by Aaron Schatz
The AFC Wild Card round presents us with two interesting games. Saturday's game looks like an even pairing, but may be a mismatch. Sunday's game looks like a mismatch, but may be an even pairing.
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. Any game charting data that appears with a bold asterisk appears courtesy of the ESPN Stats & Information Group and is complete through the end of the season. Other game charting data (such as defensive back coverage stats) is roughly 85 percent complete. Between DVOA splits and game charting data, there was a ridiculous amount of information to go through, so I'm sure I commited the sin of omission when it comes to at least a couple of interesting things. Occupational hazard, I suppose.
Some portions of the Cincinnati-Houston preview appeared as an article on ESPN Insider on Thursday.
For a scouting perspective, make sure to also check out Andy Benoit's AFC Film Room.
Cincinnati at Houston
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.
Although Houston won two more games than Cincinnati, these two teams ended up with almost the same DVOA rating for the season. We have Houston ranked 11th and Cincinnati 12th. However, as you may know by now, these two teams are also going in very opposite directions. Houston ended up with a weighted DVOA of -3.0%, 19th in the league. Cincinnati ended up with a weighted DVOA of 18.0%, which moves them up to seventh.
So one of the big questions about this game is whether that momentum means anything. Both FO writer Vince Verhei (on ESPN Insider) and former FO writer Bill Barnwell (on Grantland) argued pretty well in recent weeks that being the "hot team" at the end of the season doesn't have much bearing on how much success a team will have in the postseason. In general, this is true. Conventional media storylines tend to dramatically overemphasize how well a team played in its last three or four games. However, there are differences in momentum and then there are DIFFERENCES IN MOMENTUM, and the difference between the Bengals and Texans is large enough to deserve bold, italics, and all caps.
[ad placeholder 4]
First of all, when Football Outsiders is comparing these teams with weighted DVOA, we're not just looking at the last three or four games. The weighted DVOA formula gives full or almost-full strength to a team's last eight weeks, and then partial credit to the six weeks before that. So when I say that the Bengals have been playing better than the Texans in recent weeks, I'm talking about a much larger sample size than what most people consider "recent weeks." That larger sample makes these differences in weighted DVOA a little easier to trust than a statement like "Wow, Indianapolis sure is going into the playoffs hot because of the way they beat Houston last week."
And it's hard to go back in time and find another game where the difference in weighted DVOA (21.0% for CIN-HOU) was so much bigger than the difference in regular DVOA (0.5% for CIN-HOU). In fact, there has only been one such game in 21 years of playoffs since 1991. That came in 2002, when the Jets hosted the Colts in the Wild Card round. The 10-6 Colts had a better record than the 9-7 Jets, but DVOA said that the Jets were a signifcantly better team, with a DVOA gap of 17.3%. The Jets were also the much hotter team, since they had started 2-5 before finishing 7-2, and the gap between the two teams in weighted DVOA was colossal, 41.8%. The result pretty much matched what DVOA would have expected, although only I knew this at the time because I had just invented DVOA a week or two previously. The Jets blew out the Colts 41-0.
Of course, since the Jets were better with regular DVOA too, that doesn't tell us too much about momentum. A better comparison would be games where the two teams were close in regular DVOA but farther apart in weighted DVOA. So I went and found all the games from the first two weeks of the playoffs where the gap between the teams was less than 5.0% in regular DVOA but bigger than 10.0% in weighted DVOA. These results also suggest that end-season momentum does mean something, as long as the difference is large and our description of "momentum" considers more than just a couple of game
|Hot Teams vs. Cold Teams with Similar Total DVOA, 1991-2012 Playoffs|
| Result for
Of course, this is a really small sample, and the results aren't unquestionably persuasive. The "cold team" actually won two of these games on the road. The third win by a "cold team" is the most astonishing, the 1992 Divisional round game where Miami shut out San Diego 31-0. Does anyone out there remember what the press coverage was like going into that game? It was a bit before my time as a football fan, so I have no memory of it. Looking back, the Chargers had started the year 0-4 then ripped off 11 wins in 12 games to win the AFC West. They won five of their last six games by at least two touchdowns, then shut out Kansas City 17-0 in the first round of the playoffs. Miami, on the other hand, started 6-0 and then finished 5-5. Their last two games of the season were a 19-17 win over the 4-12 Jets and a 16-13 overtime win over the 2-14 Patriots. The momentum gap between these teams was colossal. Then Stan Humphries went out and threw three interceptions in the first 21 minutes and that was pretty much the ballgame.
