1999 DVOA Ratings and Commentary

by Aaron Schatz

For the past few months, five years worth of DVOA have been available on our website: 2000-2004.  But since the end of last season, you may have noticed me occasionally mention ratings from 1998 and 1999.  I was not able to fully analyze those seasons until this past off-season, and two months ago I published the ratings for 1998. The 1999 ratings were supposed to come two weeks later, but the book tour for PFP 2005 and other circumstances have kept me so busy that I couldn't finish my write-up. With Tennessee and St. Louis meeting this weekend in a rematch of that season's Super Bowl, this seems like the perfect time to look back at 1999: The Year of the Rams.

1999 was a year of change, with only one team from the previous year's top six finishing in 1999's top six. If you are waiting for me to reveal that some other team actually had a higher DVOA than St. Louis in 1999, well, keep waiting.  1999 is one of the three seasons where the top team in DVOA took home the Super Bowl title as well (the others are 2002 and 2004) and the Rams dominated the NFL all season. St. Louis's rating was double that of every other team except for the 14-2 Jacksonville Jaguars, and the Rams had that rating despite sitting their starters in the final week of the season and putting up their worst single-game rating of the year.

Before commentary, let's look at the Football Outsiders team efficiency ratings for 1999, measured by our proprietary Value Over Average (VOA) system that breaks down every single play and compares a team's performance to the league averaged based on situation in order to determine value over average.  (Explained further here.)

DVOA represents adjusted statistics.  OFFENSE and DEFENSE DVOA are adjusted for opponent quality and to consider all fumbles, kept or lost, as equal value.  SPECIAL DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.  NON-ADJ TOTAL VOA does not include these adjustments.  DVOA is a better indicator of team quality.  VOA is a better indicator of actual wins.  WEIGHTED DVOA gives a stronger consideration to games late in the season.  Remember that, as always, defense is better when it is NEGATIVE. LAST YEAR represents rank in 1998.

1 STL 42.2% 13-3 24 50.8% 33.3% 1 18.3% 1 -21.3% 4 2.6% 9
2 JAC 32.7% 14-2 10 41.0% 22.1% 5 4.9% 9 -21.1% 5 6.7% 2
3 TEN 21.0% 13-3 14 26.1% 23.7% 3 8.8% 6 -8.5% 18 3.7% 4
4 OAK 20.5% 8-8 23 14.0% 24.2% 2 15.9% 2 -10.4% 13 -5.8% 28
5 DAL 17.7% 8-8 6 23.5% 22.0% 6 -0.3% 11 -15.0% 8 3.1% 7
6 KC 17.6% 9-7 17 22.2% 13.9% 10 5.5% 8 -16.6% 6 -4.5% 26
7 BUF 14.4% 11-5 8 17.1% 16.3% 8 7.3% 7 -9.6% 14 -2.5% 24
8 NYJ 12.3% 8-8 3 9.1% 20.8% 7 -2.2% 12 -12.7% 11 1.8% 11
9 MIA 9.8% 9-7 7 6.3% 4.7% 16 -10.2% 21 -16.0% 7 4.0% 3
10 MIN 9.7% 10-6 4 10.0% 22.5% 4 10.2% 4 3.3% 25 2.8% 8
11 SEA 9.6% 9-7 13 5.8% 10.6% 12 -5.4% 15 -13.7% 9 1.3% 13
12 WAS 7.5% 10-6 26 17.8% -4.3% 20 11.5% 3 -0.2% 22 -4.2% 25
13 PIT 6.1% 6-10 19 8.3% 1.2% 17 -9.1% 19 -11.7% 12 3.5% 6
14 DEN 5.7% 6-10 2 -2.9% 15.6% 9 -3.1% 13 -9.3% 15 -0.5% 18
15 TB 5.1% 11-5 11 -1.3% 12.8% 11 -21.4% 26 -22.8% 3 3.6% 5
16 DET 3.6% 8-8 15 10.0% 6.0% 15 -3.3% 14 -6.6% 19 0.3% 16
17 GB 3.0% 8-8 9 3.5% -7.3% 22 -7.4% 16 -9.3% 16 1.2% 14
18 BAL 2.9% 8-8 20 0.2% 8.0% 13 -25.2% 27 -28.0% 1 0.2% 17
19 CAR 2.4% 8-8 25 2.1% 7.0% 14 4.8% 10 10.7% 27 8.4% 1
20 NE -4.2% 8-8 12 -0.4% -15.0% 24 -15.4% 24 -9.0% 17 2.2% 10
21 IND -5.3% 13-3 28 -0.4% -1.3% 18 9.0% 5 13.7% 29 -0.5% 19
22 NYG -7.3% 7-9 16 -10.5% -7.0% 21 -9.7% 20 -4.4% 20 -2.0% 23
23 PHI -10.1% 5-11 29 -18.2% -2.5% 19 -31.9% 30 -22.9% 2 -1.0% 21
24 ATL -12.9% 5-11 1 -20.5% -12.9% 23 -12.6% 23 -0.6% 21 -0.8% 20
25 SD -13.8% 8-8 27 -14.6% -22.1% 27 -28.2% 29 -12.8% 10 1.6% 12
26 CHI -14.0% 6-10 18 -19.5% -16.4% 25 -8.6% 17 4.0% 26 -1.3% 22
27 CIN -28.9% 4-12 30 -25.6% -19.7% 26 -11.1% 22 10.8% 28 -7.0% 29
28 SF -30.7% 4-12 5 -31.4% -38.1% 29 -8.8% 18 14.5% 30 -7.5% 31
29 NO -35.0% 3-13 22 -30.0% -40.0% 31 -27.5% 28 2.3% 24 -5.2% 27
30 ARI -35.6% 6-10 21 -40.1% -30.4% 28 -35.9% 31 0.7% 23 1.0% 15
31 CLE -46.6% 2-14 x -49.5% -38.4% 30 -20.9% 25 18.4% 31 -7.3% 30

