jongib369
02-24-2016, 05:55 AM
30 August 2014 | Analysis, Analytic Research, Historical Evolution | (http://www.nbaminer.com/who-is-the-best-player-in-the-last-17-seasons-of-nba/)
(This season Steph Curry Efficiency score according to their bolded criteria, 30.64, Durant 29.06, Westbrook 28.89, Harden 26.98, Kobe 11.92. LeBron in the 2010 season got 32.42)
The most basic yet the most asked question of all time about the NBA is who its best player is. It is a simple but extremely complicated question because the response may change according to your comparison metrics and about how you measure the best. Even if you can assure the objectivity and find a common criteria, you can’t normalize the effect of time and the style of plays, conditions and even the rules of the game in different decades.
For example, basketball was played for a long time without the 3-point shot and the NBA didn’t adopt the 3-pointer until 1979 which happens to be Magic Johnson and Larry Bird’s rookie season. Another example is the hand-check penalty, which was introduced by the NBA at the outset of the 2004-05 season, primarily as a way to give the offense more freedom of movement and to keep games running more smoothly. And you know, the zone defense were prohibited in the National Basketball Association prior to the 2001–2002 season… These rules are just a few examples and it is obvious that they impacted the game tremendously.
But hey, we accept the challenge! However, under some conditions:
Our analysis window started from the 1997-98 season. Thus, we try to find the best players of last 17 seasons.
We compare the players according to their overall performance based on their in-game basic stats (their efficiency score) calculated using seasonal total player stats with this formula: PTS+(FGM-FGA)+(FTM-FTA)+REB+AST+STL+BLK-TO. Then, we found the average of each player’s efficiency score per game played.
We also look the plus/minus statistics as a secondary comparison factor.
The playing time is also an important factor. So, we normalized it while implying per 36 minutes stats for all players.
Another significant factor that affects the players’ performance is their team’s strength / depth. If a player is a member of an efficient team, he also benefits from this and his game improves statistically. Let’s say you are in the same team with John Stockton, Chris Paul or Steve Nash, you would often end up with easy chances to score. And as a result, your plus/minus statistics would also be higher. Therefore, at the last stage, we also try to normalize the team effect.
If a player did not play more than half of the regular season matches, or his playing time is below 10 minutes per game, his season stats are excluded from the analysis, as he can’t provide reliable statistical data.
Overall Performances
We’ve both considered season-by-season and overall performance. But first, let’s look at the overall results. There are 1726 unique players played in at least one NBA regular season game, from 1997-98 to end of 2013-14.
The following table summarizes top 25 most effective players of NBA for last 17 seasons.
http://www.nbaminer.com/wp-content/uploads/2014/08/1.png
On the top of the list, we see LeBron James. His efficiency score is 28 per game and is far better than any other player on that span. James is followed by Durant, Garnett, Malone and Duncan. Interestingly all of those players are forwards.
Note that Michael Jordan is #17 and Charles Barkley is #18, even though nothing but their last three seasons were considered (at age35+). They merit a deep respect for that.
Shaquille O’Neal is #12. But he lost a ton of efficiency score for its poor free throw shooting performance. Maybe, he would have been on the top of this list, if he was as good as LeBron James on the line.
Now, let’s look at the same table (top 25) sorted by the efficiency score per 36 minutes this time. In this case Kevin Love is the leader with 26.4 efficiency score per 36 min, following by Shaquille O’Neal, LeBron James, Duncan and Garnett.
Kevin Durant fell to #12; Anthony Davis and Andre Drummond climbed to 6th and 7th respectively (even only with their two seasons worth of performances).
http://www.nbaminer.com/wp-content/uploads/2014/08/2.png
Seasonal Performances
It’s time to look deeper and compare seasonal performances of players. What are the best overall single season performances then? It will show us the most seasonal dominant performance as well.
We see the 2003-04 season performance of Kevin Garnett for the Minnesota Timberwolves at the first place. That year, Kevin Garnett was the MVP of the regular season to no one’s surprise and Timberwolves finished the regular season first in the West but lost at the Western Conference finals. Garnett’s 33.13 efficiency score per game is the highest of last 17 NBA seasons.
