burnet44
07-16-2007, 08:56 AM
Tuesday, July 17, 2007
Last modified Tuesday, January 10, 2006 3:41 PM MST
LANL study kicks up storm
ROGER SNODGRASS, roger@lamonitor.com, Monitor Assistant Editor
A funny thing happened to physicist Eli Ben-Naim at Los Alamos National Laboratory while he was trying to analyze competitive societies.
Ben-Naim and his collaborators from Boston University, applying statistical theories, came up with a finding that struck a nerve in the supercharged world of international sports.
"What is the most competitive sport?" Ben-Naim asked, as way of validating a theory relating the relative strength of two opponents.
The answer turned out to be English soccer, which beat out major American baseball, basketball and football leagues in the decisive measure of the frequency of upsets - when the weaker team overcomes the stronger.
How frequently that happens turns out to be a gauge of competitiveness, and that number could have a wide range of useful applications for making predictions across the broader society.
"The funny thing is, nobody thought about it before," Ben-Naim said, looking somewhat amused, during an interview Monday.
The researchers were looking for a way to validate mathematical theories about social strife that affects everything from ants and apes to rogue and nation states.
To make the question real, the scientists took a detailed look at league records for top professional baseball, basketball, football and hockey leagues from North America and the English Football Association, the premier soccer league.
The advantage of the focus on these leagues was that the data was extensive, available and trustworthy, Ben-Naim said.
Weaker and stronger teams could be readily identified by their win-loss record at the time of the contest, not an absolute evaluator, but at least a known quantity.
The voluminous statistics examined in the report included 43,350 games, going back to 1888 in the case of the soccer league, and nearly 164,000 games since 1901 for baseball.
Although the resulting article is still only in pre-print form, as the authors are still deciding where to place it, the news was picked up by two British magazines and the BBC, apparently as a matter of local pride.
"This thing bounced huge in Britain," said Kevin Roark, a LANL spokesman.
Canada and Ireland have also called, he said, complaining about the relatively disrespectful valuation of hockey and the exclusion of Irish soccer in the math.
The study found soccer and baseball to be the most competitive, and basketball and football as the least competitive. Hockey falls in the middle.
Ben-Naim finds the trend lines valuable, which show NFL football to be moving up slowly from the bottom of the chart.
The naturalist and author Stephen Jay Gould, cited as a footnote in the paper, also exploited baseball statistics for studying evolutionary behavior. In "The spread of excellence from Plato to Darwin," he addressed the question of why there were no .400 hitters since Ted Williams hit .406 in 1941, noting the adaptive behavior of teams as well as changing rules and strategies.
There are a lot of variables missing from the model, Ben-Naim points out. There are no minor leagues in the mix, no allowances for "home team advantage," deep-pocket owners or bonanza television revenues factored in.
Can statistical physics, using mathematical models derived from interacting particle systems, be useful in social scenarios?
Ben-Naim said he thought league commissioners ought to be interested in this kind of measurement, because parity and upsets create excitement - and profits.
He said, "Commentators may claim that parity is increasing in the National Football league, but how do they know?"
Now, there is a number for it.
The deeper scientific issue is about modeling international conflict, the subtext of the researchers' root investigation into conflict and hierarchy.
"This is a big question we are worried about at Los Alamos," he said.
He added, "There is nothing wrong with science being a little playful, to stimulate public interest."
The real measure of the sports study, he said, is whether others will adopt it and improve it.
Last modified Tuesday, January 10, 2006 3:41 PM MST
LANL study kicks up storm
ROGER SNODGRASS, roger@lamonitor.com, Monitor Assistant Editor
A funny thing happened to physicist Eli Ben-Naim at Los Alamos National Laboratory while he was trying to analyze competitive societies.
Ben-Naim and his collaborators from Boston University, applying statistical theories, came up with a finding that struck a nerve in the supercharged world of international sports.
"What is the most competitive sport?" Ben-Naim asked, as way of validating a theory relating the relative strength of two opponents.
The answer turned out to be English soccer, which beat out major American baseball, basketball and football leagues in the decisive measure of the frequency of upsets - when the weaker team overcomes the stronger.
How frequently that happens turns out to be a gauge of competitiveness, and that number could have a wide range of useful applications for making predictions across the broader society.
"The funny thing is, nobody thought about it before," Ben-Naim said, looking somewhat amused, during an interview Monday.
The researchers were looking for a way to validate mathematical theories about social strife that affects everything from ants and apes to rogue and nation states.
To make the question real, the scientists took a detailed look at league records for top professional baseball, basketball, football and hockey leagues from North America and the English Football Association, the premier soccer league.
The advantage of the focus on these leagues was that the data was extensive, available and trustworthy, Ben-Naim said.
Weaker and stronger teams could be readily identified by their win-loss record at the time of the contest, not an absolute evaluator, but at least a known quantity.
The voluminous statistics examined in the report included 43,350 games, going back to 1888 in the case of the soccer league, and nearly 164,000 games since 1901 for baseball.
Although the resulting article is still only in pre-print form, as the authors are still deciding where to place it, the news was picked up by two British magazines and the BBC, apparently as a matter of local pride.
"This thing bounced huge in Britain," said Kevin Roark, a LANL spokesman.
Canada and Ireland have also called, he said, complaining about the relatively disrespectful valuation of hockey and the exclusion of Irish soccer in the math.
The study found soccer and baseball to be the most competitive, and basketball and football as the least competitive. Hockey falls in the middle.
Ben-Naim finds the trend lines valuable, which show NFL football to be moving up slowly from the bottom of the chart.
The naturalist and author Stephen Jay Gould, cited as a footnote in the paper, also exploited baseball statistics for studying evolutionary behavior. In "The spread of excellence from Plato to Darwin," he addressed the question of why there were no .400 hitters since Ted Williams hit .406 in 1941, noting the adaptive behavior of teams as well as changing rules and strategies.
There are a lot of variables missing from the model, Ben-Naim points out. There are no minor leagues in the mix, no allowances for "home team advantage," deep-pocket owners or bonanza television revenues factored in.
Can statistical physics, using mathematical models derived from interacting particle systems, be useful in social scenarios?
Ben-Naim said he thought league commissioners ought to be interested in this kind of measurement, because parity and upsets create excitement - and profits.
He said, "Commentators may claim that parity is increasing in the National Football league, but how do they know?"
Now, there is a number for it.
The deeper scientific issue is about modeling international conflict, the subtext of the researchers' root investigation into conflict and hierarchy.
"This is a big question we are worried about at Los Alamos," he said.
He added, "There is nothing wrong with science being a little playful, to stimulate public interest."
The real measure of the sports study, he said, is whether others will adopt it and improve it.