First, let's get this out of the way: Because I wanted to get this commentary written this weekend, the tables are not yet on the website with the individual player stats for either 1998 or 1999. Those will go up when I do a special 1998-1999 mailbag answering questions from the 1998 discussion thread as well as this discussion thread. (I have 1997 data now as well, but likely won't be able to break it down and create DVOA until next summer.)

As you might expect, the Rams had the league's top offense with the debut of the Greatest Show on Turf. What's interesting, however, is how unimpressive that top offense was compared to the top offenses of the last few years. Check out last year's offensive ratings and you'll see that five teams were higher than the 1999 Rams. It turns out that the general offensive level of the NFL jumped just as much in 2000 as it did in 2004. The DVOA ratings currently use a baseline of 2002-2003, with those two years adding to 0%. The total league DVOA for 1999 was -6.4%, by far the lowest of any of the seven seasons I've analyzed so far.

As good as the Rams' offense was, it looked even better because of strength of schedule. You'll notice that the Rams' non-adjusted rating (VOA) is even better than their adjusted rating (DVOA), and that's almost entirely because of the poor defenses faced by Kurt Warner and crew.  Strength of schedule drops the Rams offense from 25.3% VOA to 18.3% DVOA. The difference between the Rams' unadjusted defense (-22.0% VOA) and adjusted defense (-21.3% VOA) is much smaller. Put it all together, though, and the average DVOA of a St. Louis opponent was by far the worst in the league:

1999 Easiest Schedules
(Average Opponent DVOA)

STL -13.6%
JAC -8.4%
DAL -6.2%
CAR -5.7%
WAS -4.0%

Looking at the rest of the NFL, for the most part the 1999 ratings match the 1999 records, and most of the teams with records out of line with these ratings moved in line with DVOA the next year. Oakland was 8-8 despite finishing third in DVOA, then improved to 12-4 the next year. Defending champion Denver went 6-10 mostly due to luck, and bounced back to 11-5 the next year. Like Denver, Pittsburgh finished 6-10 despite a positive DVOA; the Steelers were 9-7 the next year and 13-3 the year after.

Interestingly enough, the two teams whose improvement in 2000 was not predicted by 1999 DVOA ratings were the two teams that made the Super Bowl: Baltimore and the New York Giants. And there was one team that would have been expected to improve based on DVOA but went completely in the other direction. That team was the Dallas Cowboys.

