- Joined
- Jun 28, 2010
- Messages
- 48,187
- Name
- Burger man
http://insider.espn.go.com/nfl/story/_/id/9907461/nfl-upset-watch-nfl-week-9
Upset Watch appears every Thursday for ESPN Insider, as Football Outsiders uses a proprietary formula to forecast the expected point spread of each game based on current DVOA ratings (explained here) and, early in the season, our DVOA projections. Each week we highlight the most likely upset on a game with a line over three points, plus an additional game where a significant underdog has a strong chance to cover.
Once again, a slate of just 13 games limits our ability to search for a good upset in Week 9. Nevertheless, a number of home underdogs have intriguing opportunities to put up surprise victories. Based on Football Outsiders stats, the most likely upsets of the week (three-point spread or greater) are:
• Buffalo (+3) vs. Kansas City
• New York Jets (+6) vs. New Orleans
• Pittsburgh (+6.5) at New England
• St. Louis (+3) vs. Tennessee
The game we ended up choosing for this week's Upset Watch -- as well as our choice for this week's Cover Watch -- hangs on the question of just how much a team loses when it is forced to go to its backup quarterback.
Upset Watch: St. Louis (+3) vs. Tennessee
As noted above, our upset picks are generally based on Football Outsiders' DVOA ratings, which in turn are based on past performance of this season. However, what do you do if injuries suggest a team will not be as good as its past performance?
Right now, Tennessee and St. Louis are nearly identical in overall DVOA. Tennessee is at minus-9.6 percent, but St. Louis isn't far behind at minus-14.8 percent. A small gap like that suggests the line should be St. Louis minus-3, not the other way around -- especially when you consider St. Louis has one of the league's strongest home-field advantages. (An FO study a couple of years ago showed that over the previous decade, the four NFC West teams had the four biggest gaps between performance at home and on the road.)
Ah, but that St. Louis team that rated as slightly below average had Sam Bradford at quarterback. This team has Kellen Clemens. Can we account for that difference? Our research has shown that a replacement-level quarterback is typically 13.3 percent of DVOA worse than the starter. Obviously, this isn't true for all teams -- the drop from Peyton Manning to Brock Osweiler would be monumental -- but Bradford is basically an average NFL starter, and Clemens is your typical off-the-bench scrub.
If we take roughly 13.3 percent DVOA out of the Rams' rating, we end up with a gap of 18.5 percentage points of DVOA, which is roughly equal to the usual home-field advantage of 17 percentage points of DVOA. In other words, even if we correct for Clemens, we end up projecting the two versions of St. Louis as equal, which still makes the Rams' plus-3 a good play with a strong chance of an outright upset.
That upset is much likelier if Zac Stacy's ankle is healthy enough to allow him to start. The Rams' offense is a dismal 29th in rushing DVOA, but that's almost all due to the struggles ofDaryl Richardson. Stacy is averaging 4.6 yards per carry and ranks 10th in DVOA out of 37 backs with at least 70 carries. Richardson has just 3.1 yards per carry and ranks 33rd in DVOA. A strong running game would take advantage of the fact that Tennessee's defense is 11th in DVOA against the pass but just 25th in DVOA against the run.
The Rams' offensive strategy should also include plenty of opportunities for Jared Cook to get revenge on his former teammates. The Titans have the best defense in the league against opposing No. 1 receivers, in part because Alterraun Verner has been outstanding this year, but they rank just 27th against tight ends. And the tight ends they've had the biggest problem with are the big seam-splitters similar to Cook. They gave up 73 yards to Kellen Winslow, 62 yards to Vernon Davis and 55 yards plus a touchdown to Antonio Gates.
The Rams' defense is the opposite of the Titans': better against the run than the pass. That's not a good matchup for a Titans team that wants to ground and pound; the Titans rank fifth in the league running 46 percent of the time in the first half (i.e., when the score doesn't generally impact play calling).
The Rams also can win this game by keeping Tennessee to field goals instead of touchdowns. The Rams' defense is ninth in DVOA in the red zone, including seventh against the pass. The Titans' offense is 19th in the red zone, including 22nd against the pass.
It will be interesting to watch the Titans on third down, where there's a huge discrepancy between the Football Outsiders ratings and the official NFL numbers. Our ratings have Tennessee as the league's fourth-best offense on third down. If you check the official NFL stats, this will sound absurd, because the Titans are near the bottom of the league with a conversion rate of just 33 percent on third down. However, there are a couple of other issues to consider here.
