10
1 NBA Comeback Clutchness: Real or Just Noise? Sandy Huang and Toa Lohe Introduction In his last NBA game, Kobe Bryant famously scored 60 points against the Utah Jazz to lead the Los Angeles Lakers to a 101-96 victory. The game was entertaining not just because Kobe scored lots of points but because of how he scored those points. Let us set the scene. With about three minutes remaining in the 4 th quarter, the Lakers were down 10 points, and it seemed that Kobe would likely end his career with a loss despite having already scored 45 points. He, however, surprised us all by proceeding to score 15 straight points for the Lakers, a brilliant capstone performance to end a legendary 20-year career in the NBA. 1 At the end of the night, Kobe reminded everyone of why many of us watch sports in the first place: to feel the emotional suspense of seeing people struggle againstand sometimes even overcomeseemingly improbable odds. 2 This ability to overcome seemingly improbable odds is what we often nebulously call clutchness. Inspired by Kobe’s last game, our final project quantified the clutchness of NBA players. In particular, we were interested in situations in which a player’s team was either behind by 5 points (or less) or tied (but not ahead) in the last 5 minutes of the 4 th quarter (or overtime). Let us refer to this particular subset of clutchness as comeback clutchness. In contrast to comeback clutchness is escape clutchness—when a player’s team is either ahead by 5 points (or less) or tied (but not ahead) in the last 5 minutes of the 4 th quarter (or overtime). We made this particular distinction between comeback clutchness and escape clutchness because of both our assumption that there is a psychological difference between these two scenarios that significantly impacts how players perform and the fact the NBA does not provide data on situations where teams are only behind/ahead and not tied. Finally, one of the main aims of our project is to find out whether special comeback clutchness performances (e.g. Kobe’s last game) can be repeated or if their manifestations are just noise (i.e. random). Background Previous work on NBA clutchness has been done by 82games 3 and Liberty Ballers 4 (a SB Nation blog for the Philadelphia 76ers). The first study done by 82games focused on the most clutch players in terms of field goals attempted, points scored, effective field goal percentage, foul drawing ability, and free throw shooting. The second study done by 82games focused on plus- minus, win-loss record, rebounding, passing, and shot creation (based on unassisted field goals made per 48 minutes). The third study done by 82games focused on the difference between non- clutch and clutch moments, PER (Player Efficiency Rating), and opposing player PER. Lastly, the study done by Liberty Ballers focused on clutch shooting. None of these four studies, however, focused on comeback clutchness. 1 For the sake of Kobe’s mythology, let’s ignore the officiating shenanigans at the end (e.g. the Laker’s Julius Randle setting hilariously illegal screens on the Jazz’s Gordon Hayward). 2 Seth Partnow of VICE Sports more eloquently defined it as “[t]hat anticipation, the moment just before the roller coaster tips over the crest of the big drop [...] the thrill that keeps us coming back.” https://sports.vice.com/en_au/article/kobe-bryant-majored-in-drama-not-math 3 http://www.82games.com/clutchplayers.htm http://www.82games.com/clutchplay2.htm http://www.82games.com/clutchplay3.htm 4 http://www.libertyballers.com/2012/2/29/2832299/lebron-james-kobe-bryant-dwyane-wade-clutch-nba-playoffs-4th-quarter

NBA Comeback Clutchness: Real or Just Noise?

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: NBA Comeback Clutchness: Real or Just Noise?

1

NBA Comeback Clutchness: Real or Just Noise?

Sandy Huang and Toa Lohe

Introduction

In his last NBA game, Kobe Bryant famously scored 60 points against the Utah Jazz to

lead the Los Angeles Lakers to a 101-96 victory. The game was entertaining not just because Kobe

scored lots of points but because of how he scored those points. Let us set the scene. With about

three minutes remaining in the 4th quarter, the Lakers were down 10 points, and it seemed that

Kobe would likely end his career with a loss despite having already scored 45 points. He, however,

surprised us all by proceeding to score 15 straight points for the Lakers, a brilliant capstone

performance to end a legendary 20-year career in the NBA.1 At the end of the night, Kobe

reminded everyone of why many of us watch sports in the first place: to feel the emotional suspense

of seeing people struggle against—and sometimes even overcome—seemingly improbable odds.2

This ability to overcome seemingly improbable odds is what we often nebulously call clutchness.