Notice any similarity between the top teams in 1999 and the top teams in 2004? Like last year, the AFC dominated in 1999 with seven of the top nine teams in DVOA, while the NFC sent two 8-8 teams into the playoffs. What's strange is that one of those two NFC teams in the top nine was also one of those two NFC teams to make the playoffs at 8-8: Dallas. The conventional wisdom about the 1998 Cowboys held that Chan Gailey was losing control of the team and the triplets had gotten too old to win (Michael Irvin was lost for the season with an injury in Week 5, as described by Mike Tanier in Too Deep Zone from last week.) But Dallas just had crummy luck. Six of those eight losses came by a touchdown or less, and other two were by just ten points, both on the road against the playoff-bound Colts and Vikings. Dallas's poor win-loss record is even stranger because, as you saw a few paragraphs ago, the Cowboys had one of the league's easiest schedules.

Usually a team that finishes with such a high DVOA but a mediocre record becomes a great sleeper Super Bowl pick the following year. But between 1999 and 2000, the Cowboys made a lot of changes. Irvin retired, Deion Sanders left, Gailey was replaced by the immortal Dave Campo, and Aikman and Emmitt kept aging.

At the bottom of the ratings, San Diego had the lowest DVOA among teams with 8-8 records, and were lower than five different teams with losing records. The next year they collapsed to 1-15. The second-lowest 8-8 team was New England in the final year of the Pete Carroll era; they dropped to 5-11 the next year. The 1999 Arizona Cardinals were not quite as fluky as the 1998 Arizona Cardinals, but they still somehow won six games with the worst non-expansion DVOA in the league. They went 3-13 the next year.

One team that really stands out when you look at that table is Indianapolis, playing the same role as the 2003 Panthers and 2004 Falcons. They were the team that just kept winning, poor DVOA rating be damned. It doesn't make sense for a team to go 13-3 despite ranking 21st in DVOA.  The Colts did outperform their Pythagorean projection of 10.2 wins, but that really wasn't a big deal, because they were still the fifth-best team in the league according to Pythagorean wins. The Colts were similarly underrated by DVOA in 2002, ranking 20th despite a 10-6 record, so I wonder whether the issue here is something in the Colts' style of play that DVOA is not picking up correctly. One thing those two teams had in common was a terrible turnover differential, so perhaps the turnover penalty in DVOA needs to be reduced. Or some of it could be luck: the 1999 Indianapolis defense recovered 12 of 17 fumbles, for example.

A lot of the disconnect between the Colts' record and the Colts' DVOA rating comes from the running game. Indianapolis ranked third in passing offense but 16th in rushing offense -- even though Edgerrin James, in his rookie season, led the league with 1552 rushing yards. Yet James was nowhere near the league leaders in the Football Outsiders metrics -- he was 12th in total value (14.7 DPAR) and his play-by-play performance was actually below average (-4.6% DVOA). (DPAR explained here.)

Edge's main weakness in 1999 was actually the same as his main weakness in 2003 and 2004: he was horrible near the goal line. Edge's DVOA in goal-to-go situations was -51.1%; his DVOA in all other situations was 2.6%. He was below average from the one- or two-yard line, and horrific when given the ball farther out. The average running back scored on 26% of goal-to-go carries from the three-yard line or beyond. Edge had 14 such carries and scored on none of them. In proper Edge fashion, he also fumbled three times in goal-to-go situations.

Edge's other weakness, however, showed what a different player he was before he tore his ACL. Last year, Edge had a 57% success rate. In his rookie year, Edge had a 43% success rate, and was much more of a boom-and-bust runner. (Success rate explaned here.) He was particularly poor on first downs. Edge had 84 carries on first down that went for a yard or less. No other running back in 1999 had more than 62 such carries. That meant Edge was constantly leaving Peyton Manning in second-and-long situations, and the Colts were fortunate enough to have a top quarterback who could advance the ball anyway.

The other playoff team with a record far better than its DVOA would otherwise indicate was Tampa Bay. But much of Tampa's low rating -- and much of the reason why Oakland is ranked fourth despite finishing 8-8 -- is one game in Week 15 that ended Oakland 45, Tampa Bay 0.