Shaq’s 1999-2000 season with the Lakers is the second best seasonal performance. It was one of his prime years and he was an irresistible force around the basket. In fact, Shaquille O’Neal was the definition of domination on those years and was deliberately fouled every time around the rim because he was so good and the free throws were his only weak side. Shaq’s free throw percentage is just 55% (784fta, 482ftm), and missed 352 that year. If he could be just as good as Garnett (he shot 79%, missed only 97 of 465 free throws), Shaq’s efficiency score per game would be 34,89, far higher than Garnett’s.
But notice also that 3 of the 5 most dominant performances of the last 17 regular NBA seasons are by Kevin Garnett (2002-03 #3 and 2004-05 #5). Kevin Garnett is clearly unstoppable in early 2000s.
LeBron James’ 2009-10 season with Cavs is at the fourth place, and Kevin Durant’s amazing 2013-14 season with Thunder is at the sixth.
http://www.nbaminer.com/wp-content/uploads/2014/08/3.png
But how about the team effect and playing time differences? In order to equalize these factors we took into account the efficiency per minutes and calculated team’s seasonal average efficiencies (both per game and per minutes).
After that, we calculated a coefficient by dividing the team’s average efficiency/game per minute to maximum team’s average efficiency/game per minute. Finally we multiplied players’ efficiency scores per minute with that coefficient.
From that viewpoint, Amar’e Stoudemire’s 2007-08 season in Phoenix is phenomenal. He produced 1.08 efficiency per minute, which is above all other players.
Kevin Garnett still has 3 seasonal performances in tıp 5 on that list as well. Even adjusted by team performance and playing time, he is at the top.
http://www.nbaminer.com/wp-content/uploads/2014/08/4.png
We can say that LeBron James is the most efficient player of last 17 years at overall. But Kevin Garnett was the best one in his prime years.
Plus/Minus Statistics
Plus/minus statistics measure the score differences of playing teams when the player is on the floor. It can be a very informative piece of statistics to infer about how much a player aids its team’s success on the floor.
The following table shows top 25 players with seasonal highest plus/minus statistics per game adjusted for his team.
http://www.nbaminer.com/wp-content/uploads/2014/08/5.png
Kobe Bryant’s 2007-08 season performance for Lakers seems the most impressive one. On that year, Lakers’s average +/- statistic per game was -0.66, but Kobe’s was 7.29!
Team’s Seasonal Average Performances
I bet you also wonder what the most efficient teams are. Regarding the last 17 seasons here are the answers:
http://www.nbaminer.com/wp-content/uploads/2014/08/6.png
We look both the team efficiency score average and the +/- statistics average. For the efficiency score, Phoenix Suns of 2004-05 is on the top. But for the +/- stats, Boston Celtics of 2007-08 is leader.
Finally, we tested the correlation between efficiency and +/- stats and we saw that there is a strong correlation (99% significance level) with a correlation coefficient of 0.674. It means that if a team can augment its efficiency 1 point per minute, it will create a 0.674 +/- difference versus the other team in the game.
More precisely, we divided teams into two groups. Those who performed 0.45+ efficiency score per minute and those who didn’t; then, we compared their +/- stats. According to the results, teams with higher efficiency created 0.95 score difference while the others -1.17. The significance level of this independent t-test is >0.99% statistically.
Most Effective Basketball Positions
Lastly we analyzed also which basketball position contributes most efficiency points.
http://www.nbaminer.com/wp-content/uploads/2014/08/7.png
We’ve separated players into 5 positions: Guards, Guard-Forwards, Forwards, Forwards-Centers and Centers. Probably not exactly what one would expect (such as PG, SG, SF, PF and C), but players’ positions can change season by season or team by team. And there are many multi-position players in modern basketball. So we try to generalize as much as possible.
In the last 17 seasons, we are seeing that the guards achieve the higher average minutes per game, and centers have lowest. Forward-Centers and Centers are the most efficient players while the Guard-Forwards are the least efficient. For plus/minus statistics, forward-centers are far better than any other position.