First of all, the Titans tend to struggle converting third downs because of their problems on first and second down. They average 7.7 yards to go on third down, which ranks 28th in the NFL. Given that average, a 33 percent conversion rate isn't that bad. In addition, the Titans don't turn the ball over on third down. They have no picks and just three fumbles (one of which was a blown snap). We don't often talk about turnovers on third down, but limiting turnovers is just as much a part of the game on third down as it is on first and second. If you fail to move the chains, you would generally rather be able to punt down the field than hand it to the defense closer to the line of scrimmage.
One final element favoring the Rams: better special teams, giving them an advantage if this game becomes a field-position battle. The Titans rank 29th in Football Outsiders' special-teams ratings and are below average in all five aspects of special teams that we track. St. Louis, on the other hand, ranks 12th in special teams. The Rams are particularly good on kickoffs (first in value) and punts (second). However, they're unlikely to turn the game on a single play, because the special-teams plays that turn a game are usually returns. The Rams are average on kick returns, and first-round pick Tavon Austin has been awful on punt returns. He ranks last in the league, with returns worth an estimated minus-9.5 points worth of field position compared to an average NFL return man.
Cover Watch: Chicago (+10.5) at Green Bay
Here's another game where injuries are a factor, although this time it's about more than just the quarterback. The Bears aren't just starting Josh McCown instead of Jay Cutler. They're also stuck with fourth-round rookie Khaseem Greene instead of Lance Briggs at outside linebacker, while cornerback Charles Tillman is questionable with a knee injury.
Those are three huge injuries. But are they enough to make up for the fact that Chicago is a surprising sixth in overall DVOA this season, narrowly ahead of the Packers? The Packers are better on offense, of course, but the Bears are superior on defense and special teams. (Green Bay is particularly bad on kickoffs, which could mean a big opportunity for Devin Hester.)
You can't expect McCown to play as well as he did when he came in for Cutler and put up a 94.6 Total QBR against Washington two weeks ago. But he should still be able to take advantage of Green Bay's difficulty stopping offenses on first down. The Bears rank fifth in offensive DVOA on first down, while the Packers are 27th on defense. The real issue here isn't runs but rather passes, as the Packers are one of just six teams to allow more than 8 net yards per pass attempt on first down. The Bears also have an advantage in the red zone, where their offense is fifth and the Packers rank last. All of these elements suggest a closer game than expected, and a line over 10 points is just a bit too big for this matchup.
Upset Watch appears every Thursday for ESPN Insider, as Football Outsiders uses a proprietary formula to forecast the expected point spread of each game based on current DVOA ratings (explained here) and, early in the season, our DVOA projections. Each week we highlight the most likely upset on a game with a line over three points, plus an additional game where a significant underdog has a strong chance to cover.
Once again, a slate of just 13 games limits our ability to search for a good upset in Week 9. Nevertheless, a number of home underdogs have intriguing opportunities to put up surprise victories. Based on Football Outsiders stats, the most likely upsets of the week (three-point spread or greater) are:
• Buffalo (+3) vs. Kansas City
• New York Jets (+6) vs. New Orleans
• Pittsburgh (+6.5) at New England
• St. Louis (+3) vs. Tennessee
The game we ended up choosing for this week's Upset Watch -- as well as our choice for this week's Cover Watch -- hangs on the question of just how much a team loses when it is forced to go to its backup quarterback.
Upset Watch: St. Louis (+3) vs. Tennessee
As noted above, our upset picks are generally based on Football Outsiders' DVOA ratings, which in turn are based on past performance of this season. However, what do you do if injuries suggest a team will not be as good as its past performance?
Right now, Tennessee and St. Louis are nearly identical in overall DVOA. Tennessee is at minus-9.6 percent, but St. Louis isn't far behind at minus-14.8 percent. A small gap like that suggests the line should be St. Louis minus-3, not the other way around -- especially when you consider St. Louis has one of the league's strongest home-field advantages. (An FO study a couple of years ago showed that over the previous decade, the four NFC West teams had the four biggest gaps between performance at home and on the road.)
Ah, but that St. Louis team that rated as slightly below average had Sam Bradford at quarterback. This team has Kellen Clemens. Can we account for that difference? Our research has shown that a replacement-level quarterback is typically 13.3 percent of DVOA worse than the starter. Obviously, this isn't true for all teams -- the drop from Peyton Manning to Brock Osweiler would be monumental -- but Bradford is basically an average NFL starter, and Clemens is your typical off-the-bench scrub.
