TCU has played much better in the second half of games this year. What other schools have seen dramatic shifts of play after halftime?
07 Jan 2011
by Aaron Schatz
On Saturday night, we get a replay of last year's AFC Championship game. On Sunday afternoon, we get a team very few people thought would make the playoffs. Who says the NFL postseason doesn't offer a wide variety of matchups? In both games, the road team actually had a better win-loss record than the home team, but only one of those teams is actually favored to win this weekend.
For those who may be unfamiliar with the Football Outsiders stats, they are explained at the bottom of the page. Scroll down or click this link.
Please note that, unless otherwise specified, Game Charting data we discuss in these pieces is not yet complete. We have all charting data complete through Week 15.
(In the interest of democracy, we are providing a Make Your Own Intro opportunity here. Acknowledge historic abilities of Peyton Manning, injuries on 2010 Colts, and greatness of Darrelle Revis. Then insert joke about Rex Ryan and feet, followed by reference to Manning Mastercard commercial and/or telling kids to go sit in the port-o-let. Optional reference to Mike Westhoff inventing special teams or Bill Polian making comments dismissive of Football Outsiders.)
All readers can click here for in-game discussion on our message boards. If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.
First things first: That week-to-week graph is great fun, isn't it? For the first half of the season, the Jets were playing slightly below average and winning close games over bad teams. In the second half of the season, they were all over the map, including the second-highest single-game DVOA of the year a week ago against Buffalo.
But I digress. The good news for the Jets on Saturday night is that the Colts defense is bad enough to make this a shootout. The difference between this year's Colts and previous Colts defenses is that the biggest holes aren't in the run defense. They're in the pass defense. Not only did the run defense rank higher in the FO ratings, it also significantly improved in the second half of the year: from 29th in DVOA Weeks 1-9 to ninth Weeks 10-17. (The Colts rank only 28th in Adjusted Line Yards; the main discrepancy seems to be that opponents tended to run for medium-length but unsuccessful gains on second-and-long as well as third-and-long.)
At the same time, the pass defense went in the other direction: from 12th in Weeks 1-9 to 29th in Weeks 10-17. The Colts pass defense is reasonably average on third downs, but poor on first downs and really awful (31st) on second downs. It isn't any particular cornerback: Kelvin Hayden has given up more big plays than Jerraud Powers, but the two starters have roughly the same Success Rate. Surprisingly, it seems the pass rush may be to blame. Dwight Freeney and Robert Mathis are fine -- they combined for 21 sacks this year and had plenty of hurries. However, partly because they so rarely blitz, nobody else on the Colts really ever gets to the quarterback. Defensive tackle Eric Foster had 3.5 sacks, and every other player on the Colts roster combined for a grand total of 4.5 sacks. That's it. (Compare this to, say, the St. Louis Rams. James Hall and Chris Long combined for 18.5 sacks, while the rest of the team had 24.) The Colts were 30th in Adjusted Sack Rate this year and 29th in 2009. Slide two linemen over to Freeney, leave in a tight end or back to help with Mathis, and the pass protection should be fine.
Another issue is the team's young linebackers. The Colts were 31st against passes to tight ends and 29th against passes to running backs, which could mean a big game for Dustin Keller and perhaps even LaDainian Tomlinson. However, linebacker is also the unit where getting people back from injury probably has the biggest effect. The most important injured players on offense -- Dallas Clark, Austin Collie -- are gone for good. But while Clint Session is probably still out this week, Gary Brackett has been back for a couple weeks in the middle of the defense. In the first half of the year, with Brackett healthy, the Colts allowed -4.8% DVOA to tight ends. Since Week 10, with Brackett out three weeks and then still hobbled by turf toe for a couple more, the Colts have allowed 45.8% DVOA to tight ends.
When the Jets do run, the ALY numbers suggest they will have the most success up the middle. The Jets are sixth in ALY on runs up the gut, while the Colts are 28th on defense. However, when those runs up the gut come in short-yardage situations, the Colts did pretty well this year. The Jets were second converting power runs on offense, but the Colts were eighth against power runs on defense.
Note that while the Jets are generally not built to come back from big deficits (23rd in DVOA when losing by more than a touchdown), the Colts aren't built to protect big leads (29th in defensive DVOA when leading by more than a touchdown).
The Jets were a dominant second in DVOA run defense, and a still respectable seventh in pass defense. That's a drop from last year, but not a big one. However, the secret is to spread them out, go shotgun, and get them to bring in the nickel and dime backs. The Jets allowed 4.0 net yards per play when opponents were in conventional formations, best in the league. They allowed 6.5 net yards per play when opponents were in shotgun, 29th in the league. That difference of 2.5 yards is the highest in the league. And the Colts went shotgun on 58 percent of plays this year, more often than any team except Detroit.
