Found this article to be pretty cool actually. I am curious as to what the formula is [PER?] but the idea itself seems great.
New stats, new picks
Improved metrics reveal voting oddities in recent MVP awards
When our man John Hollinger introduced his Value Added and Estimated Wins Added metrics recently, he not only provided a sharp look at the 2008-09 MVP race (LeBron James should win the award, Dwyane Wade is a closer second than you'd think, and both Chris Paul and Dwight Howard come before Kobe Bryant), he also gave us a way to look back on recent MVP votes with a more discerning eye.
EWA is a measure of the number of victories a player contributes to a team's overall record, relative to the number of games that team would be expected to win if the player were replaced by a league-average performer. In other words, it tells us, in a nice round number, precisely how grateful Mike Brown and Dan Gilbert should be for the gift of LeBron (30.8 Estimated Wins Added). And it also provides a better understanding of just how good a job the voters did in selecting the last five MVPs.
2003-2004 MVP
(Players are listed in voting order, followed by their EWA.)
Kevin Garnett 28.90
Tim Duncan 20.79
Jermaine O'Neal 14.70
Peja Stojakovic 18.28
Kobe Bryant 16.03
EWA doesn't take defense into account, so KG's MVP year was actually even more impressive than these numbers suggest. It's hard to remember now, bathed as he is in the generous glow of the Celtics' championship, but Garnett-love was somewhat hard to come by back when he was with the Timberwolves in 2003-04, since he still hadn't advanced past the first round of the playoffs. His EWA is evidence of how ridiculous the skeptics were when they questioned whether he had what it took to win. This was a righteous MVP for a dominant player at the height of his powers and it offers an implicit critique of the sometimes simpleminded way we let wins and losses (as if they're something an individual can wish or will into being) color our analysis. The Kevin Garnett who wears a ring (and had an EWA of 16.12 while playing alongside Paul Pierce and Ray Allen in 2007-08) is a hero, while the Kevin Garnett who was head and shoulders above everyone else in the league was somehow flawed, somehow not quite good enough to win our whole-hearted approval.
2004-05
Steve Nash 14.13
Shaquille O'Neal 20.37
Dirk Nowitzki 22.00
Tim Duncan 18.11
Allen Iverson 19.38
2005-06
Steve Nash 17.10
LeBron James 29.55
Dirk Nowitzki 25.67
Kobe Bryant 28.71
Chauncey Billups 18.10
The Nash years are fascinating. EWA suggests the votes both season were way off base, and that Nash didn't even belong in the same conversation with Dirk, Shaq, LeBron and Kobe. Leadership of the savvy point guard is a romantic ideal -- an irresistible story that seems to reinforce all the notions we hold dear about basketball played "the right way" -- and Nash was the flesh-and-blood avatar of those beliefs. There was, too, almost certainly a kind of corrective, throwback appeal in the Nash votes. He seemed like the antidote to a game some folks thought had become too individualized, too ego-driven; a vote for Nash was a vote for something once thought lost.
Driven this way by story more than data, Nash's MVPs seem unjust just a few years after the fact. Surely Shaq is more deserving of two MVP awards than Nash is. Clearly we should be talking about LeBron's second or Kobe's third MVP this season. But still, it's hard to dismiss Nash solely on the basis of EWA. There is little doubt that something interesting happened when Nash joined the Phoenix Suns beginning in the 2004-05 season. There was some kind of catalytic magic in the combination of Nash, Shawn Marion, and Amar'e Stoudemire (and Mike D'Antoni, too) in those first two seasons.
Marion's PER went from 19.9 in 2003-04 to 21.7 in 2004-05 to 23.6 in 2005-06. Stoudemire's PER went from 19.8 to 26.6 before getting hurt in 2005-06. Nash's went from 20.5 in his last season in Dallas to 22.0 in 2004-05 and 23.3 in 2005-06. The Suns went from winning only 29 games in 2003-2004 to 62 in 2004-2005 and 54 the year after that (basically without Stoudamire). On the one hand, Nash's MVP awards reveal our frustrating attachment to old-school ideas, an attachment that sometimes blinds us to the greatness in our midst. On the other hand, don't the awards also indicate a moment in which the MVP was awarded as kind of imprecise, but nevertheless legitimate, marker of a rare and compelling instance of a team in collaboration, flowing and functioning at a ridiculously high level? And isn't Nash, at the helm of that hot band, a pretty reasonable recipient of our attention and praise?
