Possibly the closest Super Bowl matchup in history also poses the question: how much does it mean when certain aspects of an NFL team improve dramatically in the second half of the season?
15 Jan 2008
by Aaron Schatz
Here are postseason DVOA ratings, including the first two rounds of the playoffs. Like last year, based on reader requests, we've ranked all 32 teams, whether they are in the playoffs or not. All numbers are weighted DVOA. That means that Weeks 1-5 are not included, while Weeks 6-11 are somewhat discounted. Any team that did not appear in the postseason is considered to have two bye weeks. A number of people have asked why I don't just end the weighted DVOA ratings for non-playoff teams at Week 17 instead of Week 19. The reason is that we already did this -- those numbers are the normal end-of-season weighted DVOA ratings. Just substitute those for these when it comes to non-playoff teams, and there you go.
Just to get this out of the way: Yes, the Giants are still very low. I'll discuss this a bit after we run the numbers.
There will be the usual comments over on AOL sometime on Wednesday, and they will be linked from the "FO Goes Mainstream" page. Playoff odds will be updated soon. You'll find the matchup pages available for the two championship games in the DVOA Premium database.
* * * * *
To save people some time, we remind everyone to put their angry troll hatred into the official zlionsfan angry troll hatred Mad Libs form:
<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>
I'm not going to bother to run the whole DVOA explanation; if you are new to the website, you can read about it here. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
|W-L|| WEI OFF
| WEI DEF
| WEI S.T.
|W-L|| WEI OFF
| WEI DEF
| WEI S.T.
By beating the team that was previously ranked second (Jacksonville), the Patriots open up a healthy lead over the rest of the league that once again resembles their lead over the rest of the league during the regular season. The shocking change is that they move up from 18th to eighth in defense, but it isn't as shocking as it seems. The Week 7 game where the Pats let Miami score a bunch of meaningless points in the second half drops in weight in the formula, and the Patriots played a good game against Jacksonville. The defenses were so bunched up this year that those two small things are enough to raise the Pats 10 places in the rankings.
Before we talk about the New York Giants, here are the one-game DVOA ratings for the second round of the playoffs. Remember that these include opponent adjustments.
In the end, the Colts-Chargers game wasn't anywhere near as "unlucky" as I thought while I was watching it. The Chargers were definitively better according to DVOA (although DVOA doesn't do anything to adjust for the weird bounce that put that Kenton Keith screen in the arms of Eric Weddle). One of the odd things from this game is that San Diego won despite a huge advantage on special teams for Indianapolis -- even though this was San Diego's clear advantage over Indianapolis going into the game.
As you can see, it was a tremendously offense-driven week. Seattle was the only team with a below-average offense, and that was just running the ball; they were positive passing the ball. Part of the reason: As you know from reading Quick Reads, only two offensive holding flags were thrown all weekend, not counting flags during special teams or interception returns.
So, let's talk about the New York Giants. At the close of the regular season, the DVOA ratings said that the Giants were clearly the worst team to make the playoffs. New York was the only playoff team that didn't end the season among our top dozen teams. Two weeks later, the Giants are in the NFC Championship game after becoming the first NFC team to knock off a number-one seed in the Divisional round since the playoffs went to six teams in 1990.
Are you waiting for me to say something about this being unprecedented? Turns out it isn't. This Giants run is entirely precedented. Not only have we been through this before, but we went through this just six months after Football Outsiders went online.
|2003 Panthers (11-5)||NFL Rank||2007 Giants (10-6)||NFL Rank|
|Special Teams DVOA||0.6%||16||1.1%||14|
|Defense: Adjusted Sack Rate||7.3%||5||8.8%||1|
|Defense: Adjusted Line Yards||3.84||6||3.83||3|
|Turnover Differential (regular season)||-5||26||-9||26|
|Turnover Differential (playoffs)||+8||+3 (so far)|
|Top Offensive Weapon||RB tandem featuring veteran power back and shiftier rookie||RB tandem featuring veteran power back and shiftier rookie|
|Top Defensive Weapon||Pass rush||Pass rush|
|Eventual Result||Lost Super Bowl to New England||Unknown|
Not everything with these two teams is an exact match, of course. The 2007 Giants played a harder schedule. Their offensive line does much better according to Adjusted Line Yards. Their veteran power back is only a three-year veteran. The 2003 Panthers had better and healthier wide receivers, and a young defense-oriented coach instead of an older defense-oriented coach. The quarterback was an undrafted free agent, not a number one overall pick. Still, a consistently mediocre team that slips by with some close wins over bad opponents, then turns into a postseason juggernaut with no warning? We've been through this before.
A number of readers have asked whether the Giants' postseason success has revealed a major flaw in our advanced statistics. Now imagine the reaction if you had never heard of DVOA until two or three months previous, and this newfangled metric was based solely on information from a single season worth of play-by-play.
In the long run, it turned out that Carolina's playoff success didn't reveal any flaws in DVOA. I've worked on the formula many times, upgraded it, tried different things, and each time the 2003 Carolina Panthers come out as a mediocre football team during the regular season. The success of the 2007 Giants doesn't reveal flaws in the DVOA formula either. A team that outscored its opponents by only 22 points and lost the turnover battle for the season is one of the top four teams in football? Where's the indicator supposed to be for that? Is it supposed to be their 7-1 record on the road during the regular season? The 2001 New York Jets went 7-1 on the road and 3-5 at home, exactly the same as the 2007 Giants. They didn't go on a run through the playoffs; they went to Oakland for the wild card round and lost 38-24.
There is one reasonable idea here, which is that perhaps the weighted DVOA formula needs to consider fewer games come playoff time. That's certainly worth testing. However, we're still left with the question of when we were supposed to realize that the 2007 Giants were about to turn it around. Was one strong game against New England the indicator that the Giants were taking a leap forward? Plenty of teams have finished the year with a very strong game, only to get spanked in the first round of the playoffs. Two games? Perhaps, given that one of them is a postseason game. The weighted DVOA formula wasn't created with postseason games in mind. It is certainly worth exploring.
However, the most likely explanation for the 2007 Giants is that sometimes a mediocre team will go on a run where everything clicks, and the team plays really well for a couple of games. In other sports, a two-game winning streak over superior opponents is not a big deal. In football, because every playoff round consists of one game instead of seven, that little winning streak looks like a colossal achievement. The Giants making it to the NFC Championship is not much different from when the NCAA invites seven or eight different Big Ten teams to March Madness, and the team that surprisingly shows up in the Sweet 16 is the one that barely got in with a tenth seed.
This is what makes sports fun. We don't know what is going to happen. Predictions are about probabilities, not definites, and the better team going in isn't always the better team going out. There's no reason to believe the Giants were a good team before three weeks ago. There's very little reason to believe they will beat Green Bay. Who knows, maybe they will anyway. That's why we play the games.
LATE ADD: The playoff odds are now updated. With only four teams left, I might as well just give them here.
111 comments, Last at 18 Jan 2008, 2:26am by terry