RotoWire Partners

East Coast Offense: Defense Wins Championships

Chris Liss

Chris Liss

Chris Liss is RotoWire's Managing Editor and Host of RotoWIre Fantasy Sports Today on Sirius XM radio.

I had no idea what I was going to write about today until I got an email from Mike Salfino with his sack adjusted YPA rankings. It's just yards per passing attempt (adjusted for sacks, i.e., a sack counts as an attempt for negative yards) minus yards per passing attempt allowed by the defense. It measures net passing efficiency, and because passing is about two thirds of total offense, it's a good barometer of net productivity generally. The reason we look at per-attempt numbers rather than total yards is that the latter includes a lot of noise from lopsided games when teams are gaining and giving up big yards in garbage time.

Here's Mike's list:

Rank Team YPA-O YPA-D YPA-Net
1 San Diego (12-3) 8.5 6.2 2.3
2 Indianapolis (14-1) 7.7 5.8 1.9
3 Philadelphia (11-4) 7.5 5.8 1.7
4 New Orleans (13-2) 8.2 6.6 1.6
5 Houston (8-7) 7.9 6.4 1.5
6 Pittsburgh (8-7) 7.8 6.3 1.5
7 Green Bay (10-5) 7.5 6 1.5
8 Dallas (10-5) 7.7 6.4 1.3
9 NY Jets (8-7) 6.2 5.1 1.1
10 New England (10-5) 7.6 6.5 1.1
11 NY Giants (8-7) 7.5 6.6 0.9
12 Minnesota (11-4) 7.3 6.6 0.7
13 Arizona (10-5) 6.9 6.2 0.7
14 Denver (8-7) 6.4 5.7 0.7
15 Baltimore (8-7) 6.8 6.3 0.5
16 Cincinnati (10-5) 6.3 6 0.3
17 Washington (4-11) 6.6 6.4 0.2
18 Buffalo (5-10) 5.8 5.9 -0.1
19 Chicago (6-9) 6.1 6.3 -0.2
20 Carolina (7-8) 6 6.3 -0.3
21 Tennessee (7-8) 6.4 6.9 -0.5
22 Jacksonville (7-8) 6.6 7.4 -0.8
23 San Francisco (7-8) 5.7 6.6 -0.9
24 Atlanta (8-7) 6.3 7.3 -1
25 Seattle (5-10) 5.8 7 -1.2
26 Tampa Bay (3-12) 5.7 6.9 -1.2
27 Kansas City (3-12) 5.3 7.2 -1.9
28 Miami (7-8) 5.8 7.8 -2
29 Detroit (2-13) 5.3 7.8 -2.5
30 St Louis (1-14) 5.2 7.7 -2.5
31 Oakland (5-10) 5.1 7.7 -2.6
32 Cleveland (4-11) 4.6 7.6 -3
Keep in mind as with any formula for team rankings, there are plenty of flaws with it. But San Diego and Indy, not surprisingly rank No. 1 and 2, respectively, and it shows Houston is a dangerous team should it make the playoffs. There are a few obvious issues - Tennessee is way too low at No. 21 because (1) these are year-long stats, and the Titans aren't the same team they were when they were 0-6, and (2) they're the rare team where rushing stats make an enormous difference in overall ranking. Nonetheless, the NFL is a passing league, and this is a good shorthand for team strength generally. But looking at the rankings, another idea jum


Keep in mind as with any formula for team rankings, there are plenty of flaws with it. But San Diego and Indy, not surprisingly rank No. 1 and 2, respectively, and it shows Houston is a dangerous team should it make the playoffs. There are a few obvious issues - Tennessee is way too low at No. 21 because (1) these are year-long stats, and the Titans aren't the same team they were when they were 0-6, and (2) they're the rare team where rushing stats make an enormous difference in overall ranking. Nonetheless, the NFL is a passing league, and this is a good shorthand for team strength generally.

But looking at the rankings, another idea jumped out at me - why do we subtract defensive YPA rather than divide by it? In other words, is the net YPA really more important than the YPA ratio? When you subtract YPA allowed, you're saying that a team that generates 8.0 YPA on offense and allows 7.0 on defense is the same as one that generates 6.0 and allows 5.0. But if you use the ratio, the former is 8/7 (1.14), and the latter is 6/5 (1.2). And you can see why dividing makes more sense at least in some cases when we consider the extremes. If you averaged 13 YPA, but allowed 12 (basically far and away the greatest offense and worst defense ever), you would score on every possession, and so would your opponent. There's absolutely no difference except that you would score on slightly fewer plays. But the team that averaged six and allowed five would have a below average offense, but a great defense and would probably make the playoffs.

The reason for this is that teams get four downs and need to go 10 yards to get a new set of them. If you only had two downs and needed to go 15 yards, then the difference of YPA at the high end might matter more. At a certain point, offensive efficiency is moot - you score in fewer plays, you use less clock, but as long as you can consistently move the chains it doesn't matter. Even a team that got three yards every single play (a terrible average) would be unstoppable. For that reason, we can't just look at net production in a vaccuum - we need to look at it in reference to creating first downs. Because teams can't guarantee their average gain every play, you need a good deal more than three yards per play to get down the field, probably anything up to 10 or so would make a difference, but the further you get from the some reliable minimum needed to generate a new set of downs most of the time, the more the returns diminish. Put differently, the importance of boosting your YPA from six to seven is probably greater than going from seven to eight because of the relatively low bar needed to get a first down. And this cuts the opposite way, too - the importance of reducing your YPA allowed from eight to seven is less than the importance of reducing it from seven to six. (I'm not sure where the exact cut-off point is - clearly going from 4.0 YPA to 5.0 is more important than going from 2.0 to 3.0 where dropping back to pass is completely futile in either case. But if it's 5.0 or 5.5, or even 6.0, you can see that there's a point where a one-yard gain or loss in net efficiency matters most in terms of how it affects your ability to generate first downs.