By the way, to digress, Dave Krieg was the starting quarterback for the Chiefs when they lost to San Diego in that playoff game. I had no clue Dave Krieg ever played for Kansas City.
Let's go on now and look at the specific ways the Bengals have improved and the Texans have declined since they both had the week off in Week 8, and how those things match up.
WHEN THE BENGALS HAVE THE BALL
Actually, the Bengals haven't improved much on offense, going from -3.2% DVOA before their bye to -0.8% afterwards. And when you compare their offense before and after the bye week, it comes out the opposite of the Texans. Cincinnati has actually declined a bit passing the ball, but the running game has improved from -10.9% DVOA before Week 8 (22nd, for 4.06 yards per carry) to 5.1% DVOA since Week 8 (seventh, for 4.51 yards per carry).
Meanwhile, Houston's run defense has actually improved despite losing inside linebacker Brian Cushing in mid-October. On the surface, this doesn't seem to be the case, because the Texans have gone from allowing 4.0 yards per carry before the bye to 4.3 yards per carry since. However, when you look at the circumstances and distance to go of these runs, it (mostly) turns out the Texans have been more successful despite allowing more yardage: 13th in run defense DVOA before the bye, and fourth since.
I say "mostly" because the Texans are just getting killed by runs on third down. Since their bye week, the Texans have allowed an average of 12 yards on quarterback scrambles and 4.5 yards on other runs.
The Bengals should just forget about sending BenJarvus Green-Ellis wide around the ends. On right-end runs, the Bengals are 19th in ALY while the Texans' defense is fourth. It's even worse on the left, where the Bengals rank 31st and the Texans are first.
The real issue for the Texans is pass defense, in particular the coverage. The pass rush has been getting to the quarterback about as often as it did early in the year, but the Texans are having more problems when they only rush the conventional four. That's more likely to give the opposing quarterback time to find a guy left open by Houston's struggling defensive backs:
|Texans by Number of Pass Rushers in 2012*|
|Weeks 1-7||Weeks 9-17|
|4 pass rushers||125||4.3||158||6.7|
|5 pass rushers||74||5.3||104||6.0|
|6 pass rushers||48||6.0||55||5.1|
Part of the problem here is an injury to Johnathan Joseph, their best cornerback last year. In the first three games this season, Joseph played very well, allowing just 3.2 yards per pass according to our game charting stats. Then he tweaked his groin, and he hasn't been the same since, allowing 8.2 yards per pass. Kareem Jackson has improved greatly in his second year, and our current charting numbers have him with 6.5 yards per pass allowed (25th) and 62 percent success rate (11th). Nonetheless, it's hard to know which of these two guys is going to be covering A.J. Green, because there doesn't seem much rhyme or reason to the Texans' secondary assignments. Jackson and Joseph don't strictly play with each on one side, but neither does one of them specifically move around to face the opponent's best receiver. Despite the cornerbacks not taking consistent assignments, the Texans have a very strong split where they rank fourth in DVOA against opponent's No. 1 receivers but 28th in DVOA against No. 2 receivers.
Other injuries have the Texans reaching deep into their roster to find nickelbacks. Brice McCain is out for the season. Alan Ball was next man up, and he struggled. Now he's dealing with a foot injury, and that puts Brandon Harris and Roc Carmichael on the field, and they've really struggled. As a result, FO metrics show that the Texans have gone from the league's third-best defense against slot receivers in Weeks 1-7 to the third-worst in Weeks 9-17.
One other interesting split to note is the way the Bengals got significantly different results based on personnel group. Even though they ran most often when using conventional 21 personnel, they also had their highest average yards per play from that grouping. If you are curious about the runs from 01 personnel, i.e. no running backs, those are often plays with Andrew Hawkins or the now-injured Mohamed Sanu in the backfield.
|Bengals Offense by Personnel Group, 2012*|
WHEN THE TEXANS HAVE THE BALL
Aha, here's the problem. The Texans offense has gone from 10.8% DVOA in Weeks 1-7 (eighth) to -8.5% DVOA in Weeks 9-17 (20th). At the same time, the Cincinnati defense has gone from 17.3% DVOA (27th) to -20.0% DVOA. Since Week 9, only the Chicago Bears had a better defensive DVOA than the Cincinnati Bengals, and no team rated better on pass defense.
