Model For Evaluating Talent Worth Questioning

“I believe the player efficiency rating and Elitics Model can reveal, quantify, and communicate performance by providing comparative analysis as calculated with this player efficiency rating. I believe the PER metric is an added dimension that can guide a team and separate themselves from the rest of the NFL.”

“I believe the player efficiency rating and Elitics Model can reveal, quantify, and communicate performance by providing comparative analysis as calculated with this player efficiency rating. I believe the PER metric is an added dimension that can guide a team and separate themselves from the rest of the NFL.”

Those are the words of Tampa Bay Buccaneers’ Manager of Analytics Tyler Oberly during a presentation for the sports analytics conference at MIT Sloan back on March 1. He wasn’t a member of the organization then, instead it was six days later that he announced he had been hired by the team. But his Elitics Model would be put to the test right away as free agency started less than a week later.

Oberly’s theory is designed to help NFL teams -- in this case the Buccaneers -- place an accurate value on players while still maintaining emphasis on the salary cap. General Manager Jason Licht has talked at lengths throughout the offseason about how he and head coach Lovie Smith would use analytics, or PER, to help evaluate players both via free agency and the draft. Analytics is a tool used throughout the NFL but the data collected is used for statistical purposes. Oberly’s PER is the first of its kind and can carry flaws.

For example, for analyzing quarterbacks the PER metric takes into account positive and negative contributions to help create the player efficiency rating or total contribution.

Positive Contributions

Touchdown Percentage, Completion Percentage, Quarterback Rating, Passing yards per attempt, Game Winning Drives

Negative Contributions

Interception Percentage, Sack Percentage

The efficiency is then factored in with the player’s salary to give a true player value. Oberly’s model, after all the calculations were done, gave Bucs’ veteran quarterback Josh McCown a rating that placed him sixth in the NFL -- ranking him above Russell Wilson, Tom Brady, Ben Roethelisberger, and Aaron Rodgers. The Bucs’ manager of analytics said he used projections to determine a 16-game value for McCown and since his efficiency was so high it warranted placing McCown in the Top 10 of the NFL.

While the above contributions are legitimate factors, they can be skewed in many ways: Receiver drops, offensive line play, what the opposing defenses do, etc.

A team that took a shot at a similar style of analytics was the Philadelphia Eagles. In a January 2013 column, Dan Pompei wrote the following in the National Football Post:

Another front office man believes there is a benefit to analytics regarding player evaluation, but he said you can’t allow numbers to override what your eyes tell you. “It can help if used properly,” he said. “But you have to be careful because there are not a lot of absolutes in football. When a guy gets 12 sacks, you better figure out how he got those sacks. You have to study film to validate production.”

Case in point: a source said in 2010, the Eagles relied heavily on analytics to select Washington defensive end Daniel Te’o-Nesheim in the third round of the draft. Many were stunned he was chosen so high, and the Eagles cut him before the start of the 2011 season.

Speaking at the same conference as Oberly was Football Outsiders Editor-in-Chief Aaron Schatz, current Atlanta Falcons Assistant General Manager Scott Pioli, St. Louis Rams COO and VP of Football Operations (and former Bucs’ Senior Assistant) Kevin Demoff, and San Francisco 49ers’ COO Paraag Marathe -- which can be found on YouTube. All admitted to teams using analytics for compiling statistical data but none were to evaluate players. In fact, Pioli was one to suggest using analytics to acquire players is a failure because certain players fit in certain systems regardless of their numbers.

The Bucs have admitted to using this format to help assist in the evaluation of players. This offseason, they acquired defensive end Michael Johnson, left tackle Anthony Collins, defensive tackle Clinton McDonald, quarterback Josh McCown,  and cornerback Alterraun Verner among others. All have underperformed to this point and NFL FIlms’ Greg Cosell had this to say about Johnson when he was a guest on Fan Interference on 98.7 The Fan:

"I'm sure people in Tampa, when he got signed, thought that, 'oh you are getting a pass rusher.' He's not a pass rusher in the strict sense. He's a good football player, but in this particular scheme where four man d-line pass rush is critical to the success of the scheme, he's not that guy."

Analytics has always been around. How it is used determines a team’s real success. Football is the ultimate team sport. All statistics can be manipulated in many ways.There have been some rumblings surrounding the Bucs' free agent pickups not matching up with grades from team scouts, yet the efficiency rating trumped those grades.

The level of influence towards decisions made on player acquisitions during free agency and the draft off of this model is uncertain. Game film should never be put on the backburner for a set of numbers. But time will tell how well this new form of analytics really works.

Follow Bucs Blitz on twitter: @ScoutBuccaneers


Bucs Blitz Top Stories