Monday, March 3, 2014
Evaluating Tony Romo: Analytics or skill?
By Calvin Watkins
There was a major discussion regarding analytics and how it pertains to football last week.
Bill Barnwell wrote a nice piece about it on Grantland and this quote from Tennessee coach Ken Whisenhunt was my favorite. When asked about analytics, Whisenhunt replied, “I probably don’t understand it.”
I’ve been a big believer in either you can play or you can’t. Forget about the stats. Personnel men and scouts don’t have spreadsheets out when grading talent.
I bring this up today when I look at Cowboys quarterback Tony Romo.
I’ve seen basically all but four of his 108 career starts since he took over as starter for Drew Bledsoe in 2006. Romo is an excellent quarterback. Yet, he lacks playoff wins. Romo has one playoff victory, beating Donovan McNabb in 2009. Romo lost to Brett Favre in the divisional round the same year.
In regular season finales where the Cowboys needed to win to get into the playoffs, Romo was outplayed by Eli Manning and Robert Griffin III. Romo is 1-6 in win-or-go-home games as a starter. You can place some of the blame on Romo, but football is really a team game in the truest sense. Coach Jason Garrett likes to say quarterbacks get too much of the blame for the losses given there are 10 other men who can influence a play.
But what do the analytics say about Romo?
Football Outsiders ranked Romo 7th among quarterbacks with a DYAR of 838.
According to Football Outsiders DYAR is Defense-adjusted yards Above Replacement. It gives the value of a quarterbacks performance compared to replacement level, adjusted for situation and opponent and then translated into yardage.
Romo’s ranking was higher than Russell Wilson (779) and Aaron Rodgers (743). Two other quarterbacks who made the postseason in 2013, Andy Dalton (551) and Cam Newton (425), also had a lower DYAR than Romo.
DVOA is another Football Outsiders stat. The Defense-adjusted Value Over Average ranks players by value per play.
Romo’s value is 11.5 percent, 10th among quarterbacks. If you compare that to the best quarterback in the game, Peyton Manning, you would think Romo isn’t very good, because Manning’s DVOA is 43.5 percent. Nick Foles, who burst onto the scene for Michael Vick and the Eagles this past season was second in DVOA at 35.1.
Pro Football Focus also ranks the position value and has Romo at 11.0, 13th at his position. There were three weeks where Romo received negative grades, Weeks 6, 8 and 10. The Cowboys went 1-2 in those games.
So what do these stats, mean? It depends who is evaluating.
Based on analytics, it looks like Romo is one of the best players at his position. But does Romo’s skill set match Peyton Manning?
What about Eli Manning?
I’d still take Eli's skills over Romo’s.
Is Romo’s skill set better than Ryan? Foles? Rodgers? Newton?
It's an interesting debate, but every quarterback has his own set of skills that help them succeed. Romo has a strong arm, toughness, an excellent feel for when defenders are near him and a command of his offense.
The negative regarding Romo is that I don’t think he makes people around him better. Did Romo really help Terrell Owens, Patrick Crayton, Roy Williams and Miles Austin become better players?
If anything, some of those receivers regressed with Romo at quarterback.
The new batch of receivers: Dez Bryant, Terrance Williams and Cole Beasley are young and talented with tremendous upside. But their quarterback might have already hit his peak.
Bryant was going to be good regardless who the quarterback was. Beasley and Williams need to be utilized more.
Romo is a good quarterback, you can rank him in the top 15 at his position.
But basing that solely on analytics or solely on skill set is a mistake.
Analytics play a role when evaluating talent. But you can't look at the numbers without taking into account what your eyes tell you.