The Bengals pass defense has improved in pretty much every way possible. They've improved significantly on every down-and-distance situation except for second-and-short, and they've improved against every kind of receiver except for the No. 3/"other" receivers. That's a place they generally struggled all season, finishing just 24th in DVOA, but Houston isn't exactly the right team to take advantage, because after Andre Johnson and Kevin Walter they have a bunch of guys who you have never heard of, and most of them were not good this year. Keshawn Martin had 28 passes with -47.7% DVOA. DeVier Posey had 14 passes with -37.1% DVOA. Lestar Jean had 32.8% DVOA, but that was only on a dozen pass attempts. It's basically the effect of two plays, a 46-yard gain against Denver in Week 3 and a 54-yard touchdown against Tennessee in Week 13.
What about the Texans offense -- what changed after the bye week? Well, the clearest problem for the Texans' offense has been third down, as their ability to convert has completely crumbled over the last few weeks. In the first seven weeks, the Texans converted 46 percent of the time on third or fourth down. Since their bye week, they've converted only 32 percent of the time.
The Texans have actually struggled on third-and-long all season, but it's the short- and medium-distance chances that have become an issue:
|Texans Offense on Third/Fourth Down in 2012|
|Weeks 1-7||Weeks 9-17|
|1-2 to go||66%||4.29||59%||0.86|
|3-6 to go||55%||7.07||32%||4.34|
|7+ to go||22%||5.88||23%||5.76|
Part of the problem is that the Texans are passing too much in short-yardage situations. In Weeks 1-7, Foster and the other Texans backs combined for 24 carries on third or fourth down with 1-2 yards to go, and Schaub threw 11 passes. In Weeks 9-17, the running backs combined for just 10 carries, and Schaub threw 10 passes.
You would think that running it up the gut against the Bengals is a terrible proposition because of the presence of Geno Atkins. The Texans may have J.J. Watt, but Atkins gives the Bengals their own transcendent defensive lineman. You may remember a post from earlier this week which talked about J.J. Watt obliterating the record for defensive Defeats with 56. Well, Atkins nearly set a new record for Defeats by a defensive tackle. He finished with 30, which trails only the 32 Defeats put up by Darrell Russell of the 1998 Oakland Raiders. (Warren Sapp in 2000 and Ndamukong Suh in 2010 are tied for third at 29).
And yet, here's the amazing thing -- overall, the Bengals were terrible preventing runs up the gut. Cincinnati ranked 20th in Adjusted Line Yards against runs up the middle, and they were dead last allowing 82 percent success converting runs on third or fourth down with 1-2 yards to go. I have no idea how this is possibly schematically -- we'll have to get Andy Benoit to look at that if the Bengals move on to the next round -- but it's pretty clear that the Atkins isn't the problem against the run that the Texans might expect.
What about when third down is a clear passing situation? Should the Bengals should blitz Schaub to force mistakes on third down? Actually, no. Schaub has faced five or more pass rushers on 44 third or fourth downs. He converted 58 percent of those for a new set of downs. When the opponent only sends four pass rushers on third or fourth down, he converts just 34 percent of the time.
Some of the problems on third-and-long come because the Texans lose their abililty to go play-action. The Texans use play-action on 25.5 percent of pass plays, tied for seventh in the NFL*. Like most teams, they gained about a yard more per play when using play-action; they were actually better than that in previous years. And the Bengals do struggle against play-action passing. They allowed 1.6 yards per play more when the offense used play-action, one of the higher figures in the league. Even after their bye week, when the defense improved so much, the Bengals allowed 1.3 yards per play more against play action than they did overall. Opponents apparently understood this, because the Bengals faced play-action on 26.2 percent of pass plays, the most in the league.
The Texans' dependence on play action doesn't just make it hard for them to pass on third-and-long. It also makes them a team not well-equipped to come back from large deficits, or at least it does theoretically. The Texans ranked 26th in offensive DVOA when losing by more than a touchdown. Of course, this theory gets a bit confused when you also notice that the Texans struggled this year with a small lead, ranking 27th in DVOA when winning by 1-8 points. They were good when tied or slightly behind (10th in DVOA) and best when playing with a comfortable lead of more than a touchdown (seventh in DVOA).
Finally, here's a unit that hasn't declined since Houston's bye week -- because the Texans' special teams have been awful all year long. Houston finished last in the FO special teams ratings by a healthy margin, mostly because of issues on kickoffs. After removing onside attempts and end-of-half squibs, the Texans had only 39.0 net yards on the average kickoff. The NFL average was 42.8, and every other team was at 40.9 or higher. The Texans allowed a league-high 15 returns of 30 yards or more, and two of those were touchdowns. We estimate that after adjusting for weather, the Texans were lost 25.9 points worth of field position on kickoffs when compared to an average team. That's twice as bad as any other team.