Inspired by Kobe’s last game, our final project quantified the clutchness of NBA players.

In particular, we were interested in situations in which a player’s team was either behind by 5

points (or less) or tied (but not ahead) in the last 5 minutes of the 4th quarter (or overtime). Let us

refer to this particular subset of clutchness as comeback clutchness. In contrast to comeback

clutchness is escape clutchness—when a player’s team is either ahead by 5 points (or less) or tied

(but not ahead) in the last 5 minutes of the 4th quarter (or overtime). We made this particular

distinction between comeback clutchness and escape clutchness because of both our assumption

that there is a psychological difference between these two scenarios that significantly impacts how

players perform and the fact the NBA does not provide data on situations where teams are only

behind/ahead and not tied. Finally, one of the main aims of our project is to find out whether special

comeback clutchness performances (e.g. Kobe’s last game) can be repeated or if their

manifestations are just noise (i.e. random).

Background

Previous work on NBA clutchness has been done by 82games3 and Liberty Ballers4 (a SB

Nation blog for the Philadelphia 76ers). The first study done by 82games focused on the most

clutch players in terms of field goals attempted, points scored, effective field goal percentage, foul

drawing ability, and free throw shooting. The second study done by 82games focused on plus-

minus, win-loss record, rebounding, passing, and shot creation (based on unassisted field goals

made per 48 minutes). The third study done by 82games focused on the difference between non-

clutch and clutch moments, PER (Player Efficiency Rating), and opposing player PER. Lastly, the

study done by Liberty Ballers focused on clutch shooting. None of these four studies, however,

focused on comeback clutchness.

1 For the sake of Kobe’s mythology, let’s ignore the officiating shenanigans at the end (e.g. the Laker’s Julius Randle setting

hilariously illegal screens on the Jazz’s Gordon Hayward). 2 Seth Partnow of VICE Sports more eloquently defined it as “[t]hat anticipation, the moment just before the roller coaster tips over

the crest of the big drop [...] the thrill that keeps us coming back.”

https://sports.vice.com/en_au/article/kobe-bryant-majored-in-drama-not-math 3 http://www.82games.com/clutchplayers.htm

http://www.82games.com/clutchplay2.htm

http://www.82games.com/clutchplay3.htm 4 http://www.libertyballers.com/2012/2/29/2832299/lebron-james-kobe-bryant-dwyane-wade-clutch-nba-playoffs-4th-quarter

Page 2: NBA Comeback Clutchness: Real or Just Noise?

2

Methods

By using “Clutch Traditional Stats” 5 and “Clutch Advanced Stats”6 provided by the NBA,

we formulated a comeback clutchness version of Justin Kubatko’s Win Shares statistic (i.e. the

estimated number of wins a player produces for his team).7 Below are all the necessary conditions

for a player’s stats to be included in our analysis:

1) The statistic is accumulated in the last 5 minutes of the 4th quarter (or overtime).

2) The player’s team is either behind by 5 points (or less) or tied (but not ahead).

3) The player has played in at least a total of 50 minutes.

In order to normalize for the total number of minutes played by each player, we also calculated

comeback clutchness Win Shares Per 48.8

To calculate Win Shares, Offensive Win Shares are added to Defensive Win Shares. Below

are the details of calculating Offensive Win Shares and Defensive Win Shares:

Player Points and Defensive Rating are already included in the player data provided by the NBA.9

Additionally, Team Minutes10 and Team Pace11 are already included in the team data provided by

the NBA. Below is a table detailing how to calculate the other statistics:

Statistic Calculation

Player Possessions Player Pace12 ÷ 48 × Minutes Played × Usage Rate13

Team Defensive Possessions Opponent’s Total Points14 ÷ Team Defensive Rating15 × 100

League Points Per Possession Total League Points ÷ Sum of Each Team’s (Pace ÷ 48 × Minutes Played)

League Points Per Game Total League Points ÷ (Total Minutes Played ÷ 48)

League Pace Sum of Each Team’s (Pace ÷ 48 × Minutes Played) ÷ (Total Minutes Played ÷ 48)

