Fantasy hoops: Harden vs. Kobe, MJ, LeBron and the best perimeter players of the past 25 years
ESPN PLUS ($ MATERIAL)
2/15/19
Right around the new year, I made the argument that James Harden was the best offensive player in the NBA this season, and it's not close.
I had a lot of people skewer me on Twitter (@ProfessorDrz), but I also had a lot of people agree with me. One of those who agreed with me, unsurprisingly, was Houston Rockets general manager Daryl Morey. Morey liked and retweeted the article, and then the very next day doubled down on my position:
"You could argue him as the best offensive player of all time," Morey told ESPN.
Hey-o! That escalated quickly!
But ... could Morey possibly be correct?
Could Harden's 2018-19 campaign be better statistically than the best of Kobe Bryant? Than LeBron James? Than Michael Jordan??
That sounds like a barbershop argument to me ... and that's right up my alley!
Let's compare James Harden's 2018-19 performance through the All-Star break to the best single seasons of the 10 best perimeter performers of the past 25 seasons. ESPN has had <offer style="box-sizing: border-box;">fantasy basketball for about 25 years, and coincidentally, we have access to plus/minus numbers dating back to the 1993-94 season, as well. Thus, we've got two strong ways to compare Harden to these greats via fantasy points scoring (i.e., box scores) and via plus/minus data (i.e., team impact).
</offer>
For Harden's competition, let's compare him with:
Here is what the main fantasy/box-score stats looked like for each of these monster seasons, with fantasy points calculated using ESPN's standard points scoring system: field goals made (1 point), field goals attempted (-1), free throws made (1), free throws attempted (-1), points (1), rebounds (1), assists (1), steals (1), blocks (1), turnovers (-1).
<aside class="inline inline-table" style="box-sizing: border-box; font-family: -apple-system, BlinkMacSystemFont, Roboto, Arial, "Helvetica Neue", Helvetica, sans-serif; height: 391px; border: 1px solid rgb(220, 221, 223); clear: both; margin: 6px 0px 18px; padding: 15px; width: 565px; color: rgb(0, 0, 0); font-size: 16px;">Box Score/Fantasy Points For Peak Campaigns, Past 25 Seasons
<tbody>
</tbody></aside>
A few things to consider:
If we looked at this in terms of roto fantasy, the comparison would be more complicated to evaluate. Harden would still have a strong case, though, because he has the highest scoring average, the highest volume of free throws at a strong percentage and a huge 3-point impact (5.1 3PG, tied with 2016 Curry's 5.1 3PG for most in this group).
Harden is one of only five players listed here with more than 14 combined assists and rebounds (Nash, Paul, LeBron and the leader Westbrook), and is one of only two listed here with at least 3.0 combined steals and blocks (Wade). Harden's turnovers (tied with Westbrook for worst listed here) would be an issue in nine-category leagues, and his 44.1 FG% at heavy volume (24.4 FGA per game) isn't ideal, either, but the breadth and volume of his strengths would be enough to give him a strong roto argument in this comparison.
Team impact comparison (plus/minus stats)
We have raw plus/minus information available for each of these seasons. Raw plus/minus is something that you could calculate on a piece of paper while watching the game, much like the box scores. On-court plus/minus is the scoring margin for a team while a player is on the court, while on/off plus/minus is the team's plus/minus when a player is playing minus the team's plus/minus when the player is on the bench. This gives a general feel for how a player's team performs with and without him and, thus, can be a solid first-pass estimate for that player's impact.
There are a lot of potential confounds with raw plus/minus, though, including how best to account for teammate caliber and strength of opposition. Thus, statisticians have used various mathematical regression techniques to convert play-by-play data into various estimates of a player's impact that are more granular than the raw plus/minus values.
The real plus-minus (RPM) stat yields estimates for exactly how much an individual player improves his team's offense (ORPM), defense (DRPM) and overall scoring margins (RPM).
Here's a snapshot of several of these plus/minus values for the seasons under consideration:
<aside class="inline inline-table" style="box-sizing: border-box; font-family: -apple-system, BlinkMacSystemFont, Roboto, Arial, "Helvetica Neue", Helvetica, sans-serif; height: 391px; border: 1px solid rgb(220, 221, 223); clear: both; margin: 6px 0px 18px; padding: 15px; width: 565px; color: rgb(0, 0, 0); font-size: 16px;">+/- Stats For Peak Campaigns, Past 25 Seasons
<tbody>
</tbody></aside>
A few things to consider:
Be that as it may, even this deeper dive into the analytics supports the notion that Harden, this season, is having one of the best offensive seasons in NBA history.
