DMN Blog: Football Outsiders: How are the Cowboys the NFC East's worst team?
by Tim MacMahon
1. If the NFC East standings went by the Football Outsider team efficiency rankings, the Cowboys would be in last place. They're seventh overall, and it's tough to argue that the No. 1 Giants and No. 5 Redskins have been more impressive through five games. How the heck are the 2-3 Eagles ranked third?
It's a tricky one, a combination of several factors. Some of the issue is luck -- the Eagles' opponents have gone nine-for-nine on field goals, including several 50+ yarders. The success rate of opposing field goals, of course, is totally random from year-to-year. They've also been subject to some long turnover returns, which is also random, while their turnover returns have been shorter.
The biggest thing, though, is how they've played in their games. Among the NFC East teams, who can say they've blown out a very good team? Washington's won four games by a total of 20 points. Dallas beat Philadelphia by four, Green Bay by 11 (in their most impressive performance of the year), and Cincinnati by nine. The Giants have blown out Seattle (the best performance of any team in the league so far by DVOA) and dominated the Redskins in a game they won by nine. They also had a close win against Cincinnati and a blowout against St. Louis.
Philadelphia, meanwhile, dominated Pittsburgh in a nine-point win the same way the Giants handled the Redskins, making their opposition look silly. They lost to Dallas by four on the road, lost to Chicago by four on the road, and lost to the Redskins by six points. They've looked very impressive in their two wins (the other a blowout of the Rams), and we give them credit for outplaying the Cowboys despite losing, perhaps owing to two 47+ yard field goals by Nick Folk.
Of course, shoulda coulda doesn't mean you can go back in time and change what happened. DVOA's pretty nifty at using these sort of things to predict future performance, though, so I'm pretty confident that the Eagles are still going to have a say in the playoff picture in the NFC.
2. Jason Witten ranks first by a large margin among tight ends in your Defense-adjusted Yard Above Replacement stat (DYAR) but only eighth in your Defense-adjusted Value Over Average stat (DVOA). Why such a big difference? Which formula do you consider more indicative of a player's performance?
The difference is volume. DVOA is like batting average -- if a guy gets one throw in his direction and he catches it and goes for 40 yards, his DVOA is going to be off the charts! His DYAR, on the other hand, will recognize that he had one great play, but that there's only so much damage you can do on a single play; in that sense, it's more like runs scored or RBI.
DYAR's a much better indicator of how a player is affecting a team on the whole; if a guy is consistently dropping passes or picking up solid chunks of yardage, DYAR will tell you so. Witten's so far above everyone else in DYAR because he's been thrown 48 passes; the only other tight end to be thrown more than 40 is Tony Gonzalez, who hasn't caught enough of the ones thrown to him to be valuable.
3. Felix Jones is off the charts in those metrics for running backs, although he doesn't have enough carries to qualify for the RB rankings. How do his numbers so far compare with the best change-of-pace backs since FO started tracking those statistical formulas?
He sure does look fantastic so far -- in light of the idea that he had a pretty middling Speed Score, it will be interesting to see if he can become an everydown back or at least a guy who picks up 150 carries a year. If he can do that at the rate he's performing at now, he'd be an incredibly valuable player.
Jones' rushing DVOA at the moment is an absurd 87.7%. That will drop some once we begin to adjust for opponent, but it would be the best performance by any back with ten or more carries since 1995, the beginning of the DVOA era.
The current holder of that title is Jerry Ellison, a halfback for the 1995 Tampa Bay Buccaneers who gained 218 yards on 26 carries, scoring five touchdowns. He only had 48 mediocre carries over the rest of his career, though.
Other guys in the top 10 include DeDe Dorsey and Ahmad Bradshaw last year, Stephen Davis in 1996, Moe Williams' impressive 2002 campaign where he scored 11 times in 84 carries, Brian Mitchell's 1995 season, and surprisingly, two years from a forgotten former Cowboy -- Michael Wiley.