5 Comeback clutchness traditional player filter for analysis of 2015-2016 season:

http://stats.nba.com/league/player/#!/clutch/?CF=MIN*GE*50&Season=2015-

16&SeasonType=Regular%20Season&AheadBehind=Behind%20or%20Tied&PerMode=Totals 6 Comeback clutchness advanced player filter for analysis of 2015-2016 season:

http://stats.nba.com/league/player/#!/clutch/advanced/?Season=2015-

16&SeasonType=Regular%20Season&AheadBehind=Behind%20or%20Tied&CF=GP*GE*10 7 http://www.basketball-reference.com/about/ws.html 8 Divide by the total number of minutes played and then multiplying by 48 9 See footnotes 5 and 6 10 Comeback clutchness traditional team filter for analysis of 2015-2016 season:

http://stats.nba.com/league/team/#!/clutch/?Season=2015-

16&SeasonType=Regular%20Season&PerMode=Totals&AheadBehind=Behind%20or%20Tied 11 Comeback clutchness advanced team filter for analysis of 2015-2016 season:

http://stats.nba.com/league/team/#!/clutch/advanced/?Season=2015-

16&SeasonType=Regular%20Season&AheadBehind=Behind%20or%20Tied 12 See footnote 6 13 See footnote 6 14 Comeback clutchness opponent team filter for analysis of 2015-2016 season:

http://stats.nba.com/league/team/#!/clutch/opponent/?Season=2015-

16&SeasonType=Regular%20Season&AheadBehind=Behind%20or%20Tied&PerMode=Totals 15 See footnote 11 and Wikipedia: https://en.wikipedia.org/wiki/Defensive_rating

Page 3: NBA Comeback Clutchness: Real or Just Noise?

3

In addition to this first analysis of comeback clutchness, we also conducted subsequent

analysis on the value of comeback clutchness, the difference between comeback clutchness and

escape clutchness, the difference between comeback clutchness and regular performance, and the

repeatability of comeback clutchness.

The value of comeback clutchness was measured as the differential between comeback

clutchness Win Shares Per 48 and regular season Win Shares Per 48. The difference between

comeback clutchness and escape clutchness was measured by using F-tests and t-tests, and the

same method was used for the difference between comeback clutchness and regular performance.

Finally, the repeatability of comeback clutchness was measured as the year-to-year correlation.

We did these analyses for every season since the 2001-2002 season and also looked at comeback

clutchness plus-minus and PIE (Player Impact Estimate).16

Results

Comeback clutchness for the 2015-2016 season:

Top 10 Comeback Clutchness (WS/48) Bottom 10 Comeback Clutchness (WS/48)

1) Andre Drummond (1.10) 1) Isaiah Canaan (-0.92)

2) Stephen Curry (1.06) 2) Brandon Knight (-0.59)

3) Draymond Green (1.06) 3) Al-Farouq Aminu (-0.45)

4) Klay Thompson (0.90) 4) Mason Plumlee (-0.41)

5) Dirk Nowitzki (0.87) 5) Trevor Booker (-0.33)

6) LeBron James (0.85) 6) Damian Lillard (-0.29)

7) Deron Williams (0.79) 7) Rajan Rondo (-0.28)

8) Chandler Parsons (0.78) 8) PJ Tucker (-0.27)

9) Jeff Green (0.70) 9) Khris Middleton (-0.23)

10) Danilo Gallinari (0.69) 10) Robert Covington (-0.22)

Value of comeback clutchness for the 2015-2016 season:

Top 10 Differential (WS/48) Bottom 10 Differential (WS/48)

1) Andre Drummond (0.97) 1) Isaiah Canaan (-0.95)

2) Draymond Green (0.87) 2) Brandon Knight (-0.64)

3) Klay Thompson (0.76) 3) Mason Plumlee (-0.55)

4) Stephen Curry (0.75) 4) Al-Farouq Aminu (-0.54)

5) Dirk Nowitzki (0.73) 5) Damian Lillard (-0.46)

6) Deron Williams (0.71) 6) Trevor Booker (-0.45)

7) Chandler Parsons (0.67) 7) Rajan Rondo (-0.36)

8) Jeff Green (0.64) 8) Russell Westbrook (-0.36)

9) LeBron James (0.61) 9) Khris Middleton (-0.33)

10) Jeremy Lin (0.60) 10) Serge Ibaka (-0.32)

16 PIE’s explanation can be found here: http://stats.nba.com/help/glossary/

Page 4: NBA Comeback Clutchness: Real or Just Noise?