Is Harden the best of the best?
So ... is Harden, this season, playing better than Kobe's best? Based on fantasy/box-score volume and efficiency, as well as on-court impact analytics, you could argue effectively that he is.
Harden's box-score stats compare favorably with post-first-retirement Jordan but come up short compared with some of Jordan's pre-retirement seasons, and impact-wise it appears as if Jordan's footprint in 1995-96 was more impressive than Harden's this season.
Harden's fantasy/box-score numbers this season might edge even peak LeBron, but, once the team impact is factored in, King James was still clearly the better player.
Overall, Harden might not be quite the best offensive player in history, based on his numbers this season. But, to GM Morey's point, he has at least put himself into the discussion. And that is an extremely big deal.
ESPN PLUS ($ MATERIAL)
2/15/19
Right around the new year, I made the argument that James Harden was the best offensive player in the NBA this season, and it's not close.
I had a lot of people skewer me on Twitter (@ProfessorDrz), but I also had a lot of people agree with me. One of those who agreed with me, unsurprisingly, was Houston Rockets general manager Daryl Morey. Morey liked and retweeted the article, and then the very next day doubled down on my position:
"You could argue him as the best offensive player of all time," Morey told ESPN.
Hey-o! That escalated quickly!
But ... could Morey possibly be correct?
Could Harden's 2018-19 campaign be better statistically than the best of Kobe Bryant? Than LeBron James? Than Michael Jordan??
That sounds like a barbershop argument to me ... and that's right up my alley!
Let's compare James Harden's 2018-19 performance through the All-Star break to the best single seasons of the 10 best perimeter performers of the past 25 seasons. ESPN has had <offer style="box-sizing: border-box;">fantasy basketball for about 25 years, and coincidentally, we have access to plus/minus numbers dating back to the 1993-94 season, as well. Thus, we've got two strong ways to compare Harden to these greats via fantasy points scoring (i.e., box scores) and via plus/minus data (i.e., team impact).
</offer>
For Harden's competition, let's compare him with:
- 1995-96 Michael Jordan (MVP), his best statistical season after his first retirement.
- 2002-03 Tracy McGrady, his career-best season when he led the NBA in scoring.
- 2004-05 Steve Nash (MVP), first of two straight MVPs and monster offensive impact.
- 2005-06 Kobe Bryant, his best statistical season, including his 81-point detonation.
- 2008-09 Dwyane Wade, best statistical season, which would be MVP-worthy most seasons.
- 2008-09 Chris Paul, second straight MVP-worthy season, one of four top-five MVP finishes.
- 2009-10 LeBron James (MVP),one of several monster seasons that could have been used.
- 2013-14 Kevin Durant (MVP), peaked statistically with the Thunder with Westbrook injured.
- 2015-16 Stephen Curry (MVP), second straight MVP, first unanimous award in NBA history.
- 2016-17 Russell Westbrook (MVP), second player in NBA history to average a triple-double.
Here is what the main fantasy/box-score stats looked like for each of these monster seasons, with fantasy points calculated using ESPN's standard points scoring system: field goals made (1 point), field goals attempted (-1), free throws made (1), free throws attempted (-1), points (1), rebounds (1), assists (1), steals (1), blocks (1), turnovers (-1).