After a very mediocre 2000, Wiley had huge numbers in 2001 and 2002, putting up a DVOA of 48.4% and 58.0%, respectively. He averaged over seven yards a carry in both seasons. He was out of the league by 2003, but obviously, the man still had at least something to offer.
4. The Cowboys have the league's best TD-to-drive ratio (.341). How does that compare to their ratio last season? What's the best ratio since FO has tracked that stat?
Last year, Dallas was 4th in the league, averaging .282 touchdowns per drive. The best that we've seen is the 2007 Patriots, who averaged .424 touchdowns per drive, as well as a league-leading 3.37 points per drive. Kinda scary. So far this year, the Giants lead the league with 3.08 points per drive. These stats take out drives like kneeldowns and Hail Mary's, so they're a good way of judging how well a team is playing when they're actually trying to score.
5. I have Larry Fitzgerald on my fantasy team. Can you crunch some numbers and tell me what kind of stats he'll put up against the Cowboys secondary?
Unfortunately, our stats aren't quite that advanced yet. What I do for my ESPN column on fantasy matchups each week is run a regression analysis based upon several factors (like opposing defense, performance against similar wide receivers, likelihood of a team to be winning/losing in the second half, injuries on offense and defense, etc) to spit out a percentage value above or below the player's expected fantasy performance against an average team, as determined by KUBIAK, our fantasy football algorithm. Yes, it's named after Gary Kubiak. It's a long story.
This week, we have Fitzgerald at eight percent above average. That's primarily due to two things. The first is the expected outcome and pace of the game; the Cowboys are likely going to win this game, which means that the Cardinals will be throwing the ball to catch up. That also likely indicates a fast-paced game, meaning more drives (and points) for each team.
Second is injuries -- it looks like the Cowboys will be missing Terence Newman this week, and at the time of writing that column, it appeared that Adam Jones would be suspended. At this point, who knows whether he'll play or how good he'll be. Either way, Fitzgerald's the better receiver, and while he'll see more double coverage without Anquan Boldin, he's still the Cardinals' only real target downfield. If I had to guess what his line will be... I'll say seven catches, 118 yards, and a score. Maybe two.
1. If the NFC East standings went by the Football Outsider team efficiency rankings, the Cowboys would be in last place. They're seventh overall, and it's tough to argue that the No. 1 Giants and No. 5 Redskins have been more impressive through five games. How the heck are the 2-3 Eagles ranked third?
It's a tricky one, a combination of several factors. Some of the issue is luck -- the Eagles' opponents have gone nine-for-nine on field goals, including several 50+ yarders. The success rate of opposing field goals, of course, is totally random from year-to-year. They've also been subject to some long turnover returns, which is also random, while their turnover returns have been shorter.
The biggest thing, though, is how they've played in their games. Among the NFC East teams, who can say they've blown out a very good team? Washington's won four games by a total of 20 points. Dallas beat Philadelphia by four, Green Bay by 11 (in their most impressive performance of the year), and Cincinnati by nine. The Giants have blown out Seattle (the best performance of any team in the league so far by DVOA) and dominated the Redskins in a game they won by nine. They also had a close win against Cincinnati and a blowout against St. Louis.
Philadelphia, meanwhile, dominated Pittsburgh in a nine-point win the same way the Giants handled the Redskins, making their opposition look silly. They lost to Dallas by four on the road, lost to Chicago by four on the road, and lost to the Redskins by six points. They've looked very impressive in their two wins (the other a blowout of the Rams), and we give them credit for outplaying the Cowboys despite losing, perhaps owing to two 47+ yard field goals by Nick Folk.
Of course, shoulda coulda doesn't mean you can go back in time and change what happened. DVOA's pretty nifty at using these sort of things to predict future performance, though, so I'm pretty confident that the Eagles are still going to have a say in the playoff picture in the NFC.
2. Jason Witten ranks first by a large margin among tight ends in your Defense-adjusted Yard Above Replacement stat (DYAR) but only eighth in your Defense-adjusted Value Over Average stat (DVOA). Why such a big difference? Which formula do you consider more indicative of a player's performance?