4

Comeback clutchness for the past 15 seasons:

Season 1 2 3 4 5

2015-2016 Andre Drummond Stephen Curry Draymond Green Klay Thompson Dirk Nowitzki

2014-2015 Pau Gasol Joakim Noah Roy Hibbert Zach Randolph Jimmy Butler

2013-2014 Greivis Vasquez Mike Conley James Harden Dirk Nowitzki Damian Lillard

2012-2013 Chris Bosh Paul Pierce Jeff Green Ray Allen LeBron James

2011-2012 Andrew Bynum Joe Johnson Udonis Haslem Roy Hibbert Zaza Pachulia

2010-2011 Mo Williams Dirk Nowitzki Tyson Chandler Manu Ginobili Serge Ibaka

2009-2010 LeBron James Kobe Bryant Dirk Nowitzki Marc Gasol Chauncey Billups

2008-2009 Pau Gasol Dwight Howard Kobe Bryant Jason Terry Tony Parker

2007-2008 LeBron James Stephen Jackson Chris Paul Amar'e Stoudemire Jeff Green

2006-2007 Josh Howard Dirk Nowitzki Jason Terry Samuel Dalembert Jason Kapono

2005-2006 Keith Bogans Chauncey Billups Chris Kaman Rasheed Wallace Earl Boykins

2004-2005 Ben Gordon Chauncey Billups Tyson Chandler Kirk Hinrich Kenny Thomas

2003-2004 Kobe Bryant Shaquille O'Neal Reggie Miller Metta World Peace Gary Payton

2002-2003 Tony Battie Tony Delk Jason Terry Chauncey Billups Shareef Abdur-Rahim

2001-2002 Raef LaFrentz Ervin Johnson Dirk Nowitzki Ryan Bowen Antonio Davis

Top 5 Players by Comeback Clutchness Win Shares Per 48

Season 1 2 3 4 5

2015-2016 Isaiah Canaan Brandon Knight Al-Farouq Aminu Mason Plumlee Trevor Booker

2014-2015 Tim Hardaway Jr. Jeremy Lin Rudy Gay Jordan Hill Wayne Ellington

2013-2014 Trey Burke Gordon Hayward Ricky Rubio Ben McLemore Zaza Pachulia

2012-2013 Greivis Vasquez Andrea Bargnani Arron Afflalo Jared Dudley Evan Turner

2011-2012 Brandon Jennings Kris Humphries Tayshaun Prince Monta Ellis Brandon Knight

2010-2011 Kirk Hinrich Stephen Curry Dorell Wright Luke Ridnour Monta Ellis

2009-2010 George Hill Tyreke Evans Manu Ginobili Tayshaun Prince Mike Bibby

2008-2009 Jason Richardson Peja Stojakovic Steve Nash Andrea Bargnani Rasheed Wallace

2007-2008 Rashad McCants Tim Thomas Anthony Parker Carlos Delfino Chris Kaman

2006-2007 Dahntay Jones Jarvis Hayes Luther Head Adam Morrison DeShawn Stevenson

2005-2006 Luke Ridnour Donyell Marshall Steve Blake Jumaine Jones Rashard Lewis

2004-2005 Tim Thomas Jarvis Hayes Keith Bogans Tony Parker Carlos Boozer

2003-2004 Bobby Simmons Toni Kukoc Michael Finley Damon Jones Doug Christie

2002-2003 Lamar Odom Corey Maggette Andre Miller Elton Brand Jason Richardson

2001-2002 Bob Sura Bryon Russell Mark Jackson Larry Hughes Erick Dampier

Bottom 5 Players by Comeback Clutchness Win Shares Per 48

Page 5: NBA Comeback Clutchness: Real or Just Noise?