Which brings me to my final point - in today's NFL, passing is king. There are a whopping 10 teams with a sack-adjusted YPA of 7.5 or better - a number that would have led the league in many seasons. And when you're talking about playoff teams, virtually all of them will have positive Net YPAs. But Net YPA being equal, I'd expect that the teams with the better YPA ratios, i.e., the ones with lesser offenses and better defenses would have better balance at preventing and converting new sets of downs. Those teams are closer to the sweet spot where each incremental gain in offensive YPA or incremental reduction in YPA allowed means the most in terms of continuing and stopping drives.

In other words, Net YPA being equal, the team with the better defense should have an advantage. And when you look at the last 10 Super Bowl winners, most of them - the Giants (at least at the time of the Super Bowl), Steelers twice, Pats twice, Bucs and Ravens were the better defensive team. Only the Colts (and it's arguable with Bob Sanders they were better than the Bears without Tommie Harris), the 2004 Patriots (the Eagles were barely better defensively, and the game was very close) and the 1999 Rams were offense heavy, and in the latter case, St. Louis' Net YPA was so much better than the Titans' on both sides of the ball.

Are the Colts Toast?

I know I just made the argument that Net YPA with an emphasis on defense is the key, and under that criteria, the Colts, with the No. 1 seed locked up, are probably the favorite. But Bill Polian's decision to give away an incredibly rare opportunity at a perfect season is a blunder from which they might not recover.

Peter Schoenke summarized the situation on the RotoSynthesis Blog better than I can here, but the bottom line is that Polian's narrow focus on protecting his players from injury at the expense of an achievement far more historic than winning a Super Bowl has already alienated the fan base and has to be eating away at the team.

The problem is that if the Colts mail it in against the Bills on Sunday (and I don't see how Polian can reverse himself and risk injury now that the perfect season has already been blown), the players will have four weeks to stew over what the organization took away from them. It's possible they somehow channel that frustration positively into their performance on January 16 or 17, but I think there's a better chance the malaise festers. The undefeated season matters because players and fans care about historical achievements whether or not they count in the NFL's particular way of settling things in the postseason. To ignore that is to dismiss their aspiration for unique greatness, and I don't think it's something the Colts - as professional as they are - will shrug off lightly.

Things to Take Away from Week 16

* That the Giants could dominate Washington on the road so completely, and get annihilated so thoroughly at home six days later by the Panthers shows just how great the game-to-game variance is in the NFL. Granted, the Giants have been one of the most Jekyll and Hyde teams in recent memory, but I watched a 14-2 team become a 2-14 team with the exact same personnel. That's got to be a coaching issue, and while it's a done deal that Bill Sheridan gets fired, Tom Coughlin needs to reconnect with this group, too. The handling of Osi Umenyiora coming off ACL surgery this year was also poor. I'd hate to see him leave and play at a Pro Bowl level for another team.

* The Chargers road blowout of the Titans cements them as the favorite to win the Super Bowl right now. Unlike Indy, who's needed fourth-quarter comebacks against every decent team, the Saints - who just lost at home to Tampa, and the Vikings, who have lost three of four - the Chargers are peaking at the right time. Let's call it Chargers-Packers in the SB, though the Pats will be dangerous to SD if they meet, more so than the Colts.

* What happened to Matt Hasselbeck? He seems to be taking the Marc Bulger career path.

* How does a guy like Jerome Harrison go from "too small to carry the load" to getting 39 carries in a meaningless game?

* Jonathan Stewart is a top-10 real life back. DeAngelo Williams is top five. I suppose depth at running back is necessary, but this team would be so much better if it had a Stewart or Williams-level receiver opposite Steve Smith instead.

* While the Raiders high-profile skill position picks have all been busts of late (JaMarcus Russell, Darren McFadden, Darrius Heyward-Bey), they sure have a hell of a punter and kicker.

Things to Watch in Week 17

* Can Philly lock up a first-round bye by winning in Dallas?

* Can the Packers win in Arizona?

* Will New Orleans mail it in at Carolina and go into the playoffs on a three-game losing streak and five-game streak of weak performances? Even if New Orleans goes all out, will it win in Carolina?

* How much will Bill Belichick play his starters at Houston?

* Can the Steelers stay alive in Miami?

Beating the Book

Bears -3 at Lions

This is a nice clean game between two teams not in the hunt that should treat this like any other week. The Bears just won their Super Bowl by upsetting their division-winning rival at home, and now they have to travel to face Detroit - a game for which they probably won't be all that excited. The Lions got beaten in San Francisco last week, but they played hard and will probably show up for this one. Back Detroit.

Lions 27 - 23

We're 10-6 in this forum on the season, 123-115-2 overall (11-5 last week). We were 12-5 in this forum last year, but 124-122 on the season overall. From 1999-2008 we're 1308-1140 (53.4%, not including ties).

The full article comes out on Thursday morning.

Surviving Week 17

Last week, I took San Francisco who won easily, but of the big favorites only New Orleans went down, and I doubt too many people had them available.

For this week, I'd take the Ravens if they were available (they're not). Next I'd probably take the Broncos (my pick) at home against the Chiefs. Denver still has playoff hopes, but I don't trust them that much against a division rival in what could be a cold weather game (bad weather levels the playing field). After that, probably the Jets at home against the Bengals - as I'm not sure Cincy will play its starters. Then the Vikings (vs. Giants), Texans (vs. Pats) and Bills (vs. Colts). Of course, I reserve the right to change my mind before the full article comes out Wednesday night.