So who are these forwards-centers? You know; Kevin Garnett, Tim Duncan, Kevin Love type players… Unlike the guards and guards-forward era in 90s, forward-centers become the cr
(This season Steph Curry Efficiency score according to their bolded criteria, 30.64, Durant 29.06, Westbrook 28.89, Harden 26.98, Kobe 11.92. LeBron in the 2010 season got 32.42)
The most basic yet the most asked question of all time about the NBA is who its best player is. It is a simple but extremely complicated question because the response may change according to your comparison metrics and about how you measure the best. Even if you can assure the objectivity and find a common criteria, you can’t normalize the effect of time and the style of plays, conditions and even the rules of the game in different decades.
For example, basketball was played for a long time without the 3-point shot and the NBA didn’t adopt the 3-pointer until 1979 which happens to be Magic Johnson and Larry Bird’s rookie season. Another example is the hand-check penalty, which was introduced by the NBA at the outset of the 2004-05 season, primarily as a way to give the offense more freedom of movement and to keep games running more smoothly. And you know, the zone defense were prohibited in the National Basketball Association prior to the 2001–2002 season… These rules are just a few examples and it is obvious that they impacted the game tremendously.
But hey, we accept the challenge! However, under some conditions:
Our analysis window started from the 1997-98 season. Thus, we try to find the best players of last 17 seasons.
We compare the players according to their overall performance based on their in-game basic stats (their efficiency score) calculated using seasonal total player stats with this formula: PTS+(FGM-FGA)+(FTM-FTA)+REB+AST+STL+BLK-TO. Then, we found the average of each player’s efficiency score per game played.
We also look the plus/minus statistics as a secondary comparison factor.
The playing time is also an important factor. So, we normalized it while implying per 36 minutes stats for all players.
Another significant factor that affects the players’ performance is their team’s strength / depth. If a player is a member of an efficient team, he also benefits from this and his game improves statistically. Let’s say you are in the same team with John Stockton, Chris Paul or Steve Nash, you would often end up with easy chances to score. And as a result, your plus/minus statistics would also be higher. Therefore, at the last stage, we also try to normalize the team effect.
If a player did not play more than half of the regular season matches, or his playing time is below 10 minutes per game, his season stats are excluded from the analysis, as he can’t provide reliable statistical data.
Overall Performances
We’ve both considered season-by-season and overall performance. But first, let’s look at the overall results. There are 1726 unique players played in at least one NBA regular season game, from 1997-98 to end of 2013-14.
The following table summarizes top 25 most effective players of NBA for last 17 seasons.
http://www.nbaminer.com/wp-content/uploads/2014/08/1.png
On the top of the list, we see LeBron James. His efficiency score is 28 per game and is far better than any other player on that span. James is followed by Durant, Garnett, Malone and Duncan. Interestingly all of those players are forwards.
Note that Michael Jordan is #17 and Charles Barkley is #18, even though nothing but their last three seasons were considered (at age35+). They merit a deep respect for that.
Shaquille O’Neal is #12. But he lost a ton of efficiency score for its poor free throw shooting performance. Maybe, he would have been on the top of this list, if he was as good as LeBron James on the line.
Now, let’s look at the same table (top 25) sorted by the efficiency score per 36 minutes this time. In this case Kevin Love is the leader with 26.4 efficiency score per 36 min, following by Shaquille O’Neal, LeBron James, Duncan and Garnett.
Kevin Durant fell to #12; Anthony Davis and Andre Drummond climbed to 6th and 7th respectively (even only with their two seasons worth of performances).
http://www.nbaminer.com/wp-content/uploads/2014/08/2.png
Seasonal Performances
It’s time to look deeper and compare seasonal performances of players. What are the best overall single season performances then? It will show us the most seasonal dominant performance as well.
We see the 2003-04 season performance of Kevin Garnett for the Minnesota Timberwolves at the first place. That year, Kevin Garnett was the MVP of the regular season to no one’s surprise and Timberwolves finished the regular season first in the West but lost at the Western Conference finals. Garnett’s 33.13 efficiency score per game is the highest of last 17 NBA seasons.