A look at the FO game charting stats shows you why you want to bring in three and four wide receivers against the Jets, even if those third and fourth wide receivers are Blair White and Taj Smith. (No, you've never heard of him, and you won't see him, because the Colts don't go four-wide, they go three-wide with a flexed tight end.) Darrelle Revis suffered from hamstring issues this year and overall was not as good as 2009, not a shock since that was a season for the ages. He was still pretty good, allowing 6.1 net yards per pass with 67 percent Success Rate. Antonio Cromartie was also good, with 5.9 net yards per pass and 62 percent Success Rate. However, Drew Coleman and Kyle Wilson combined for a 48 percent Success Rate with 7.9 net yards allowed per pass.
The big man to look for in this game may be Pierre Garcon. It was Garcon and Austin Collie who carved up the Jets last year while Revis was draped over Reggie Wayne. This season, the Jets are 24th covering number-two receivers. And Garcon's numbers have improved dramatically over the last few weeks. In the first 10 games, Garcon had a catch rate of 47 percent and -26.8% DVOA. Then, Manning began including him in the pre-game route tree-running drills he runs with Wayne (and used to run with Marvin Harrison). Perhaps coincidentally, perhaps not, Garcon has been on fire since then. In the last six games, Garcon had a 69 percent catch rate and 19.1% DVOA.
The passing issue becomes even bigger in the red zone, where the Colts had the best red-zone passing DVOA in the NFL and the Jets ranked 25th against passing. The Colts generally score seven, not three.
What about Rex Ryan dialing up the pressure, you say? Well, my colleague Bill Barnwell covered that issue fairly substantially in last year's AFC Championship preview. Although Manning has had a mix of good and bad games against Ryan defenses, blitzing hasn't significantly affected the proceedings, with very few hurries, hits, or sacks. Last year's game wasn't any different. We unfortunately don't have a chart on the first half -- it seems to be the only half of 2009 football we're missing from charting -- but in the second half, the Jets only hurried Manning twice. He completed one of those passes anyway, turning a Bryan Thomas hurry into a 15-yard gain by Pierre Garcon. The Jets big-blitzed Manning seven times in that second half, and he hit four out of six passes, all for first downs, while the seventh pass got a first down with an illegal contact penalty.
And if you think the Jets are going to sack Manning on third down, dream on. The Jets' ASR is 9.8 percent on first and second down, but just 1.0 percent on third down. The Colts' offensive ASR is 3.7 percent on first and second down, and 0.0 percent on third down. Yes, zero-point-zero. Once we adjust for the fact that sacks generally are more frequent on third down, Manning is relatively un-sackable. He has been sacked on third down only three times all year, and not once since Week 9.
One other quirky stat: The Jets have the fourth-best defense in the league in the first half, and the best defense in the league in the fourth quarter, but rank only 21st in the third quarter. Oddly enough, the third quarter is also the worst quarter for the Colts offense, although the differences are smaller.
This part of the game is all Gang Green. The Jets were fifth in our special teams ratings, while the Colts -- as is their wont -- were awful, ranking 31st. Special teams may be the least consistent unit in the NFL, but not for the Colts. They've had dismal special teams for a decade. As a result of this, the Jets start their average drive only 67.3 yards away from a touchdown (second in NFL) while the Colts start their average drive 72.5 yards away from a touchdown (30th).
The biggest contrast is on kick returns, where the Jets were number one and the Colts were dead last. The Jets should also have a huge field position advantage on punts, where they ranked second while the Colts were 31st on punt returns. Steve Weatherford was actually number one on gross punt value. Oddly enough, the one area of special teams where the Jets struggled -- field goals -- is the only area where the Colts were above average. Adam Vinatieri had a good year, but do remember that he doesn't have a strong leg. He's only attempted five field goals of 50+ yards in his five seasons with Indianapolis, and only connected on two of them.
Most of the talk about this game will concentrate on strength vs. strength: Peyton Manning against the Jets defense. If the Jets pull off an upset in this game, it will because of weakness vs. weakness: Mark Sanchez against the Colts pass defense. Remember that Sanchez won't have to go as far as Manning will in order to score points. It may seem like the Jets declined in the second half of the year, but really they just became more inconsistent. For some reason, it seems harder to remember their good games than to remember the bad ones. The Colts may have home-field advantage, but the Jets were the better team this year overall.
All readers can click here for in-game discussion on our message boards. If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.
Remember that website "Hot or Not?" In order to figure out which of these teams will move on to the Divisional Round, we need to play "Kansas City Chiefs: Hot or Not?"