5

Value of comeback clutchness for the past 15 seasons:

Season 1 2 3 4 5

2015-2016 Andre Drummond Draymond Green Klay Thompson Stephen Curry Dirk Nowitzki

2014-2015 Joakim Noah Roy Hibbert Pau Gasol Zach Randolph Jimmy Butler

2013-2014 Greivis Vasquez Mike Conley Taj Gibson Paul Pierce Klay Thompson

2012-2013 Jeff Green Paul Pierce Chris Bosh Ray Allen Avery Bradley

2011-2012 Andrew Bynum Joe Johnson Udonis Haslem Roy Hibbert Zaza Pachulia

2010-2011 Mo Williams Dirk Nowitzki Tyson Chandler Manu Ginobili Jeff Green

2009-2010 Kobe Bryant LeBron James Dirk Nowitzki Marc Gasol Jared Jeffries

2008-2009 Jason Terry Pau Gasol Tony Parker Rashard Lewis Raja Bell

2007-2008 LeBron James Stephen Jackson Jeff Green Baron Davis Nick Collison

2006-2007 Josh Howard Dirk Nowitzki Jason Terry Samuel Dalembert Jason Kapono

2005-2006 Keith Bogans Chris Kaman Earl Boykins Rasheed Wallace Carmelo Anthony

2004-2005 Ben Gordon Tyson Chandler Chauncey Billups Kenny Thomas Kirk Hinrich

2003-2004 Kobe Bryant Shaquille O'Neal Metta World Peace Al Harrington Alvin Williams

2002-2003 Tony Battie Tony Delk Jason Terry Antoine Walker Glenn Robinson

2001-2002 Raef LaFrentz Ervin Johnson Ryan Bowen Antonio Davis Dirk Nowitzki

Top 5 by Comeback Clutchness Win Shares Per 48 Differential

Season 1 2 3 4 5

2015-2016 Isaiah Canaan Brandon Knight Mason Plumlee Al-Farouq Aminu Damian Lillard

2014-2015 Rudy Gay Jeremy Lin Andre Drummond Chris Paul Jordan Hill

2013-2014 Ricky Rubio Gordon Hayward Trey Burke Ramon Sessions Ben McLemore

2012-2013 Greivis Vasquez Ryan Anderson Jared Dudley Blake Griffin John Wall

2011-2012 Brandon Jennings Kris Humphries Tayshaun Prince Monta Ellis Greg Monroe

2010-2011 Kirk Hinrich Stephen Curry Dorell Wright Luke Ridnour Monta Ellis

2009-2010 George Hill Manu Ginobili Tyreke Evans Deron Williams Mike Bibby

2008-2009 Jason Richardson Steve Nash Chris Paul Peja Stojakovic Nate Robinson

2007-2008 Rashad McCants Anthony Parker Carlos Delfino Tim Thomas Chris Kaman

2006-2007 Dahntay Jones Amar'e Stoudemire Luther Head Jarvis Hayes DeShawn Stevenson

2005-2006 Luke Ridnour Donyell Marshall Steve Blake Rashard Lewis Jumaine Jones

2004-2005 Tim Thomas Tony Parker Kobe Bryant Carlos Boozer Jarvis Hayes

2003-2004 Bobby Simmons Toni Kukoc Michael Finley Damon Jones Jason Kidd

2002-2003 Lamar Odom Corey Maggette Elton Brand Andre Miller Gilbert Arenas

2001-2002 Bob Sura Bryon Russell Mark Jackson Andrei Kirilenko Mike Miller

Bottom 5 by Comeback Clutchness Win Shares Per 48 Differential

Page 6: NBA Comeback Clutchness: Real or Just Noise?

6

Difference between comeback clutchness and escape clutchness (includes past 15 seasons):

Since F < (F Critical one-tail), the variances of the two samples are equal, and, since (t Stat) >> (t

Critical two-tail) and the p-value << 0, player performance in comeback clutchness situations is

significantly different from that of escape clutchness situations.

Difference between comeback clutchness and regular performance (includes past 15 seasons):

Since F >> (F Critical one-tail) and the p-value = 0, the variances of the two samples are unequal,

and, since (t Stat) >> (t Critical two-tail) and the p-value ≈ 0, player performance in comeback

clutchness situations is significantly different from that of regular performance situations.