<aside class="inline inline-table" style="box-sizing: border-box; font-family: -apple-system, BlinkMacSystemFont, Roboto, Arial, "Helvetica Neue", Helvetica, sans-serif; height: 391px; border: 1px solid rgb(220, 221, 223); clear: both; margin: 6px 0px 18px; padding: 15px; width: 565px; color: rgb(0, 0, 0); font-size: 16px;">Box Score/Fantasy Points For Peak Campaigns, Past 25 Seasons
| PLAYER | YEAR | FGM | FGA | FTM | FTA | REB | AST | STL | BLK | TO | PTS | FP |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Michael Jordan | 1995-96 | 11.2 | 22.6 | 6.7 | 8.0 | 6.6 | 4.3 | 2.2 | 0.5 | 2.4 | 30.4 | 28.9 |
| Tracy McGrady | 2002-03 | 11.1 | 24.2 | 7.7 | 9.7 | 6.5 | 5.5 | 1.7 | 0.8 | 2.6 | 32.1 | 28.9 |
| Steve Nash | 2004-05 | 5.7 | 11.4 | 2.8 | 3.2 | 3.3 | 11.5 | 1.0 | 0.1 | 3.3 | 15.5 | 22.0 |
| Kobe Bryant | 2005-06 | 12.2 | 27.2 | 8.7 | 10.2 | 5.3 | 4.5 | 1.8 | 0.4 | 3.1 | 35.4 | 27.8 |
| Dwyane Wade | 2008-09 | 10.8 | 22.0 | 7.5 | 9.8 | 5.0 | 7.5 | 2.2 | 1.3 | 3.4 | 30.2 | 29.3 |
| Chris Paul | 2008-09 | 8.1 | 16.1 | 5.8 | 6.7 | 5.5 | 11.0 | 2.8 | 0.1 | 3.0 | 22.8 | 30.3 |
| LeBron James | 2009-10 | 10.1 | 20.1 | 7.8 | 10.2 | 7.3 | 8.6 | 1.6 | 1.0 | 3.4 | 29.7 | 32.4 |
| Kevin Durant | 2013-14 | 10.5 | 20.8 | 8.7 | 9.9 | 7.4 | 5.5 | 1.3 | 0.7 | 3.5 | 32.0 | 31.9 |
| Stephen Curry | 2015-16 | 10.2 | 20.2 | 4.6 | 5.1 | 5.4 | 6.7 | 2.1 | 0.2 | 3.3 | 30.1 | 30.7 |
| Russell Westbrook | 2016-17 | 10.2 | 24.0 | 8.8 | 10.4 | 10.7 | 10.4 | 1.6 | 0.4 | 5.4 | 31.6 | 33.9 |
| James Harden | 2018-19 | 10.8 | 24.4 | 10.0 | 11.5 | 6.7 | 7.7 | 2.2 | 0.8 | 5.4 | 36.6 | 33.5 |
<tbody>
</tbody>
A few things to consider:
- Jordan's 1995-96 season was not his statistical peak; that peak came before the 25-year window used here. But, for the record, Jordan would have averaged 35.0 fantasy points per game during his 1987-88 MVP season.
- McGrady's 2002-03 season was truly remarkable, considering that this was near the peak of the monster defense/handcheck era before the 2005 rule change. Even in that low-scoring era, McGrady's 28.9 fantasy points per game matched 1996 Jordan and stood up until Wade and Paul in 2008-09.
- Nash's box-score stats aren't on the level of any of the other greats mentioned here, but Nash took home back-to-back MVPs in 2004-05 and 2005-06. In the next section, which looks at impact according to plus/minus stats, we'll get a better idea of why.
- Kobe's 2005-06 season is legendary, but when compared with some of his direct rivals in the same time span, guys like McGrady and Wade arguably had more impressive peak seasons, statistically.
- The main reason Wade and Paul never won MVPs is that they peaked at the same time that LeBron was taking over the world. Both of them produced MVP-level seasons.
- LeBron's 2008-09 and 2017-18 seasons were just as statistically dominant as 2009-10, so any of them could have been used here. His 2009-10 season was the highest fantasy points mark until the past few seasons, when scoring in the NBA has gone nuts.
- Westbrook's first triple-double season of 2016-17 represents the highest fantasy points scoring average for a perimeter-based, offensive-focused player in this comparison.
If we looked at this in terms of roto fantasy, the comparison would be more complicated to evaluate. Harden would still have a strong case, though, because he has the highest scoring average, the highest volume of free throws at a strong percentage and a huge 3-point impact (5.1 3PG, tied with 2016 Curry's 5.1 3PG for most in this group).
Harden is one of only five players listed here with more than 14 combined assists and rebounds (Nash, Paul, LeBron and the leader Westbrook), and is one of only two listed here with at least 3.0 combined steals and blocks (Wade). Harden's turnovers (tied with Westbrook for worst listed here) would be an issue in nine-category leagues, and his 44.1 FG% at heavy volume (24.4 FGA per game) isn't ideal, either, but the breadth and volume of his strengths would be enough to give him a strong roto argument in this comparison.