Here's a look at the best and worst individual players of 1999. Remember, DVOA measures value per play, while DPAR measures total value:

  • The top quarterback was, of course, Kurt Warner, but the gap between Warner (102.1 DPAR) and a big second-year leap forward by Peyton Manning (97.0 DPAR) was very small.  If you don't adjust for strength of schedule, the gap goes from 5 points to 32 points.
  • Guess which of the quarterbacks from the Class of 1999 had the best rookie season? Daunte Culpepper didn't throw a pass all season, so he's out. Would you believe the answer is Cade McNown?

Player Team DPAR Rank
Cade McNown CHI -2.8 34
Akili Smith CIN -26.7 47
Tim Couch CLE -41.5 48
Donovan McNabb PHI -41.6 49

  • At the start of the 1999 season, Jake Plummer was being celebrated as one of the NFL's best young quarterbacks, the man who would make the Cardinals respectable again. By the end of the 1999 season, Plummer ranked as the league's worst quarterback with -47.4 DPAR.
  • Stephen Davis was just a monster in 1999 with 1407 yards (second in the league) and 17 touchdowns (nobody else scored more than 13). He led the league with a 60% success rate, and nobody else with at least 75 carries was over 55%. He led the league with 49.1 rushing DPAR and also had the highest DVOA of all backs with at least 50 carries.
  • The other running back who had a huge year, of course, was Marshall Faulk. Since Faulk is now an aged backup who specializes in catching four-yard passes on 3rd-and-6, let's take a moment to celebrate just how awesome he was at his peak. Faulk was second in the league with 37.8 DPAR rushing. How far ahead of the league were Faulk and Davis? The difference between Faulk and the number three running back, Robert Smith (24.2 DPAR) is bigger than the difference between Smith and the 17th-ranked running back, Eddie George (10.7 DPAR). But wait, we haven't even touched Faulk's astonishing year in the passing game. Faulk had 36.9 DPAR receiving, more than twice as high as the second-ranked receiver among running backs, Tiki Barber (16.5 DPAR). Faulk had 1048 receiving yards and caught 84% of the passes thrown to him. No other running back with more than 25 passes had a catch percentage higher than 78%.
  • Although Davis and Faulk dominated the league in rushing DPAR, one other running back finished ahead of Faulk in rushing DVOA, but he only carried the ball 89 times: some no-name scrub from Baltimore named Priest Holmes.
  • The least valuable running back of 1999 was ... wait for it ... wait for it ... Ricky Williams (-18.2 DPAR). The second-worst running back in rushing DPAR was Warrick Dunn, but he was also the third-best running back in receiving DPAR.
  • The top wide receivers in DPAR were Jimmy Smith, Cris Carter, Tim Brown, Isaac Bruce, and Randy Moss. Slightly lower, in the "whatever happened to" file, are Patrick Jeffers of Carolina (eighth) and Michael Westbrook of Washington (ninth). The league's least valuable receiver was J.J. Stokes of San Francisco.
  • Does a strong performance in a limited role mean that a player will be successful as a starter? That's one of the oft-debated questions here at Football Outsiders. Can a player with a high DVOA as a backup running back succeed in a starting role? (We'll ask LaMont Jordan in a year, I guess.) Does a high DVOA for a third receiver mean that player can become a number one receiver? 1999 tosses a few more players into this discussion. We rank receivers with a minimum of 50 passes, and the top three receivers in DVOA were all guys with between 50 and 60 passes. Each of those guys eventually became a starting receiver for another team, with differing levels of success. Number one was Atlanta's Tim Dwight. Number two was St. Louis's Az Hakim. Number three was a 27-year-old career scrub from the Chiefs named Joe Horn.
  • Two tight ends were a step ahead of the pack: one player having his first good year (Tony Gonzalez) and one player having his last good year (Rickey Dudley). Although Gonzalez had a slight advantage in DPAR, Dudley had a significantly higher DVOA -- but he only was thrown 59 passes, compared to 108 for Gonzalez. The least valuable tight end was Jed Weaver of Philadelphia, probably because he was the safety valve for a struggling rookie. Weaver caught only 39% of intended passes; the league average for tight ends was 60%.
  • One final note. There was a punt in 1999 that was just ridiculous, far past any baseline I had ever created. In Week 5, on the road against the Jets, Bryan Barker punted the ball from the Jaguars' own 5-yard line to the Jets' 12-yard line, an 83-yard punt. What the hell? That's unreal.