2006-07
Dirk Nowitzki 22.74
Steve Nash 17.17
Kobe Bryant 24.41
Tim Duncan 21.15
LeBron James 22.30
Two voter predispositions to note here: First, Kobe finished two spots behind Dirk, despite an EWA 1.67 points higher. Credit the Mavericks' great record (67-15) that season, as compared to the Lakers' 42-40 campaign. All things being equal, the guy whose contributions coincide with a realistic chance for the title will win the award, and those guys' numbers were close enough that Dirk seems like a legit choice.
Second, LeBron placed fifth, despite an EWA nearly identical to Nowitzki's. It's as if we couldn't quite believe our eyes yet on LeBron, as if it were difficult to wrap our minds around exactly how good he already was, at age 22 (and on the way to his first NBA Finals). There's a slow burn on these things sometimes, which you'll notice in last year's results, too.
2007-08
Kobe Bryant 21.58
Chris Paul 26.01
Kevin Garnett 16.12
LeBron James 28.09
Dwight Howard 18.45
You know the theory about Al Pacino winning the Oscar for "Scent of a Woman" in 1992? It's universally understood to be a subpar Pacino performance (nowhere near as good as his work in either of the first two "Godfather" movies, or in "Dog Day Afternoon"), but he won the award because he'd been passed over so many times before (once, inexplicably, by Art Carney for "Harry and Tonto" in 1974). The same premise works for Kobe's MVP in 2007-08. He was crushed by the Nash craze of 2005-06, undercut by an otherwise anemic Lakers team in 2006-07, but then finally, thanks to the arrival of Pau Gasol, he had a team and a record that seemed to justify the hardware. (The way blindness, depression, and a tango with Gabrielle Anwar once provided the perfect context for Pacino.) It's not that Kobe wasn't excellent last season, it's just that he was pretty clearly outdone by Paul and James, who were hit by the stigma of youth.
In the end, the EWA numbers are illuminating because they give us an increasingly precise way to measure performance, and it's exciting to think of how they may come to shape voting in future years, the way we've seen new metrics like OPS and WHIP begin to have a greater influence on MVP and Cy Young Award voting in baseball. But until they become the new common sense, what's most intriguing about the EWA numbers, particularly as we look back on recent MVP decisions, is the way they hint at the frames through which we see what we see, and the conventions by which we make sense of it.
New stats, new picks
Improved metrics reveal voting oddities in recent MVP awards
When our man John Hollinger introduced his Value Added and Estimated Wins Added metrics recently, he not only provided a sharp look at the 2008-09 MVP race (LeBron James should win the award, Dwyane Wade is a closer second than you'd think, and both Chris Paul and Dwight Howard come before Kobe Bryant), he also gave us a way to look back on recent MVP votes with a more discerning eye.
EWA is a measure of the number of victories a player contributes to a team's overall record, relative to the number of games that team would be expected to win if the player were replaced by a league-average performer. In other words, it tells us, in a nice round number, precisely how grateful Mike Brown and Dan Gilbert should be for the gift of LeBron (30.8 Estimated Wins Added). And it also provides a better understanding of just how good a job the voters did in selecting the last five MVPs.
2003-2004 MVP
(Players are listed in voting order, followed by their EWA.)
Kevin Garnett 28.90
Tim Duncan 20.79
Jermaine O'Neal 14.70
Peja Stojakovic 18.28
Kobe Bryant 16.03
EWA doesn't take defense into account, so KG's MVP year was actually even more impressive than these numbers suggest. It's hard to remember now, bathed as he is in the generous glow of the Celtics' championship, but Garnett-love was somewhat hard to come by back when he was with the Timberwolves in 2003-04, since he still hadn't advanced past the first round of the playoffs. His EWA is evidence of how ridiculous the skeptics were when they questioned whether he had what it took to win. This was a righteous MVP for a dominant player at the height of his powers and it offers an implicit critique of the sometimes simpleminded way we let wins and losses (as if they're something an individual can wish or will into being) color our analysis. The Kevin Garnett who wears a ring (and had an EWA of 16.12 while playing alongside Paul Pierce and Ray Allen in 2007-08) is a hero, while the Kevin Garnett who was head and shoulders above everyone else in the league was somehow flawed, somehow not quite good enough to win our whole-hearted approval.