Team impact comparison (plus/minus stats)
We have raw plus/minus information available for each of these seasons. Raw plus/minus is something that you could calculate on a piece of paper while watching the game, much like the box scores. On-court plus/minus is the scoring margin for a team while a player is on the court, while on/off plus/minus is the team's plus/minus when a player is playing minus the team's plus/minus when the player is on the bench. This gives a general feel for how a player's team performs with and without him and, thus, can be a solid first-pass estimate for that player's impact.
There are a lot of potential confounds with raw plus/minus, though, including how best to account for teammate caliber and strength of opposition. Thus, statisticians have used various mathematical regression techniques to convert play-by-play data into various estimates of a player's impact that are more granular than the raw plus/minus values.
The real plus-minus (RPM) stat yields estimates for exactly how much an individual player improves his team's offense (ORPM), defense (DRPM) and overall scoring margins (RPM).
Here's a snapshot of several of these plus/minus values for the seasons under consideration:
<aside class="inline inline-table" style="box-sizing: border-box; font-family: -apple-system, BlinkMacSystemFont, Roboto, Arial, "Helvetica Neue", Helvetica, sans-serif; height: 391px; border: 1px solid rgb(220, 221, 223); clear: both; margin: 6px 0px 18px; padding: 15px; width: 565px; color: rgb(0, 0, 0); font-size: 16px;">+/- Stats For Peak Campaigns, Past 25 Seasons
| PLAYER | YEAR | ON +/- | ON/OFF +/- | ORPM | RPM |
|---|---|---|---|---|---|
| Michael Jordan | 1995-96 | 16.7 | 15.2 | N/A | N/A |
| Tracy McGrady | 2002-03 | 3.2 | 13.0 | 7.7 | 7.4 |
| Steve Nash | 2004-05 | 12.7 | 14.9 | 7.3 | 5.7 |
| Kobe Bryant | 2005-06 | 4.6 | 12.5 | 7.4 | 5.9 |
| Dwyane Wade | 2008-09 | 3.2 | 14.2 | 6.9 | 8.8 |
| Chris Paul | 2008-09 | 6.5 | 19.6 | 7.4 | 10.9 |
| LeBron James | 2009-10 | 15.0 | 21.2 | 8.7 | 10.8 |
| Kevin Durant | 2013-14 | 8.0 | 6.0 | 6.9 | 7.0 |
| Stephen Curry | 2015-16 | 17.9 | 22.2 | 7.7 | 8.5 |
| Russell Westbrook | 2016-17 | 4.0 | 12.5 | 6.7 | 6.3 |
| James Harden | 2018-19 | 3.1 | 3.3 | 7.8 | 8.1 |
<tbody>
</tbody>
A few things to consider:
- The raw plus/minus scores tell a story of their own. Players on good teams have the highest plus/minus values (led by Curry, Jordan and LeBron), but, for the most part, the on/off plus/minus values for these seasons are in similar ranges.
- The regressed RPM scores suggest that each of the individual offensive impacts of these players were similar, adding between 6.7 and 8.7 points to their teams on offense, which are monster values.
- Jordan's 1995-96 season is before the RPM era, but his raw plus/minus scores are among the best on this list, which indicates that his regressed values would similarly be near the top.
- Wade, Paul and LeBron were the three players on this list who also had significant positive impact on defense in these seasons, which contributes to them having the highest measured overall impacts as measured by RPM on this list. Jordan would likely have been in this crew, as well, based on his 1997-98 regressed plus/minus results that indicated he was also having a strong defensive impact, even late in his career.
Be that as it may, even this deeper dive into the analytics supports the notion that Harden, this season, is having one of the best offensive seasons in NBA history.
Is Harden the best of the best?
So ... is Harden, this season, playing better than Kobe's best? Based on fantasy/box-score volume and efficiency, as well as on-court impact analytics, you could argue effectively that he is.
Harden's box-score stats compare favorably with post-first-retirement Jordan but come up short compared with some of Jordan's pre-retirement seasons, and impact-wise it appears as if Jordan's footprint in 1995-96 was more impressive than Harden's this season.
Harden's fantasy/box-score numbers this season might edge even peak LeBron, but, once the team impact is factored in, King James was still clearly the better player.
Overall, Harden might not be quite the best offensive player in history, based on his numbers this season. But, to GM Morey's point, he has at least put himself into the discussion. And that is an extremely big deal.
