Packers exploring benefits of analytics in player personnel, scouting
Associated Press
Analytics helped convince the Jacksonville Jaguars to draft tackle Luke Joeckel (right) No. 2 overall in 2013.
It appears as if the Green Bay Packers soon will be next.
"We are beginning the process of dabbling in analytics," general manager Ted Thompson said in a recent interview. "First of all, I've got to get a couple really smart people around me and teach me what it is. We're working on that."
The Packers have had scholarly Mike Eayrs on staff since March 2001 as director of research and development. But Eayrs, 63, is a former collegiate coach and spends most of his time working for coach Mike McCarthy.
During the regular season, Eayrs is given about 10 minutes at midweek to address the team about trends for that week's officiating crew. On game day, he charts the time it takes for the offense to get the play off.
Eayrs has been described by former colleagues as one of the NFL's foremost authorities on all game-day tactical decisions.
"He helps us quite a bit during the draft in terms of trade value," Thompson said. "But his primary job is to work with the coaching staff on game day...those kind of analytics."
The Packers hired Melanie Marohl as salary cap analyst in July 2001 to evaluate trends in player salaries across the league under head negotiator Andrew Brandt and, since 2008, Russ Ball.
Green Bay also has turned to analytics in trying to stem its recent failures in injury prevention. The GPS tracking devices some players wear in practice probably are a result of that.
The area of analytics in which the Packers don't have a full-time employee is player personnel and scouting.
Analytics is a fancy word for using available data to draw conclusions for the task at hand. The Packers are in the business of winning football games, and to do it they constantly must find better players.
Thompson, 61, graduated in business from Southern Methodist. After a 10-year playing career, he obtained a Series 7 stockbroker's license and worked in his native Texas in the late 1980s. He missed football, however, and was hired by the Packers as a scout in 1992.
Despite running the football division of a massive corporation, Thompson remains a scout not only in heart but also in practice.
How can a career football man who believes in beating the bushes seeking players believe in metrics presented by a non-football newcomer?
"I'm not saying I believe in it, but I'm not going to shut my ears to it, 61 or no," replied Thompson. "We're going to be open to new opportunities and things like that.
"People have made fun of it in terms of being, quote-unquote, 'Moneyball.' I know the Astros, one of my teams, has really started to go all in on this stuff. I think there's a place for it.
"There's always been some analytics. It's a 6-5 guy instead of a 6-4 guy. That's analytics. So you don't want to make it too complicated. But we are starting to work a little bit in that area."
"Moneyball," the 2003 book written by Michael Lewis, told the story of how the low-revenue Oakland A's and general manager Billy Beane came to rely heavily on statistical evaluation to gain an edge in player procurement.
By definition, baseball has always been a numbers game. Football lagged behind until the last 10 to 15 years, when several companies, such as STATS in suburban Chicago, sprouted up to collect and sell the results of their data in what had been largely unexplored areas of football.
In the 1960s, the Dallas Cowboys were the first team to use a computer for scouting. Ron Wolf, the Packers' retired general manager, recalled teams like Atlanta and Cleveland using a secretive device known only as the "black box" to assist the draft.
One of the first football analytic hires in the NFL without a background in the sport was Paraag Marathe by the San Francisco 49ers in 2001.
Now the team's chief operating officer and lead contract negotiator, Marathe continues heading up its analytic effort.
"It's a lot more widely used than most people would advertise," Atlanta Falcons general manager Thomas Dimitroff told USA Today in May 2013. "It's been around a long time. We use analytics to eliminate as much guesswork as we possibly can."
Last summer, the Chicago Bears hired Mitch Tanney away from STATS to become their director of analytics. He played quarterback at Division III Monmouth (Ill.), graduating in mathematics.
"He is going to help us in analyzing (veteran) player statistical information to compare and contrast with other players in the league," Bears general manager Phil Emery said at the time.
"He will help us in analyzing future prospects with all the various statistical information each player presents: performance scores on the field, medical scores, background testing and psychological testing. He will dial down the important information and make it clear how it relates to performance.
"I always had a strong interest in this area. Going to that conference (MIT Sloan Sports Analytics Conference in March 2013) helped me get my head around the analytics equation."
The New England Patriots have been heavily involved in analytics under Ernie Adams, their brainy research director since 2000 and a key resource for Bill Parcells and the New York Giants in the mid-1980s.
Tony Khan, the son of owner Shahid Khan, directs the Jacksonville Jaguars' three-man analytics staff. The Minnesota Vikings have two full-time people involved in analytics.
Two years ago, Baltimore Ravens GM Ozzie Newsome appointed Sandy Weil as director of football analytics. Weil owns a math degree from Yale and a master's in computational finance from Carnegie Mellon.
"You've got to know how to blend your scouts in with the analytics," one NFL general manager said. "It doesn't set your board for you, but it's a cross-check and a reference on guys you might have low on your board.
"You say, 'Let's look at these guys that have analytics that are high and see what the true value of that player is.'"
Just as it was in the "Moneyball" experience in Oakland, resistance is commonplace in the NFL.
"You know what? It's all a crock of (expletive)," one veteran scout said. "They're going to tell you if a guy can play or cannot play? C'mon."
Said another: "Who knows what it (analytics) is? To me, it's common sense. If you can't evaluate players, I don't give a damn what type of analytics you have."
And a third: "We got an analytics guys who said, 'Well, the metrics of the height-weight-speed proportion of the lean muscle mass are to a point, why not take a shot on him as a free agent?'
"I go, 'The (expletive) never makes any plays.'"
Before drafting tackle Luke Joeckel with the No. 2 pick in 2013, Tony Khan told GQ magazine that the Jaguars examined every first-round pick from 1990-2008. They tabulated games played and started, and found tackles taken in the top 10 were higher in both categories than any other position.
Another example of football analytics was a study done by the Ravens a few years ago that revealed a strong correlation between sack production regardless of collegiate level and NFL success.
Today, technology has progressed to the point where metrics can detail almost every aspect of football.
"It only takes a foot," said Packers tight end Ryan Taylor, referring to the final play of the 34th Super Bowl when Tennessee's Kevin Dyson was tackled by St. Louis' Mike Jones inches short of the goal line as time expired.
"It's smart. That's a big part of the game now. You can really use something to improve your football team."
More than half of NFL teams have an analytics person listed in their staff directory. Others are more clandestine and don't even announce hires in this endeavor.
The Packers did have an analytics consultant on board in 2009, and his input was considered significant in the decisions to select B.J. Raji and Clay Matthews in the first round.
Thompson, however, was unwilling to commit to that consultant and his services were discontinued after a year or two.
"Ted has an open mind," said an executive in personnel for an NFL team and a friend of Thompson's. "He's of the 'Old Guard.' But the 'Old Guard,' with progressive thinking, should always take an understanding of what analytics can do."
No comments:
Post a Comment