At first glance, it looks like the Chiefs enter the playoffs with reasonable momentum. Sure, they lost to Oakland last week, but it was a game that mattered only for playoff seeding. In Weeks 10-16, when they had to win in order to salt away the AFC West title, the Chiefs went 5-2. On the other hand, look at who the Chiefs have been winning against. Their opponents in those seven weeks included three NFC West teams, Denver twice, and a Tennessee team which declined significantly after midseason.
If we look at the Football Outsiders DVOA ratings, the importance of schedule trumps the basic totals of wins and losses. Since Week 10, Kansas City's ratings have collapsed in pretty much every category.
The biggest problem has been defense. In the first eight weeks of the season, the Chiefs ranked 11th against the pass and sixth against the run. In the last eight weeks, they ranked 24th against the pass and dead last against the run. The offense has also declined, although the decline has been smaller: from ninth passing and seventh rushing to 19th passing and 16th rushing. (They would be 14th in both categories if we took out the loss started by backup Brodie Croyle.)
Oddly enough, Baltimore's offense has suffered from a similar, but smaller, drop. The Ravens ranked third passing and 13th rushing through Week 9; since, they are 12th passing and 21st rushing. Those struggles on the ground may be the biggest secret about this year's Ravens defense: We think of them as a run-first offense, but this year, at least, they've been far more efficient when they put the ball in the air.
When the Ravens acquired T.J. Houshmandzadeh before the season, many of us expected them to run more spread sets with three wide receivers. Actually, they haven't. We have them with three or more wideouts on 43 percent of plays this year, actually slightly less than last year's 44 percent. And while Anquan Boldin may draw the most attention in coverage, Derrick Mason has been the Ravens' best receiver this year, 14th in DVOA among receivers with at least 50 targets.
That brings up the question of which Kansas City cornerback is covering who, but it may not be as important as we expect. Brandon Flowers is widely acknowledged as the top dog in the Kansas City secondary, but when we look closer at the numbers it is astonishing how much of that value came in the first three weeks of the season. In the first three weeks, we recorded Flowers with a ridiculous 95 percent Success Rate. He allowed only two completions over 10 yards, and one of those was a failed conversion on third-and-20. But since the Kansas City bye in Week 4, our numbers have Flowers as a very average player, perhaps in part due to a hamstring injury suffered in Week 11. The numbers tell a pretty astonishing story about the Kansas City defense after the bye week:
|Charting Stats for Kansas City Cornerbacks, Weeks 1-15|
| Suc Rate
| Suc Rate
Arenas' yards allowed per pass figure is low because he's generally covering slot receivers, but that Success Rate since Week 5 is awful. If the Chiefs end up with Arenas lined up on Boldin in the slot, it is not going to be pretty.
Have we bored you yet with our non-stop comparisons of Thomas Jones and Jamaal Charles? Too bad, because I'm going to do it again. Charles led the league in both DYAR and DVOA, running for a near-record 6.4 yards per carry. Thomas Jones ranked 44th in DYAR and 42nd in DVOA, gaining just 3.7 yards per carry. Yet somehow, Jones actually had 15 more carries than Charles did. As an added bonus, Charles caught 45 passes for 468 yards and three touchdowns, while Jones had just 10 receptions for 122 yards and no touchdowns. Look, nobody is more cognizant of running back workload issues than we are here at Football Outsiders, but you guys can take the chains off now. Charles still has 70 carries to go just to hit 300. It is okay to give him three-quarters of the carries in the playoffs. Please, for the sake of your own fans and our sanity.
Charles' best bet for big gains is to run around right end. Kansas City ranked second in the NFL with 5.47 ALY on right end runs. The Ravens' run defense was 13th in ALY overall, but next-to-last on runs around right end (4.98 ALY). Jones' best bet for big gains is to find a time machine that will take him back to the 2008 Jets, but barring that miracle, his best bet is to run up the middle. The Chiefs line had 4.2 ALY when Jones was running up the middle, 3.6 ALY when he was running to either side.
In the passing game, as Vince Verhei pointed out in this week's Any Given Sunday, Dwayne Bowe is pretty much a one-man show. Bowe was responsible for 36.4 percent of Kansas City's receiving yardage, second in the NFL behind Roddy White. The Ravens did pretty well against number-one receivers this year, seventh in DVOA. And talented rookie tight end Tony Moeaki is going to run into trouble against a defense that ranked second against opposition tight ends.
One surprising fact about the Chiefs: They went shotgun a lot less than in Todd Haley's first year as head coach. Last year, the Chiefs ran shotgun 47 percent of the time, but they were actually pretty terrible from shotgun. This year, they ran shotgun on just 31 percent of plays, and once again they were a better offense from standard formations, probably because of the quality of their running game and the questionable quality of the receivers down the depth chart.
