What email address or phone number would you like to use to sign in to Docs.com?
If you already have an account that you use with Office or other Microsoft services, enter it here.
Or sign in with:
Signing in allows you to download and like content, and it provides the authors analytical data about your interactions with their content.
Embed code for: NBA Player Collisions 2016-17
Select a size
Who wins and who loses collisions in NBA games.
Figure 1: Outcomes for Defenders with Most CollisionsPlayer Collisions 2016-17
Ignoring screens and off the ball collisions, many critical plays in NBA games involve contact, even collisions, between the ball handler and a defender. Regarding collisions, the question has been
http://fivethirtyeight.com/features/the-nba-player-whos-so-good-at-taking-charges-we-created-a-stat-for-him/?src=obsidebar=sb_1posed before – what players do well or poorly on defensive side of collision – and a few players who did well since 2014-15 listed in an article by
http://fivethirtyeight.com/contributors/chris-herring/Chris Herring. While I am familiar with data sources mentioned, I am not sure exactly how the author defined and compiled the numbers published.
Despite my uncertainty, I thought I would try to address the situation – that is collisions – in a broader way addressing both offense and defense. While most fouls involve some kind of physical contact, I was interested in collisions and not in mere leaning, grapping, or shoving. The collisions that I investigated had one of two outcomes – a charging foul called on the offensive player or a blocking foul on the defensive player. As at different times – and different collisions – the same player plays either offense or defense, a player’s separate success rates on both sides of the ball are of interest. Indeed, they can be quite different. In addition to considering them separately, I also combined them to see if players existed that were success on both sides of the ball and to what extent. In addition, I took a brief look at collisions at the team level.
This piece is both an extension of my
https://doc.co/hnJvg6/J8ryRdwork on offensive fouls (thanks for the many hits) in that it considers both success and failure for players on both sides of the ball and narrower in that it includes only charge-or-blocking-foul collisions rather than all offensive fouls.
I used data from the 2016-17 regular season extended through April 24. Of the 441 players who participated in these charge-or-block collisions, 413 participated in collisions as a defender, and 115 participated in 10 or more collisions as a defender. Of these 115, 53 (46%) won half or more of their collisions – that is they drew as many or more charges as they committed blocking fouls – the definition of success for a defender.
This did not sound as bad as I expected since charges seemed hard to come by, to me. Indeed, in the aggregate my impression was closer to accurate. Of the almost 3300 collisions, 40.7% were successes for defenders – charging calls. This, of course, means the offense won 59.3% of these collisions by drawing a blocking foul on the defender. Sixty percent of blocking fouls called were shooting blocking fouls with free throws as opposed to simple personal blocking fouls.
Detailed data for all players is in the Annex. It also includes percentiles.
Players on Defense
Figure 1 (first page) shows the numbers of successes and failures for players who participated in the most collisions as defenders. Note that some such as Kemba Walker and DeMarus Cousins have relatively high success rates at drawing charges (blue), and others such as Marquise Chris and Giannis Antetokounmpo commit a high percentage of blocking fouls (orange) from their collisions as defenders.
So, what defenders have high percentage success rates resulting from the hits they take. Figure 2 defenders with 50% or higher success at drawing chares from their collisions.
Ish Smith (8 of 10, 80%), Aaron Brooks (10 of 13, 76.9%), Tim Frazier (10 of 13, 76.9%), and Marcus Smart (23 of 30, 76.7%) head the 2016-17 list (minimum 10 collisions).
Figure 2: Successful Defenders in CollisionsAnthony Tolliver, who was singled out in
http://fivethirtyeight.com/features/the-nba-player-whos-so-good-at-taking-charges-we-created-a-stat-for-him/?src=obsidebar=sb_1Herring’s article for multi-year performance in this area finished tied for fifth with Draymond Green (both 13 of 17, 76.5%) on the 2016-17 season’s list. DeMarcus Cousins (24 of 32, 75%) was sixth.