Escape Clutchness Comeback Clutchness

Mean 0.399744153 0.314479098

Variance 0.116108959 0.115400413

Observations 1630 1630

df 1629 1629

F 1.006139892

P(F<=f) one-tail 0.450852491

F Critical one-tail 1.084946995

F-Test Two-Sample for Variances

Escape Clutchness Comeback Clutchness

Mean 0.399744153 0.314479098

Variance 0.116108959 0.115400413

Observations 1630 1630

Pooled Variance 0.115754686

Hypothesized Mean Difference 0

df 3258

t Stat 7.154521473

P(T<=t) one-tail 5.16225E-13

t Critical one-tail 1.645321462

P(T<=t) two-tail 1.03245E-12

t Critical two-tail 1.960692388

t-Test: Two-Sample Assuming Equal Variances

Comeback Clutchness Regular Performance

Mean 0.283649742 0.11589044

Variance 0.120707021 0.002725993

Observations 1862 1862

df 1861 1861

F 44.28001294

P(F<=f) one-tail 0

F Critical one-tail 1.079261758

F-Test Two-Sample for Variances

Comeback Clutchness Regular Performance

Mean 0.283649742 0.11589044

Variance 0.120707021 0.002725993

Observations 1862 1862

Hypothesized Mean Difference 0

df 1945

t Stat 20.60443832

P(T<=t) one-tail 8.74521E-86

t Critical one-tail 1.645637431

P(T<=t) two-tail 1.74904E-85

t Critical two-tail 1.961184408

t-Test: Two-Sample Assuming Unequal Variances

Page 7: NBA Comeback Clutchness: Real or Just Noise?

7

Year-To-Year Correlation using comeback clutchness WS/48:

R2 = 0.02 (no correlation)

Year-to-Year Correlation using comeback clutchness WS/48 differential:

R2 = 0.01 (no correlation)

Page 8: NBA Comeback Clutchness: Real or Just Noise?

8

Discussion

Looking at our results, we seem to have produced some pretty intuitive results—and also

some very counterintuitive results.

For instance, in the 2015-2016 season, Andre Drummond had both the highest comeback

clutchness Win Shares Per 48 (1.10) and the highest comeback clutchness Win Shares Per 48

differential (0.97). To be clear, this means Drummond’s regular season Win Shares Per 48 was

0.13.17 This is a very counterintuitive result given that opposing teams often hack Drummond near

the end of games (i.e. foul him on purpose to send him to the free throw line where he shoots an

abysmal 37.1%). It could be the case, however, that Drummond’s comeback clutchness is so high

because teams who are ahead may be less likely to employ the hacking strategy and thus Pistons

coach Stan Van Gundy is not forced to take Drummond out of the game. It could also be the case

that Drummond is often taken out of the game immediately after the first hacking foul and thus his

comeback clutchness numbers are inflated. Other counterintuitive results include random

appearances in the top five by some very marginal players18 and bottom-five appearances by some

marquee players in their prime like Damian Lillard (2015-2016 season), Chris Paul (2014-2015

season), Manu Ginobili (2009-2010 season), and Kobe Bryant (2004-2005 season).19

In contrast to these counterintuitive results, we produced very intuitive ones like Stephen

Curry, Draymond Green, and Klay Thompson being in the top five for both comeback clutchness

Win Shares Per 48 and comeback clutchness Win Shares Per 48 differentials. Teams like the Dallas

Mavericks who repeatedly won close games were also rewarded in our rankings (e.g. Dirk

Nowitzki, Deron Williams, and Chandler Parsons) while an awful team like the Philadelphia 76ers

was represented by their point guard Isaiah Canaan. Finally, the ever efficient LeBron James

appears in both of our 2015-2016 rankings.

Conclusions

Comeback clutchness is just noise (i.e. rarely repeatable) given its non-existent year-to-

year correlation. It is, however, significantly different from both regular season performance and

escape clutchness. The latter result confirms our assumption that there probably exists a

psychological difference between playing behind and playing ahead that significantly impacts

player performance, especially when it comes to clutch situations.

Extensions of our project include performing the same kinds of analyses but with adjusted

plus-minus and PIE. By using adjusted plus-minus, one could better account for differences in

competition. By using PIE, our hypothesis is that more intuitive results can be gained (i.e. less

random appearances in the top five by very marginal players), at least according to our preliminary

analysis20 of PIE. A downside, however, to using adjusted plus-minus and PIE is that neither

statistic really tells you how many wins a player can produce for their team (although one could

potentially argue that this is also true of Win Shares Per 48).