Shaq’s 1999-2000 season with the Lakers is the second best seasonal performance. It was one of his prime years and he was an irresistible force around the basket. In fact, Shaquille O’Neal was the definition of domination on those years and was deliberately fouled every time around the rim because he was so good and the free throws were his only weak side. Shaq’s free throw percentage is just 55% (784fta, 482ftm), and missed 352 that year. If he could be just as good as Garnett (he shot 79%, missed only 97 of 465 free throws), Shaq’s efficiency score per game would be 34,89, far higher than Garnett’s.
But notice also that 3 of the 5 most dominant performances of the last 17 regular NBA seasons are by Kevin Garnett (2002-03 #3 and 2004-05 #5). Kevin Garnett is clearly unstoppable in early 2000s.
LeBron James’ 2009-10 season with Cavs is at the fourth place, and Kevin Durant’s amazing 2013-14 season with Thunder is at the sixth.
http://www.nbaminer.com/wp-content/uploads/2014/08/3.png
But how about the team effect and playing time differences? In order to equalize these factors we took into account the efficiency per minutes and calculated team’s seasonal average efficiencies (both per game and per minutes).
After that, we calculated a coefficient by dividing the team’s average efficiency/game per minute to maximum team’s average efficiency/game per minute. Finally we multiplied players’ efficiency scores per minute with that coefficient.
From that viewpoint, Amar’e Stoudemire’s 2007-08 season in Phoenix is phenomenal. He produced 1.08 efficiency per minute, which is above all other players.
Kevin Garnett still has 3 seasonal performances in tıp 5 on that list as well. Even adjusted by team performance and playing time, he is at the top.
http://www.nbaminer.com/wp-content/uploads/2014/08/4.png
We can say that LeBron James is the most efficient player of last 17 years at overall. But Kevin Garnett was the best one in his prime years.
Plus/Minus Statistics
Plus/minus statistics measure the score differences of playing teams when the player is on the floor. It can be a very informative piece of statistics to infer about how much a player aids its team’s success on the floor.
The following table shows top 25 players with seasonal highest plus/minus statistics per game adjusted for his team.
http://www.nbaminer.com/wp-content/uploads/2014/08/5.png
Kobe Bryant’s 2007-08 season performance for Lakers seems the most impressive one. On that year, Lakers’s average +/- statistic per game was -0.66, but Kobe’s was 7.29!
Team’s Seasonal Average Performances
I bet you also wonder what the most efficient teams are. Regarding the last 17 seasons here are the answers:
http://www.nbaminer.com/wp-content/uploads/2014/08/6.png
We look both the team efficiency score average and the +/- statistics average. For the efficiency score, Phoenix Suns of 2004-05 is on the top. But for the +/- stats, Boston Celtics of 2007-08 is leader.
Finally, we tested the correlation between efficiency and +/- stats and we saw that there is a strong correlation (99% significance level) with a correlation coefficient of 0.674. It means that if a team can augment its efficiency 1 point per minute, it will create a 0.674 +/- difference versus the other team in the game.
More precisely, we divided teams into two groups. Those who performed 0.45+ efficiency score per minute and those who didn’t; then, we compared their +/- stats. According to the results, teams with higher efficiency created 0.95 score difference while the others -1.17. The significance level of this independent t-test is >0.99% statistically.
Most Effective Basketball Positions
Lastly we analyzed also which basketball position contributes most efficiency points.
http://www.nbaminer.com/wp-content/uploads/2014/08/7.png
We’ve separated players into 5 positions: Guards, Guard-Forwards, Forwards, Forwards-Centers and Centers. Probably not exactly what one would expect (such as PG, SG, SF, PF and C), but players’ positions can change season by season or team by team. And there are many multi-position players in modern basketball. So we try to generalize as much as possible.
In the last 17 seasons, we are seeing that the guards achieve the higher average minutes per game, and centers have lowest. Forward-Centers and Centers are the most efficient players while the Guard-Forwards are the least efficient. For plus/minus statistics, forward-centers are far better than any other position.
So who are these forwards-centers? You know; Kevin Garnett, Tim Duncan, Kevin Love type players… Unlike the guards and guards-forward era in 90s, forward-centers become the cr