Reaching for Rhythm
Two weeks ago on The Disciples of Clyde Podcast, Bethlehem Shoals and Dan Filowitz spent some time discussing the occasional, infuriating phenomenon of players trying to shoot their way into a rhythm from the outside. The discussion focused on LeBron James and Kobe Bryant, and with their various marked advantages scoring inside, how frustrating it can be to watch them start a game seemingly determined to find a rhythm by shooting 18 ft. jumpers. Towards the end of the podcast they wondered whether there was any data available to track the frequency of this occurence, and if not there was a casual challenge for a listener to put some together. I should note that I don’t have a Synergy Sports membership so it’s possible that this data is available there, making many hours of work on my part completely redundant. That being said, here is my answer to their challenge.
The questions I focused on were these: Do certain players have discernible shot patterns to start a game? How do those differ from their shot patterns for the rest of the game?
To create a sample I began by putting together a list of the Top 20 players from last season in terms of Field Goal Attempts per game. I then used Hoopdata‘s player pages to calculate what percentage of their Overall Field Goal Attempts came from each area of the floor. For their game starting shot patterns I went through and looked at the first 4 Field Goal Attempts of each game for each player during the 2009-2010, what location those shots came from, and turned that into a percentage.
There are admittedly some flaws with the way I choose to gather the data. I did not include Free Throw Attempts, which are an indicator of attacking the rim. Despite being high volume shooters, not all of these players started the game as their team’s primary scoring option. For example Aaron Brooks and Derrick Rose often didn’t take their 4th shot attempt until the 2nd Quarter. However, the technique I used was the easiest and most consistent way for me to assemble this information. While there is more data which factors into this equation, what I have here certainly has a story to tell. Below is a color-coded table of the percentage of each player’s shots which came from each area of the floor.
This is a lot of data and not the easiest for comparing patterns, so let’s take these numbers and put them into individual charts. For space reasons, I only included the individual charts for LeBron and Kobe. To see the individual charts for the rest of the Top 20 click here. The blue line on each table represents a player’s Overall Shot Percentage for each area. The red line represents the Shot Percentages on each player’s 1st 4 shots.
Here it’s much easier to see the patterns emerging. Kobe appears to have a slight tendency to begin games relying on his mid-range jumpshot. LeBron on the other end appears to hold back his 3PT attempts and take more shots At the Rim to start the game. LeBron’s graph also illustrates the rather large dichotomy in his shot selection. As Shoals and Filowitz alluded to, he doesn’t have a reliable floater and doesn’t seem as comfortable with true mid-range jumpshots. More than 80% of his shots are either taken at the rim, or are long jump shots. Kobe has a much more regular distribution of shots from different areas. (The scales are different for each graph which makes Kobe’s look slightly more irregular.) I don’t know that I will have the time to put it together but I would be very interested to see if similar graphs from the beginning of Kobe’s career would look more similar to LeBron’s.
To look at the data for all 20 players at once, I took each player’s 1st 4 Shot Percentages for each area and subtracted their Overall Shot Percentages. A red value means a player was less likely to shoot from that area at the beginning of the game than overall. A green value means a player was more likely to shoot from that area at the beginning of the game than overall.
Here you can see that only 4 of the 20 players, Tyreke Evans, Dirk Nowitzki, Dwayne Wade and LeBron James, seemed inclined to begin a game by getting to the rim more than they normally would. On the other hand 18 out of 20 began a game by shooting more mid-range jumpers, and 11 of 20 had an increase in long jumpers to start the game. Even primarily post players David Lee, Chris Kaman and Zach Randolph increased their long jumper percentages at the beginning of a game. Across this sample, the occurrence of players starting a game trying to establish a rhythm from the outside seems extremely prevalent. The strange thing is that LeBron and Kobe, the two players who sparked this discussion, seem to fit the pattern less than most.
To cycle back to the case of LeBron and Kobe, the phenomenon of them shooting long jumpers to begin a game seems as much an issue of perception as anything else. Kobe sees an increase in the percentage of shots he takes from the 16-23ft. range to start a game, but it is a modest increase when compared to players like Brandon Roy, Zach Randolph, Kevin Durant and Monta Ellis. The percentages show that LeBron is actually more likely to attack the rim at the beginning of the game. In going through the play-by-play data I did find 5 or 6 games where LeBron started off with 4 straight long jump shots. Kobe fell into the same trap on occasion. Even though these games are outliers for both players, they stand out because they seem to be such a blatant under-utilization of their skills. These players are so effective at the rim, LeBron off penetration and Kobe in post-ups, that when they neglect these advantages it stands out that much more.
In going through this information I wondered several times whether these same patterns hold true for the playoffs as well. If this NBA postseason has taught us anything it’s that regular season samples can be an extremely poor indicator of what will happen in the playoffs. I will be working on those numbers and hopefully will have them together by the end of next week.