2017 playoff dominance, it may be in their best interest to continue running a system that is more conducive to the play styles of Curry, Green, and Thompson.
Limitations
To accurately determine how important a player is to his team is difficult. This is because a player’s importance can be influenced by other players on his team, especially when defining “importance” as the extent to which the player’s team outscores the other team when he is on and off the court.
The earlier example of Draymond Green having a higher net value in the 2015 playoffs than Steph Curry (despite Curry having led the team in points and assists per game) can be used to show where this metrics falls short. It was concluded that Green was the more “important” player because he had a higher net value than Curry (13.8 to 7.5). However, would the Warriors still outscore opponents by a margin of 0.25 points per minute with Green on the court if Curry did not play at all? In other words, it is possible that the reason Green had such a high on court plus-minus to begin with is because he played most of his minutes alongside Curry. Curry’s presence on the court may have helped Green to be as effective as he was in those playoffs. And the same could hold true for Curry in that he needs Green on the floor to have a greater impact on the team.
So, player “importance” in this case is almost never mutually exclusive. Only if a player were to play on several different teams with different teammates and within different systems would the bias of other players’ impacting his net value be removed. This scenario applies to very few players; however, looking at players’ net value over multiple years could help to remove some of this bias. The table below shows cumulative net values for players who have played at least 1,000 minutes in playoff games over the span of the last 11 seasons (since 2007–08).