2004-05
Steve Nash 14.13
Shaquille O'Neal 20.37
Dirk Nowitzki 22.00
Tim Duncan 18.11
Allen Iverson 19.38
2005-06
Steve Nash 17.10
LeBron James 29.55
Dirk Nowitzki 25.67
Kobe Bryant 28.71
Chauncey Billups 18.10
The Nash years are fascinating. EWA suggests the votes both season were way off base, and that Nash didn't even belong in the same conversation with Dirk, Shaq, LeBron and Kobe. Leadership of the savvy point guard is a romantic ideal -- an irresistible story that seems to reinforce all the notions we hold dear about basketball played "the right way" -- and Nash was the flesh-and-blood avatar of those beliefs. There was, too, almost certainly a kind of corrective, throwback appeal in the Nash votes. He seemed like the antidote to a game some folks thought had become too individualized, too ego-driven; a vote for Nash was a vote for something once thought lost.
Driven this way by story more than data, Nash's MVPs seem unjust just a few years after the fact. Surely Shaq is more deserving of two MVP awards than Nash is. Clearly we should be talking about LeBron's second or Kobe's third MVP this season. But still, it's hard to dismiss Nash solely on the basis of EWA. There is little doubt that something interesting happened when Nash joined the Phoenix Suns beginning in the 2004-05 season. There was some kind of catalytic magic in the combination of Nash, Shawn Marion, and Amar'e Stoudemire (and Mike D'Antoni, too) in those first two seasons.
Marion's PER went from 19.9 in 2003-04 to 21.7 in 2004-05 to 23.6 in 2005-06. Stoudemire's PER went from 19.8 to 26.6 before getting hurt in 2005-06. Nash's went from 20.5 in his last season in Dallas to 22.0 in 2004-05 and 23.3 in 2005-06. The Suns went from winning only 29 games in 2003-2004 to 62 in 2004-2005 and 54 the year after that (basically without Stoudamire). On the one hand, Nash's MVP awards reveal our frustrating attachment to old-school ideas, an attachment that sometimes blinds us to the greatness in our midst. On the other hand, don't the awards also indicate a moment in which the MVP was awarded as kind of imprecise, but nevertheless legitimate, marker of a rare and compelling instance of a team in collaboration, flowing and functioning at a ridiculously high level? And isn't Nash, at the helm of that hot band, a pretty reasonable recipient of our attention and praise?
2006-07
Dirk Nowitzki 22.74
Steve Nash 17.17
Kobe Bryant 24.41
Tim Duncan 21.15
LeBron James 22.30
Two voter predispositions to note here: First, Kobe finished two spots behind Dirk, despite an EWA 1.67 points higher. Credit the Mavericks' great record (67-15) that season, as compared to the Lakers' 42-40 campaign. All things being equal, the guy whose contributions coincide with a realistic chance for the title will win the award, and those guys' numbers were close enough that Dirk seems like a legit choice.
Second, LeBron placed fifth, despite an EWA nearly identical to Nowitzki's. It's as if we couldn't quite believe our eyes yet on LeBron, as if it were difficult to wrap our minds around exactly how good he already was, at age 22 (and on the way to his first NBA Finals). There's a slow burn on these things sometimes, which you'll notice in last year's results, too.
2007-08
Kobe Bryant 21.58
Chris Paul 26.01
Kevin Garnett 16.12
LeBron James 28.09
Dwight Howard 18.45
You know the theory about Al Pacino winning the Oscar for "Scent of a Woman" in 1992? It's universally understood to be a subpar Pacino performance (nowhere near as good as his work in either of the first two "Godfather" movies, or in "Dog Day Afternoon"), but he won the award because he'd been passed over so many times before (once, inexplicably, by Art Carney for "Harry and Tonto" in 1974). The same premise works for Kobe's MVP in 2007-08. He was crushed by the Nash craze of 2005-06, undercut by an otherwise anemic Lakers team in 2006-07, but then finally, thanks to the arrival of Pau Gasol, he had a team and a record that seemed to justify the hardware. (The way blindness, depression, and a tango with Gabrielle Anwar once provided the perfect context for Pacino.) It's not that Kobe wasn't excellent last season, it's just that he was pretty clearly outdone by Paul and James, who were hit by the stigma of youth.
In the end, the EWA numbers are illuminating because they give us an increasingly precise way to measure performance, and it's exciting to think of how they may come to shape voting in future years, the way we've seen new metrics like OPS and WHIP begin to have a greater influence on MVP and Cy Young Award voting in baseball. But until they become the new common sense, what's most intriguing about the EWA numbers, particularly as we look back on recent MVP decisions, is the way they hint at the frames through which we see what we see, and the conventions by which we make sense of it.
Comment