17 1.10 – 0.97 = 0.13 18 Specifically: Greivis Vasquez, Udonis Haslem, Zaza Pachulia, Jason Kapono, Keith Bogans, Earl Boykins, Kenny Thomas, Tony

Battie, Tony Delk, Raef LaFrentz, Ervin Johnson, Ryan Bowen, Antonio Davis, Jared Jeffries, Nick Collison, and Alvin Williams 19 Along with Steve Nash (2008-2009 season), Rashard Lewis (2005-2006 season), Tony Parker (2004-2005 season), Carlos Boozer

(2004-2005 season), Michael Finley (2003-2004 season), Lamar Odom (2002-2003 season), Elton Brand (2002-2003 season), John

Wall (2012-2013 season), Brandon Jennings (2011-2012 season), Deron Williams (2009-2010 season), and Mike Miller (2001-

2002 season) 20 Looking at the top players for each of the last fifteen seasons

Page 9: NBA Comeback Clutchness: Real or Just Noise?

9

Appendix

Season 1 2 3 4 5

2015-2016 Stephen Curry Kevin Durant Andre Drummond Jeremy Lin Jamal Crawford

2014-2015 Anthony Davis Dirk Nowitzki LaMarcus Aldridge James Harden Danny Green

2013-2014 Mike Conley James Harden Damian Lillard Chris Bosh Iman Shumpert

2012-2013 Dirk Nowitzki Ray Allen Jeff Green Jarrett Jack Jimmy Butler

2011-2012 Andrew Bynum Kyrie Irving Kobe Bryant Nene Dwight Howard

2010-2011 Dirk Nowitzki Mo Williams Eric Gordon Manu Ginobili Jason Richardson

2009-2010 Kobe Bryant LeBron James Joe Johnson Steve Nash Dirk Nowitzki

2008-2009 Zach Randolph Jason Terry Kobe Bryant Tony Parker Paul Pierce

2007-2008 LeBron James Stephen Jackson Manu Ginobili Tim Duncan Vince Carter

2006-2007 Mehmet Okur Zach Randolph Chauncey Billups Jason Kapono Dirk Nowitzki

2005-2006 Jason Terry Earl Boykins Primoz Brezec Chauncey Billups Dwyane Wade

2004-2005 Ray Allen Ben Gordon Earl Boykins Larry Hughes Chauncey Billups

2003-2004 Kevin Garnett Kobe Bryant Yao Ming Metta World Peace Shareef Abdur-Rahim

2002-2003 Shaquille O'Neal Derek Anderson Jason Terry Dirk Nowitzki Chris Webber

2001-2002 Bonzi Wells Dirk Nowitzki Pat Garrity Ray Allen Antoine Walker

Top 5 Players by Comeback Clutchness Offensive Win Shares Per 48

Season 1 2 3 4 5

2015-2016 Draymond Green Andre Drummond Jeff Green Reggie Jackson LeBron James

2014-2015 Derrick Rose Joakim Noah Roy Hibbert Zach Randolph Luc Mbah a Moute

2013-2014 Greivis Vasquez Joakim Noah Kirk Hinrich Taj Gibson Courtney Lee

2012-2013 Chris Bosh Kevin Garnett Paul Pierce Rajon Rondo Avery Bradley

2011-2012 Udonis Haslem Roy Hibbert Zaza Pachulia Jeff Teague Mario Chalmers

2010-2011 Mo Williams Tyson Chandler Jason Kidd Tim Duncan Serge Ibaka

2009-2010 Andrew Bogut Marc Gasol Jared Jeffries Lamar Odom Mike Conley

2008-2009 Hedo Turkoglu Pau Gasol Dwight Howard Rashard Lewis Lamar Odom

2007-2008 Jeff Green Shaquille O'Neal Chris Paul Nick Collison Daniel Gibson

2006-2007 Josh Howard Dirk Nowitzki Jason Terry Monta Ellis Andre Miller

2005-2006 Keith Bogans Chris Kaman Elton Brand Marcus Camby Chauncey Billups

2004-2005 Tyson Chandler Kirk Hinrich Ben Gordon Stephen Jackson Ben Wallace

2003-2004 Shaquille O'Neal Kobe Bryant Al Harrington Reggie Miller Jermaine O'Neal