If we take roughly 13.3 percent DVOA out of the Rams' rating, we end up with a gap of 18.5 percentage points of DVOA, which is roughly equal to the usual home-field advantage of 17 percentage points of DVOA. In other words, even if we correct for Clemens, we end up projecting the two versions of St. Louis as equal, which still makes the Rams' plus-3 a good play with a strong chance of an outright upset.
That upset is much likelier if Zac Stacy's ankle is healthy enough to allow him to start. The Rams' offense is a dismal 29th in rushing DVOA, but that's almost all due to the struggles ofDaryl Richardson. Stacy is averaging 4.6 yards per carry and ranks 10th in DVOA out of 37 backs with at least 70 carries. Richardson has just 3.1 yards per carry and ranks 33rd in DVOA. A strong running game would take advantage of the fact that Tennessee's defense is 11th in DVOA against the pass but just 25th in DVOA against the run.
The Rams' offensive strategy should also include plenty of opportunities for Jared Cook to get revenge on his former teammates. The Titans have the best defense in the league against opposing No. 1 receivers, in part because Alterraun Verner has been outstanding this year, but they rank just 27th against tight ends. And the tight ends they've had the biggest problem with are the big seam-splitters similar to Cook. They gave up 73 yards to Kellen Winslow, 62 yards to Vernon Davis and 55 yards plus a touchdown to Antonio Gates.
The Rams' defense is the opposite of the Titans': better against the run than the pass. That's not a good matchup for a Titans team that wants to ground and pound; the Titans rank fifth in the league running 46 percent of the time in the first half (i.e., when the score doesn't generally impact play calling).
The Rams also can win this game by keeping Tennessee to field goals instead of touchdowns. The Rams' defense is ninth in DVOA in the red zone, including seventh against the pass. The Titans' offense is 19th in the red zone, including 22nd against the pass.
It will be interesting to watch the Titans on third down, where there's a huge discrepancy between the Football Outsiders ratings and the official NFL numbers. Our ratings have Tennessee as the league's fourth-best offense on third down. If you check the official NFL stats, this will sound absurd, because the Titans are near the bottom of the league with a conversion rate of just 33 percent on third down. However, there are a couple of other issues to consider here.
First of all, the Titans tend to struggle converting third downs because of their problems on first and second down. They average 7.7 yards to go on third down, which ranks 28th in the NFL. Given that average, a 33 percent conversion rate isn't that bad. In addition, the Titans don't turn the ball over on third down. They have no picks and just three fumbles (one of which was a blown snap). We don't often talk about turnovers on third down, but limiting turnovers is just as much a part of the game on third down as it is on first and second. If you fail to move the chains, you would generally rather be able to punt down the field than hand it to the defense closer to the line of scrimmage.
One final element favoring the Rams: better special teams, giving them an advantage if this game becomes a field-position battle. The Titans rank 29th in Football Outsiders' special-teams ratings and are below average in all five aspects of special teams that we track. St. Louis, on the other hand, ranks 12th in special teams. The Rams are particularly good on kickoffs (first in value) and punts (second). However, they're unlikely to turn the game on a single play, because the special-teams plays that turn a game are usually returns. The Rams are average on kick returns, and first-round pick Tavon Austin has been awful on punt returns. He ranks last in the league, with returns worth an estimated minus-9.5 points worth of field position compared to an average NFL return man.
Cover Watch: Chicago (+10.5) at Green Bay
Here's another game where injuries are a factor, although this time it's about more than just the quarterback. The Bears aren't just starting Josh McCown instead of Jay Cutler. They're also stuck with fourth-round rookie Khaseem Greene instead of Lance Briggs at outside linebacker, while cornerback Charles Tillman is questionable with a knee injury.
Those are three huge injuries. But are they enough to make up for the fact that Chicago is a surprising sixth in overall DVOA this season, narrowly ahead of the Packers? The Packers are better on offense, of course, but the Bears are superior on defense and special teams. (Green Bay is particularly bad on kickoffs, which could mean a big opportunity for Devin Hester.)
You can't expect McCown to play as well as he did when he came in for Cutler and put up a 94.6 Total QBR against Washington two weeks ago. But he should still be able to take advantage of Green Bay's difficulty stopping offenses on first down. The Bears rank fifth in offensive DVOA on first down, while the Packers are 27th on defense. The real issue here isn't runs but rather passes, as the Packers are one of just six teams to allow more than 8 net yards per pass attempt on first down. The Bears also have an advantage in the red zone, where their offense is fifth and the Packers rank last. All of these elements suggest a closer game than expected, and a line over 10 points is just a bit too big for this matchup.