23 comments, Last at 26 Sep 2005, 9:56pm

2 Re: 1999 DVOA Ratings and Commentary

Hey Aaron, do you have the DVOA of the 62-7 jags-dolphins playoff game from that year? Just curious to see.

4 Re: 1999 DVOA Ratings and Commentary

If a punt bounces away from the returner, who knows where it will go?

On that note, CFL punter Jon Ryan, who is on his way to breaking the CFL's all time punting record for punting average in a season will probably be playing in the NFL next year. Ladies and gentlemen, keep your eyes peeled. This year, he has been booming for over 51 yards per kick. (See link)

6 Re: 1999 DVOA Ratings and Commentary

I don't get what is so amazing about that. If the punter kicks is 42 yards in the air, and it rolls for another 41 yards because the returner fell down or something, that goes in the books as an 83 yard punt.

7 Re: 1999 DVOA Ratings and Commentary

Woohoo! Go Browns! Highlights from 99, in order:

1) Winning the Hall of Fame game. Maybe winning it wasn't so important (although it was against the Cowboys), what really mattered was that they were back! This is, and likely will remain, the only preseason game I have ever held a party for.

2) Beating almost-as-pathetic New Orleans on a hail mary (Couch to Johnson) for the first win. They were still in the game thanks largely to a fumble by worth-ten-picks Ricky Williams while driving for a clinching score (or running out the clock). Thanks, Ricky.

3) Winning at Pittsburgh, on a last-second field goal. The attempt was set up by a boneheaded penalty against some backup Steelers linebacker from OSU, who will never amount to anything, and certainly won't catch TD passes in consecutive Super Bowls.

That's it! What a dreadful year. Every advantage given to them in the draft, decent expansion players available, a clean salary cap slate, and that's what they ended up with? Matt Millen could hardly have done worse.

On a more fun note, this was one of the years with an odd number of teams, meaning that someone had to have a bye every week. This is reason #1 that the league should stick with 32 teams and a wonderfully balanced, regular schedule. How dumb was it for teams to have byes in weeks 1 or 17? I think this year's byes began a week or two early, but week 1?

8 Re: 1999 DVOA Ratings and Commentary

83 yard punt?

Randall Cunningham once kicked a sneak 91 yard punt against the Giants I believe.

He always claimed he could have lead the league in punting if Buddy Ryan would have let him.

I find it interesting that Baltimore, Philadelphia, and Tampa Bay all had the best defenses in the league in 1999. That truly foreshadowed things to come given their 26th, 27th, and 30th ranking in offense.

9 Re: 1999 DVOA Ratings and Commentary

Would it be possible to restrict the contribution of any single game to the overall DVOA score? This would help with situations like the 45-0 shellacking of Tampa Bay as well as the games without relevance at the end of the season (without resorting to an arbitrary reduction).

One could actually take it one step further: reduce the possible contribution of any single PART of a game, in an effort to weed out garbage time scoring.

I'm not sure how you weigh the contributions of individual plays, but it might actually also make sense to use a cap there as well - similar to what you already do for line yards.

10 Re: 1999 DVOA Ratings and Commentary

Say, it's great that we have DVOA numbers for all of 1999, but where are the numbers for Week 2 of 2005? I want to get my money's worth, here. . .

11 Re: 1999 DVOA Ratings and Commentary

I'm not surprised by the wild difference in James' success rate at all. I had him on my fantasy team (I had the most loaded fantasy team in the universe in 1999- Warner, Bruce, Faulk, James, Jimmy Smith, Randy Moss...what?) and so I paid close attention to Colts games. All I remember is that he would run into the line for no gain over and over, until finally he ran into the line and kind of bounced off someone into the clear and then proceed to run for 60 yards. It was about the most unimpressive great season imaginable.

13 Re: 1999 DVOA Ratings and Commentary

Great stuff. I am really surprised at the high performance of the Jets. They finished 8th despite having some very bad losses early in the year with Rick Mirer at quarterback. In a few of those games, the offense was just abysmal. Ironically, the game that cost the Jets the most may have been at #4 Oakland (a game in which the Jets blew a sizeable 2nd half league).