The Texans also have to live with the fact that Shayne Graham can't hit field goals over 45 with any consistency whatsoever. Graham went 6-for-13 on field goals over 42 yards, and 12 of those attempts were indoors.
The Bengals were one of the better special teams units according to our ratings, particularly when it came to gaining field position on punts. Kevin Huber hit 12 punts that landed inside the 5 without a return, the most in the league. He also had only seven touchbacks, and three of those were on punts where he started behind or at his own 40.
Like I wrote at the beginning of this preview, this isn't your typical case of one team being hot going into the playoffs and the other team being cold. This is a huge difference that has lasted for two months. That's enough to at least balance out the home-field advantage and make this game a 50-50 proposition, and it's probably enough to make Cincinnati the favorites. J.J. Watt is dominant, but the man can't do it all alone. Plus, that Geno Atkins guy can play a little bit too.
Indianapolis at Baltimore
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.
If you are reading Football Outsiders, you know that our metrics say the Colts were massively overrated and lucky this season. The Colts finished 25th overall in DVOA, with the worst rating ever for a team with 10 or more wins. So on the surface, it sure seems like the Baltimore Ravens should be massive favorites in this game.
However, there are three problems with making that an easy conclusion: one intangible, one possibly coincidental, and one very mathematical.
The intangible reason: #chuckstrong.
The coincidental reason: This is weird and I have no way to explain it, but eight of the ten worst playoff teams by DVOA actually won their first playoff game, not counting this year's Colts. Four of those eight teams won on the road.
|Worst Playoff Teams by DVOA, 1991-2012|
|2004||STL||8-8||-27.2%||31||W 27-20 at SEA, L 47-17 at ATL|
|2010||SEA||7-9||-22.9%||30||W 41-36 vs. NO, L 35-24 at CHI|
|1998||ARI||9-7||-17.1%||25||W 20-7 at DAL, L 41-21 at MIN|
|1996||MIN||9-7||-14.5%||24||L 40-15 at DAL|
|2006||SEA||9-7||-13.0%||24||W 21-20 vs. DAL, L 27-24 at CHI|
|1996||IND||9-7||-12.2%||22||L 42-14 at PIT|
|2011||DEN||8-8||-11.8%||24||W 29-23 vs. PIT (OT), L 45-10 at NE|
|1995||IND||9-7||-9.8%||23||W 35-20 at SD, W 10-7 at KC, L 20-16 at PIT|
|2004||MIN||8-8||-8.1%||20||W 31-17 at GB, L 27-14 at PHI|
|2000||MIN||11-5||-6.3%||22||W 34-16 vs. NO, L 41-0 at NYG|
The mathematical reason: The 2012 Baltimore Ravens just weren't really very good. They were essentially an average team on offense and defense that happened to have the league's best special teams. Special teams do matter, but they have a much higher variability than offense or defense. You can't trust special teams to get you a win on their own. While the Colts are the weakest team in the AFC field, the Ravens are the second-weakest if you take out special teams.
WHEN THE COLTS HAVE THE BALL
The Baltimore Ravens have had one of the NFL's strongest defenses for a very, very long time. Baltimore ranked sixth or higher in defensive DVOA every single year from 1999 to 2011. That's a period of thirteen straight seasons. This year, that run ended. The Ravens didn't just have a defense that was less than excellent; they had a defense that DVOA rated as below average.
A big reason for this decline was injury, of course. The Ravens go into this game healthier on defense than they have been all season, but healthier doesn't mean perfectly healthy. A lot of important defenders are expected to play, but given their appearance on the injury list, it's hard to imagine that they are at 100 percent. Ray Lewis should finally return from the torn triceps suffered in Week 6, but it's hard to imagine he's all the way back. Terrell Suggs is listed as probable with a biceps injury, and he's been in and out of the lineup since returning from the Achilles tear that cost him the first half of the season. Ed Reed is also listed as probable with a shoulder injury. Haloti Ngata, Bernard Pollard, and Pernell McPhee are all listed as questionable.