The difference is volume. DVOA is like batting average -- if a guy gets one throw in his direction and he catches it and goes for 40 yards, his DVOA is going to be off the charts! His DYAR, on the other hand, will recognize that he had one great play, but that there's only so much damage you can do on a single play; in that sense, it's more like runs scored or RBI.
DYAR's a much better indicator of how a player is affecting a team on the whole; if a guy is consistently dropping passes or picking up solid chunks of yardage, DYAR will tell you so. Witten's so far above everyone else in DYAR because he's been thrown 48 passes; the only other tight end to be thrown more than 40 is Tony Gonzalez, who hasn't caught enough of the ones thrown to him to be valuable.
3. Felix Jones is off the charts in those metrics for running backs, although he doesn't have enough carries to qualify for the RB rankings. How do his numbers so far compare with the best change-of-pace backs since FO started tracking those statistical formulas?
He sure does look fantastic so far -- in light of the idea that he had a pretty middling Speed Score, it will be interesting to see if he can become an everydown back or at least a guy who picks up 150 carries a year. If he can do that at the rate he's performing at now, he'd be an incredibly valuable player.
Jones' rushing DVOA at the moment is an absurd 87.7%. That will drop some once we begin to adjust for opponent, but it would be the best performance by any back with ten or more carries since 1995, the beginning of the DVOA era.
The current holder of that title is Jerry Ellison, a halfback for the 1995 Tampa Bay Buccaneers who gained 218 yards on 26 carries, scoring five touchdowns. He only had 48 mediocre carries over the rest of his career, though.
Other guys in the top 10 include DeDe Dorsey and Ahmad Bradshaw last year, Stephen Davis in 1996, Moe Williams' impressive 2002 campaign where he scored 11 times in 84 carries, Brian Mitchell's 1995 season, and surprisingly, two years from a forgotten former Cowboy -- Michael Wiley.
After a very mediocre 2000, Wiley had huge numbers in 2001 and 2002, putting up a DVOA of 48.4% and 58.0%, respectively. He averaged over seven yards a carry in both seasons. He was out of the league by 2003, but obviously, the man still had at least something to offer.
4. The Cowboys have the league's best TD-to-drive ratio (.341). How does that compare to their ratio last season? What's the best ratio since FO has tracked that stat?
Last year, Dallas was 4th in the league, averaging .282 touchdowns per drive. The best that we've seen is the 2007 Patriots, who averaged .424 touchdowns per drive, as well as a league-leading 3.37 points per drive. Kinda scary. So far this year, the Giants lead the league with 3.08 points per drive. These stats take out drives like kneeldowns and Hail Mary's, so they're a good way of judging how well a team is playing when they're actually trying to score.
5. I have Larry Fitzgerald on my fantasy team. Can you crunch some numbers and tell me what kind of stats he'll put up against the Cowboys secondary?
Unfortunately, our stats aren't quite that advanced yet. What I do for my ESPN column on fantasy matchups each week is run a regression analysis based upon several factors (like opposing defense, performance against similar wide receivers, likelihood of a team to be winning/losing in the second half, injuries on offense and defense, etc) to spit out a percentage value above or below the player's expected fantasy performance against an average team, as determined by KUBIAK, our fantasy football algorithm. Yes, it's named after Gary Kubiak. It's a long story.
This week, we have Fitzgerald at eight percent above average. That's primarily due to two things. The first is the expected outcome and pace of the game; the Cowboys are likely going to win this game, which means that the Cardinals will be throwing the ball to catch up. That also likely indicates a fast-paced game, meaning more drives (and points) for each team.
Second is injuries -- it looks like the Cowboys will be missing Terence Newman this week, and at the time of writing that column, it appeared that Adam Jones would be suspended. At this point, who knows whether he'll play or how good he'll be. Either way, Fitzgerald's the better receiver, and while he'll see more double coverage without Anquan Boldin, he's still the Cardinals' only real target downfield. If I had to guess what his line will be... I'll say seven catches, 118 yards, and a score. Maybe two.
posted by The Gryphon at 11:27 AM
<< Home