However as a Jet fan, this chart just adds to my pain. Not only would Vinny have made a difference, but if they made the play-offs the Jets might have been able to get to at least Round 2. Painful memories indeed.

14 Re: 1999 DVOA Ratings and Commentary

2005 stats will be available starting Tuesday, and on Monday you'll find out why I couldn't do them for two weeks. I have the 2005 DVOA ratings now through Week 2 (Pittsburgh is on top, duh) but rather than write a commentary that would just be rendered moot by this weekend's games, I decided to hold off until after the weekend. The 1999 stuff was done months ago, I just never finished writing the article until yesterday.

15 Re: 1999 DVOA Ratings and Commentary

I'd like to see the Jets breakdown with Mirer as QB and Lucas as QB, I imagine it's fairly large

16 Re: 1999 DVOA Ratings and Commentary

#12: VarlosZ,
What money?

Damnit, I forgot to turn on the laugh track.

Thanks for all the hard work, Aaron -- I think these things are fascinating. Out of curiousity, how far back are play-by-play summaries available for all teams? In other words, how many of these articles could there eventually be?

17 Re: 1999 DVOA Ratings and Commentary

I am a Raider fan and remember this season quite well. A few comments. First, Rickey Dudley??? This guy was always thought of as a colossal dissapointment but he was apparently pretty good for a while.
Second, do you have rankings for variance? I reckon the Raiders were the least consistent team in the NFL that year. You mention the Tampa Bay game, but they also beat the Bills, Vikings and Chiefs (once) but lost to the Chargers, the Broncos twice and the Packers.
Third, I am surprised the special teams weren't even worse. They lost about three or four games due to special teams mistakes (these losses caused them to draft Janikowski in the first round next year. Ugh.)
Finally a lot of people wanted Jon Gruden fired this year because the team was so inconsistent and underacheiving. In particular people were angry about losing to the Broncos twice but they seem to have been better than their 6-10 record. Many think he saved his job by beating Kansas 41-38 in week 17 (my favourite game ever), ending a string of about 193 consecutive losses to the Chiefs. I wonder how things would have turned out if he had been fired. Is there someting about Gruden that causes his teams to badly underachieve their DVOA?

18 Re: 1999 DVOA Ratings and Commentary

Aaron, I'm pretty sure the record for the longest punt in NFL history is 99 yards, and I'm also pretty sure it happened against Denver. Can't remember who did it, though.

Take a second to think about that. You're pinned on your own 1 yard line. Your offense takes three shots to move the ball. No luck. You're punting from your own end zone. You boom a punt, the other team lets hit fall. It takes a favorable bounce, rolls, rolls, rolls... stops dead on the other 1 yard line. You just completely reversed the field. It's the one NFL record that I'm pretty sure will never be broken (unless we start adopting CFL rules).

19 Re: 1999 DVOA Ratings and Commentary


Wouldn't what you described be a 98 yard punt? (100 minus 1 yd at each end)

A 99 yard punt would have to be from your own one to the other team's endzone, for a touchback, right? Or am I missing something here?

20 Re: 1999 DVOA Ratings and Commentary

Yes, that would be a 98 yard punt, the current NFL record, by Steve O'neal against Denver. He kicked it from the back of the endzone, it first bounced at the opponents 30, then came to rest at the 1 yard line. That's a freak play.

21 Re: 1999 DVOA Ratings and Commentary

As a Titans fans, I'm extremely surprised to see an offensive ranking of #6 and a defensive ranking of #18. I mean, I remember who the wideouts were that year, and Kevin Dyson, the post-Pittsburgh Yancey Thigpen, Isaac Byrd, and Chris Sanders were not exactly names that struck fear into the hearts of opposing defensive backs. Then again, when you have a team that can go 14-2 without beating anybody with a winning record, it's a weird year.

22 Re: 1999 DVOA Ratings and Commentary

Looking at the Eagles' weighted DVOA showing a good bit of improvement, I can't wait to see the numbers for the Doug Pederson experience. Oooh boy. Looking at the three guys who did start right away, I can't argue with Reid waiting to put McNabb in, but still...