Relevant considerations go beyond how many charges were drawn and blocking fouls committed. Blocking fouls can be shooting blocking fouls with free throws or simply personal fouls involving blocking. For example, Marreese Speights looks good in his collision success rate drawing 36 charges to committing 26 blocking fouls, but of his 26 blocking fouls 21 were shooting blocking fouls with free throws for the offensive player. He greatly exceeds the 60% average of blocking fouls that are shooting blocking fouls. While this may be indicative of a problem, one must beware of the small sample size and the context in which both drawing charges and commiting blocking fouls occur, particularly location.
In aggregate, this bias towards shooting blocking fouls is not a major problem as only 17 players committed 9 or more such fouls in 2016-17. However, to each team, it is yet another factor that tends toward losing games.
Some sets of videos are more awkward to assemble than others, but click on their names for the blocking fouls by the player:
http://stats.nba.com/events/Giannis Antetokounmpo, and
Players on Offense
On offense, 394 players had charge-or-block collisions. Of these, 116 had 10 or more collisions. The 50 with 17 or more collisions while on offense are shown in Figure 3. Once again, blue is good, but on offense good is a blocking foul. Orange is for charges committed, a bad thing on offense, of course.
A slightly fewer number of offensive players had 17 collisions than the number on defense (49 to 53). However, the difference is not enough to justify saying, “Offensive players can run into as many players as they chose with some hope of a positive outcome while defensive players have little or no chance of a positive outcome if they run into an offensive player – particularly one with the ball.” I have already mentioned that the offense won 59.3%, say 60% (to two significant digits), so maybe this statement is a reasonable partial description for what causes the bias towards offense in collision calls. (I almost wrote, bang-bang collision calls.)
Westbrook and the players listed at the top of Figure 3 had the most collisions while on offense. The names are not surprising, but maybe, Harden being 17th is.
Figure 3: Collision Outcomes for Players on Offense
Figure 4: Offensive Collision Success PercetageAs can be seen from Figure 3, some offensive players do quite well at not committing charges – for example, Isaiah Thomas, Kyle Lowery, De vin Harris, Chris Paul, and Kemba Walker. Looking at the offensive collision success percentages for players with at least 10 collisions on offense, I found 44 had been successful two-thirds of the time or more. This is in striking contrast with players on defense only 15 of whom were successful two-thirds of the time or more.
Videos of collisions for particular players on offense are available but sometimes hard to sort out. However, for example click their name for video of the charging fouls committed by
http://stats.nba.com/events/Ish Smith, and
My first thought was that the offensive and defensive players in a charge-or-block collision face such different situations that looking at all collisions of a player might not be meaningful. However, I later decided that the number of collisions a player was involved in could be an interesting indicator of style of play, and possibly I would find other aspects of interest.
The players with the most collisions for the season was Eric Bledsoe. His frequency of collisions was higher than his 77 collision might lead one to believe as he played in only 66 games, but he did play 2176 minutes. Thus, he participated in 1.167 collisions per game played with a collision every 28 minutes of play on average. Giannis Antetokounmpo also had 77 collisions. However, he played in 80 games and for 2845 minutes (35.6 min/G). Thus, his frequency of collisions was significantly lower as judged against the amount of time he played.
The players who are considered the NBA’s best were spread out some, but three were near the top. Russel Westbrook was fifth, Lebron James was 16th, and James Harden was 22nd. Kawhi Leonard was 60th with 28 charge-or-block collisions, Kevin Durant was 73rd with 26 (missed several games), and Stephen Curry was 118 with 20 collisions.
In my initial look beyond who collided the most, I saw several items pointing in potentially fruitful directions. Nine of the top 15 are littles and 6 are bigs. Altogether more littles than bigs appear in this list of the top 49. Almost all of the players with 40 or more combined collisions have a close to 50% or higher success rate.
Figure 5:Player Combined Defensive and Offensive Collisions and SuccessSome teams have several players on the list of high-collision players – for example, MIL, PHX, DAL, and MIN. A few teams have no players on the list – for example, SAS and GSW.
As mentioned at the beginning, my definition of a collisions is an event that results in of one of two fouls – a charging call or a blocking foul. At different times a participating player is playing on each of the two sides, offense or defense. Thus, from a particular player's viewpoint four events exist. Figure 6 uses these four categories of outcomes to show how each player with 32 or more collisions performed. The outcomes associated with a player being on defense are surrounded by a reddish glow-fringe.