One important player who isn't coming back is Lardarius Webb, who looked like a lockdown corner in the early weeks of the season and then tore his ACL. Former Bears special teams dynamo Corey Graham has actually been a reasonable replacement on the left side (offensive right). Graham played almost entirely in the last half of the season, so he has a lot of plays still uncharted, but the numbers we have so far give Graham a 64 percent Success Rate, fourth in the NFL among corners with at least 32 charted targets. He's allowed 6.1 yards per pass, which ranks 17th. On the offensive left, however, our charting numbers suggest that Cary Williams gives up a lot of successful shorter completions. Williams has allowed 7.3 yards per pass, close to league average, but his 43 percent Success Rate is 82nd out of 93 cornerbacks. He's spent most of the season giving up a steady stream of 8-to-12-yard completions, although with very little yards after the catch. His average of 2.0 YAC allowed is one of the ten lowest for starting corners. Still, this doesn't sound like a particularly strong skill set if you have to cover Reggie Wayne, master of the 10-yard curl. Wayne was targeted this year on 43 passes that went between 8 and 12 yards in the air, more than any receiver except Brandon Marshall.
Chykie Brown has replaced Jimmy Smith as the primary nickelback, and we don't really have enough charting data on him to make any kind of judgment.
The cornerback who gets to cover Donnie Avery will probably come out looking better than the one who has to cover T.Y. Hilton; Avery is the starter, but the rookie has much better numbers. Hilton was 26th with a 10.4% DVOA, while Avery was 80th out of 86 wide receivers with -18.9% DVOA. Their average route length was similar (Hilton 12.4 yards, Avery 12.9 yards) but Hilton had a better catch rate (56 percent vs. 48 percent) and a lot more yards after the catch (7.6 YAC vs. 3.5 YAC).
Andrew Luck is a phenomenal talent, and definitely had a strong rookie year, but the FO numbers don't support the argument that Luck should be in the Rookie of the Year conversation with Robert Griffin and Russell Wilson. Luck was just 19th with -5.2% DVOA. He finishes much better in ESPN's Total QBR rating because of his habit of making clutch fourth-quarter comebacks. Whether you believe that's a sustainable talent is up to you.
Other than the fourth quarter, the place Andrew Luck really stood out this year was third-and-long. However, it's also where the Ravens stand out compared to other NFL defenses. Both the Colts' offense and the Ravens' defense rank number one in DVOA when it is third or fourth down with 7+ yards to go.
However, the Colts are also very good on third-and-short, and the Ravens defense is not. That starts with short-yardge running situations. The Colts converted 72 percent of these situations, second in the NFL. The Ravens allowed conversions on 76 percent, which ranked 29th. What's odd is that the Colts were very good at preventing stuffs at the line when they absolutely needed a yard, but otherwise they got stuffed at the line all the time. Only Arizona and Philadelphia saw its running backs stuffed more often for a loss or no gain.
Baltimore's success against offenses in third-and-long situations ties into one of the most interesting statistical splits I found while preparing for the AFC playoffs. The Ravens' defense ranked fifth in DVOA against teams in the shotgun (-10.0%) but 30th against plays where the quarterback was under center (9.1%). They were one of only two defenses (along with Denver) that allowed more yards per play on plays with the quarterback under center (5.3) than they did on plays with the quarterback in shotgun (5.2).
I asked Andy Benoit about how this split might tie into the X's and O's and he had two theories. First, that shotgun spread formations are when the Ravens really like to use quarters coverage; yesterday's Film Room piece dissected why this is so effective for the Ravens' defense. Second, shotgun lessens the likelihood that opponents can use a play fake, and the Ravens were just killed by play fakes this year, in part because their linebackers aren't particularly strong when it comes to play recognition. Baltimore allowed 8.6 yards per pass when opponents used play-action, compared to 6.1 yards per pass overall. That gap of 2.5 yards is the highest of any defense in the league.*
How do these Ravens weaknesses match up with the Colts' tendencies? Not particularly well, but not particularly poorly either. The Colts used shotgun 45 percent of the time, which ranked 18th in the league, and like most teams were about 10 percentage points worth of DVOA better when they were in shotgun. They used play action 18 percent of the time, which is a little bit less than the NFL average, and gained roughly a yard more per play, which is about the NFL average.
The Ravens could use blitzes to disrupt Luck when he goes spread shotgun, as he was awful against big blitzes in 2012. Luck gained 6.5 yards per play against three or four pass rushers and 6.8 yards per play against five pass rushers, but only 3.7 yards per play against a big blitz of six or more.*
WHEN THE RAVENS HAVE THE BALL
This is where the Colts have a problem, because their defense was awful this year. It's even worse than it looked on film, because the Colts played the league's easiest schedule of opposing offenses. The Colts couldn't stop the run. They couldn't stop the pass. And they couldn't rush the passer: even though Robert Mathis and Dwight Freeney are still on this team, the Colts ranked 27th in Adjusted Sack Rate.