So, at some point we need to address the Billy Cundiff touchbacks thing, and this is as good a time as any. Cundiff has been the best kickoff man in the league this year, by leaps and bounds. He has hit touchbacks on 40 of 79 kickoffs, 53 percent. FO metrics have his kickoffs worth 15.0 points of field position over average; Olindo Mare is second at 7.7 points. (If you look at the Baltimore net kickoff value above, you'll notice their coverage team was actually slightly below average.) This kind of value is ridiculous, of course, especially when you consider that Cundiff had only 11 touchbacks in his entire career before this season, on over 200 kickoffs.
Has anything like this ever happened before? Well, no, and sort of. First, no: Since the kickoff line was moved back from the 35 to the 30-yard line in 1994, the highest year-over-year increase in touchback percentage (min. 25 kickoffs each season) belongs to Michael Husted, who hit only 14 percent of his kickoffs in both 1995 and 1996. In 1997, that went up to 36 percent of kickoffs. Still, that's a 22 percent rise. Not really comparable.
Before 1993, we have two instances that get almost somewhat close to Cundiff's crazy rise. Brad Daluiso went from 36 percent touchbacks as a rookie in 1991, mostly with Buffalo, to 80 percent touchbacks in 1992 with... oh yeah, Denver. So that 44 percent rise makes a little sense. The other rise belonged to Jason Hanson of Detroit, and this one is really strange. Hanson had 22 percent touchbacks as a rookie in 1992. He never had a year higher than 33 percent after the kickoff was moved back to the 30-yard line in 1994. But in his second season, Hanson hit 46 touchbacks on 79 kickoffs. That's 59 percent, even better than Cundiff this season.
Fluke? Real change in ability? Could it be... (gasp) performance-enhancing drugs? Cundiff told Stefan Fatsis of Slate (listen at around 38 minutes in) that it was a combination of technique (trying to strike higher on the ball, to get more distance and less hang time) and more confidence. No, really, he attributed some of this 53 percent rise in kickoff rate to feeling more comfortable in Baltimore. Okay, then. The only explanation I buy is his explanation that he finally has a special teams coach who doesn't berate him for not trying for as much hang time as possible.
As for the other special teamers in this game, Kansas City started off strong on special teams, but everything else except punt returns declined over the rest of the year. Baltimore had the best punting in the league this year, thanks to Sam Koch (second in gross punt value behind Weatherford), and Cundiff was good on field goals to go with his crazy kickoff season, but the Ravens are mediocre on returns.
Football Outsiders has been driving the Kansas City bandwagon since before the season began, but this is where we get off. Although weighted DVOA isn't much more accurate than total DVOA when it comes to predicting playoff results, it is hard to ignore how much the Chiefs have declined since midseason. Their record is built off a schedule that would rank as the easiest in the league if not for the San Diego Chargers, and they've been extremely inconsistent, particularly in the last two months. Baltimore, on the other hand, has been very consistent, with negative DVOA in only two games all year -- both against Cincinnati. Even with their home-field advantage, one of the biggest in the NFL, the Chiefs are clear underdogs in this game. Their chances of winning are bigger than Seattle's chances of beating New Orleans over in the NFC -- but honestly, not by much.
DVOA (Defense-adjusted Value Over Average) breaks down each play of the season and compares it to the NFL average based on situation and opponent. You'll find it explained further here. Since DVOA measures ability to score, a negative DVOA indicates a better defense and worse offense, and a positive DVOA indicates a better offense and worse defense.
SPECIAL TEAMS numbers are different; they represent value in points of extra field position gained compared to NFL average. Field goal rating represents points scored compared to average kicker at same distances. All special teams numbers are adjusted by weather and altitude; the total is then translated into DVOA so it can be compared to offense and defense. Those numbers are explained here.
Each team is listed with DVOA for offense and defense, total along with rush and pass, and rank among the 32 teams in parentheses. (If the DVOA values are difficult to understand, it is easy to just look at the ranks.) We also list red zone DVOA and WEIGHTED DVOA (WEI DVOA), which is based on a formula which drops the value of games early in the season to get a better idea of how teams are playing now (explained here).
Each team also gets a chart showing their performance this year, game-by-game, according to total DVOA. In addition to a line showing each game, another line shows the team's trend for the season, using a third-power polynomial trendline. That's fancy talk for "the curve shifts direction once or twice." Note that even though the chart appears in the section for when each team has the ball, it represents total performance, not just offense.
26 comments, Last at 11 Jan 2011, 1:58pm by tshark