Charges drawn (defense) are blue and charges committed (offense) are in gray. Blocking fouls (defense) are orange with a boundary and blocking fouls drawn (offense), so both sides of blocking fouls have a boundary around their portions of the bar.
The glow-fringe allows one to easily see which players do their collisions mainly on defense such as Speights, Illyasova, and Dieng. Others do their hitting while on offense such as Westbrook, Wall, Wiggins, and James.
The long early blue portions of the bars of several players indicates the players prowess in taking charges. Long portions of yellow-orange at the right ends of bars indicates collisions drawing blocking fouls – a good result for the offensive portion of play of the player with which a bar is associated.
The Annex has a table with even more detailed numbers for all 441 players who participated in collisions. Thus, if you want be more precise, Figure 6 data is available.
As we have examined the players with the most collisions, we have seen many recognizable names. Interestingly, no GSW or SAS players are in the top group for total collisions, but some were on the top offense and defensive volume of collision charts. Kawhi Leonard was mid-range on the offensive volume of collisions chart (Figure 3), and Patty Mills, Manu Ginobili, and Draymond Green were in the lower half of the defensive collision volume chart (Figure 1). However, several GSW and SAS players were included in the charts for highest percentages of success.
Figure 6: High Collision Players' Drawing and Committing Charges and Blocking FoulsSo, what is the relationship between number of collisions and winning them, and winning games? Before I address this more directly, first let's look at the numbers for each team during the 2016-17 regular season.
Figure 7: Collisions by TeamTeams
I have included one additional aspect to Figure 7 not in the prior Figures – the division of blocking fouls into personal blocking fouls and shooting blocking fouls. The latter, shooting block fouls, awards mandatory free throws to the offense while a simple personal blocking foul only results in free throws if the defense is in the penalty, a result of committing four prior fouls.
Since I separated blocking fouls into personal and shooting, I decided to indicate their both being blocking fouls by putting black borders around them as I had in prior chart.
I have included the numbers of each event in Figure 7. For example, DAL, which had the second highest number of collisions, drew 84 charges, committed 61 personal blocking fouls, 73 shooting blocking fouls, 5 charge fouls, drew40 personal blocking fouls and 61 shooting blocking fouls. PHX had the most collisions.
The final chart showing similar information but with the teams ordered by winning percentage is on the next page. It allows one to see if any obvious relationships exist with winning.
Ordering the teams by winning percentage as I have done in Figure 8 does not immediately make any relationships with winning obvious.
Running correlations at the team level with winning percentage for each of the four outcomes of collisions was of limited help as you can see in Table 1.
Table 1: Correlations with Winning Percentage
Correlations with Win%
Commit Blocking Foul (Personal)
Commit Bling foul (Shooting)
Draw Personal Blocking Foul
Draw Shooting Blocking Foul
The last two correlations with drawing blocking fouls are not even in the expected direction. However, this may not be the expectation of all teams because the teams with the best winning percentages, GSW and SAS, have few of them, indeed few collisions altogether.
Possessions by possessions and player by player the outcomes of collisions are significant. Some players are vastly superior to overs in winning collisions some on defense and some on offense as well as a few on both.
At the team level, differing team strategies and confounding factors lead to results that were mainly inconclusive and sometimes inconclusive but counterintuitive. Nevertheless, I found the results for teams interesting – though not as interesting as those for individuals.
Table 2: Player Statistics for Collisions
Player in Collision
Charge Drawer total (Defense)
Blocking Foul Personal (Defense)
Blocking Foul Shooting (Defense)
Blocking Foul Total (Defense)
Total Collisions as Defender
Defender Collision Success Percentage
Drawer BF Personal (Offense)
DBF Shooting (Offense)
Drawer BF Total (Offense)
Charge Committer Total (Offense)
Total Collisions on Offense
Offense Collision Success Percentage
Total Successful Collisions
Percentile Total Collisions
Total Collision Percentage
Percentile Collision Percentage
Derrick Jones, Jr.
Gary Payton II
Glenn Robinson III
James Ennis III
James Michael McAdoo
Kelly Oubre Jr.
Larry Nance Jr.
Luc Mbah a Moute
Metta World Peace
Otto Porter Jr.
Tim Hardaway Jr.
Wade Baldwin IV