Looking for relative strengths for the Colts defense is difficult. The Colts best down-and-distance split, for example, is third-and-long, where they rank 16th in DVOA. Unfortunately, that's also the Ravens' best down-and-distance split, where they rank fourth.
(This doesn't mean that the Ravens are better off getting stuck in third-and-long, of course. The Ravens convert 30 percent of the time on third or fourth down with 7+ yards to go. They rank only 28th in DVOA on third-and-short, but even still, convert those situations 55 percent of the time.)
The other split where the Colts defense stands out is on our "defense vs. receivers" table, where they rank sixth preventing success on passes to running backs in the passing game. So Ray Rice is unlikely to pick up 30 yards on a fourth-and-29 dumpoff this week. That's okay, though, because there's plenty of opportunity for him on the ground, where Rice had one of his better years by FO stats (seventh in DVOA) although his success rate was a lousy 44 percent (33rd out of 42 qualifying backs).
That poor success rate suggests that Rice was a "boom-and-bust" back this season, but when facing the Colts, running backs generally became "boom-and-not-bust." Somehow the Colts kept Adrian Peterson to 60 yards in Week 2, but that was a massive aberration. The Colts rarely stuff runners at the line (30th in the NFL) and are horrible when it comes to preventing long runs. Their 1.49 Open Field Yards per Carry is dead last in the NFL and tied with the 2008 Lions for the worst figure since 1995. Our Adjusted Line Yards metrics suggest that the Ravens have a significant edge running up the middle as well as on the right side.
Questions of left vs. right also play a role in figuring out how the Ravens can have success in the passing game, because Joe Flacco this season was remarkably better throwing to his right than he was to his left.
|Joe Flacco Passes by Direction, 2012 (includes DPI)|
The game charting metrics suggest there hasn't been much difference between Indy's two starting corners, youngster Cassius Vaughn (8.3 yd/pass, 44% suc rate) and veteran Vontae Davis (10.8 yd/pass, 46% suc rate). However, most of our currently compiled charting data for Davis comes from the first half of the season, before he missed four games with a knee injury. In small sample size, he's been better since returning. Certainly if we go by reputation, Vaughn is the weaker of the two, and he's usually on the offensive right. That could be a problem against Flacco.
By the way, if you haven't watched the Colts much this year, get ready for the new era. With former Ravens defensive coordinator Chuck Pagano and his hybrid defense, the Colts have gone from the team that never blitzes to a team that sends five or more pass rushers 39 percent of the time, tied for fourth in the NFL.*
[ad placeholder 3]
Here's how the Ravens have been winning games all year. The Ravens had one of the ten best special teams units in DVOA history. They finished in the top ten in all five areas of special teams that we track. Justin Tucker was particularly strong, averaging over 70 yards per kickoff. Kick returns are even better if we only look at Jacoby Jones (16.0 estimated points above average) and not Deonte Thompson, who returned kicks in September (-2.7 estimated points below average).
This is where we would normally continue by talking about how the Colts have poor special teams just like every other year, except this year they didn't. The new regime found some rookies who could actually cover kicks and punts, and T.Y. Hilton was a good return man, so the Colts finished 12th in special teams DVOA. The Colts had not ranked that high in special teams since 1996. Nonetheless, special teams significantly favor the Ravens, and they're more likely than the Colts to make one big special teams play to turn the game.
When it comes "feel," there's a lot pointing towards the Colts here. They're on this great run, they're a great story, and you've got the weird record of bad teams pulling upsets in the playoffs. But when you look at things objectively, you have to feel the Ravens are the clear favorite here. No, they aren't a particularly great team, but they were a lot better than the Colts during the regular season, and they get the home field advantage. Luck should be able to have success against the still injury-riddled Ravens defense, but it's not likely to match the success that Joe Flacco and Ray Rice can have against the Colts. Things are particularly set up for long highlight runs by Rice and big shot plays downfield to Torrey Smith, the kind of stuff that breaks open a ballgame.
Most fans want to see the Chuck Pagano win, but the people of Baltimore want to see the Irsays lose. The people of Baltimore are more likely to get their wish.
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 rolling average of the last five games. Note that even though the chart appears in the section for when each team has the ball, it represents total performance, not just offense.