2002-2003 Tony Battie Tony Delk Antoine Walker Darius Miles Christian Laettner

2001-2002 Ryan Bowen Raef LaFrentz James Posey Ervin Johnson Vin Baker

Top 5 Players by Comeback Clutchness Defensive Win Shares Per 48

Page 10: NBA Comeback Clutchness: Real or Just Noise?

10

Season 1 2 3 4 5

2015-2016 Robert Covington Derrick Rose Khris Middleton Isaiah Canaan Brandon Knight

2014-2015 Derrick Rose Bradley Beal Rudy Gay Michael Carter-Williams Steven Adams

2013-2014 Brandon Jennings Dion Waiters Ramon Sessions Ben McLemore Gerald Green

2012-2013 John Wall Carlos Delfino Jared Dudley Luis Scola Byron Mullens

2011-2012 Brandon Jennings O.J. Mayo Jameer Nelson Nicolas Batum Jason Kidd

2010-2011 Luke Ridnour Vince Carter Tony Parker Brandon Jennings DeMarcus Cousins

2009-2010 Rodney Stuckey Corey Brewer Eric Gordon Jamal Crawford Aaron Brooks

2008-2009 Metta World Peace Rasheed Wallace Baron Davis LaMarcus Aldridge Jason Richardson

2007-2008 Kirk Hinrich Gerald Wallace Antonio McDyess Chris Kaman Dwyane Wade

2006-2007 Luther Head Randy Foye Jason Kidd Hedo Turkoglu Jameer Nelson

2005-2006 Jumaine Jones Anthony Johnson Mark Blount Raef LaFrentz Deron Williams

2004-2005 Keith Bogans Kobe Bryant David Wesley Tony Delk Speedy Claxton

2003-2004 Allen Iverson Toni Kukoc Michael Finley Eric Williams Jamal Crawford

2002-2003 Darius Miles Larry Hughes Scottie Pippen Troy Hudson Kenyon Martin

2001-2002 LaPhonso Ellis Bryon Russell Derrick Coleman Baron Davis Terry Porter

Bottom 5 Players by Comeback Clutchness Offensive Win Shares Per 48

Season 1 2 3 4 5

2015-2016 Isaiah Canaan Mason Plumlee CJ McCollum Damian Lillard Brandon Knight

2014-2015 Jeremy Lin Tim Hardaway Jr. Jordan Hill James Harden Marcin Gortat

2013-2014 Trey Burke Gordon Hayward Corey Brewer Ricky Rubio Zaza Pachulia

2012-2013 Ben Gordon Ryan Anderson Chris Paul Jamal Crawford Anthony Davis

2011-2012 Kris Humphries Monta Ellis Greg Monroe Brandon Jennings Brandon Knight

2010-2011 Kirk Hinrich Stephen Curry Dorell Wright Monta Ellis DeMar DeRozan

2009-2010 Tyreke Evans Mike Bibby George Hill Jason Richardson Beno Udrih

2008-2009 Jason Richardson Steve Nash Chris Paul Peja Stojakovic Rasual Butler

2007-2008 Rashad McCants Anthony Parker Carlos Delfino Ryan Gomes Al Jefferson

2006-2007 Amar'e Stoudemire LaMarcus Aldridge Dahntay Jones DeShawn Stevenson Jarvis Hayes

2005-2006 Donyell Marshall Rashard Lewis P.J. Brown Luke Ridnour Steve Blake

2004-2005 Tim Thomas Carlos Boozer Manu Ginobili Josh Childress Jarvis Hayes

2003-2004 Bobby Simmons Ray Allen Damon Jones Brad Miller Steve Nash

2002-2003 Lamar Odom Corey Maggette Charlie Ward Andre Miller Elton Brand

2001-2002 Bonzi Wells Bob Sura Bryon Russell Erick Dampier Ron Mercer

Bottom 5 Players by Comeback Clutchness Defensive Win Shares Per 48