SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.
Also, our existing PFR blog rss feed will be redirected to the new site's feed.
Pro-Football-Reference.com ยป Sports Reference
For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.
ShareThisTony Romo’s career stats aren’t as misleading as I thought
If you go to PFR's leaderboard page, you can see that Tony Romo ranks 4th all time in yards per pass attempt, and all three quarterbacks ahead of him retired by 1960. Romo also ranks first in net yards per attempt (PFR has NY/A data beginning in 1969, which is the earliest year for which we have complete individual sack data) and first in adjusted net yards per attempt among all players. You can use PFR-'s play index to see a list of the top active quarterbacks in yards per attempt, and I've heard multiple sources cite Romo's incredible career average in that category.
My initial take when I heard such propaganda was that is was incredibly misleading. Why?
1) Romo obviously plays in a much more pass-friendly era than nearly every quarterback in league history.
2) Romo, because he was an undrafted free agent, didn't see the field during his early years. Young quarterbacks tend to play poorly, and sitting on the bench during your first couple of years will help boost your career averages, especially when you have only played for a few seasons. Which brings me to ...
3) Romo just turned 30 years old in April; he hasn't yet played during the final years of his career, when most quarterbacks tend to kill their career averages. For example, the Patriots Steve Grogan retired with a career yards per attempt average of 7.48, but from age 25 to 34, Grogan averaged 7.98 yards per attempt. Romo, unless he retires early, will sink his career averages at least a little bit by the end of his career.
Sure, the player season finder tells us that Romo's 8.10 yards per attempt average is the fourth best ever, but we know that is really misleading. What we need to do is adjust for those three factors isolated above. And, in fact, that's really easy to do. We can use the player season finder to only look at quarterback seasons at ages 26, 27, 28 and 29 -- Romo has never thrown a pass at any other age.
Figuring out a cut-off is a matter of preference, but the results don't change too significantly here. If we require only 600 pass attempts over those four seasons, Romo still ranks 7th all-time; raising the cut-off to 900 attempts (Romo has 1,857) brings Romo back into 4th place, albeit in a tie with Bart Starr.
Okay, but what about adjusting for era? Last July, Doug introduced the Advanced Passing tables to PFR. He wrote:
In each of nine different passing rate stats, we have computed each player's standard deviations above (or below) league average for that season and converted it to an "IQ score" (average = 100, standard deviation = 15). From the glossary:
- On all stats, 100 is league average.
- On all stats (including sack percentage and interception percentage), a higher number means better than average
- The greatest passing seasons of all time are in the 140s. A typical league-leading season in most categories will be in the high 120s or the low-to-mid 130s.
The advanced passing stats (represented by acronyms like Y/A+) are loaded into the player season finder, which makes this a simple exercise. And, according to that tool, where does Romo rank? With a cut-off of 600 attempts, Romo is in a four-way tie for 5th place in league history. If we use 900 attempts as the cut-off, Romo is in a three-way tie for 2nd place, all-time in Y/A+ by quarterbacks between the ages of 26 and 29. Romo does drop into a tie with Joe Namath for 8th place since 1969 by a quarterback between those ages in ANY/A+ (which is why Jason thinks that Namath is a legit, HOF quarterback).
One reason Romo still looks good when we adjust for era? It may be a passer friendly era in terms of pass attempts, sack rates, and interception rates, but yards per attempt numbers aren't nearly as inflated as you might think. This page breaks down all passing numbers per team game for each season in NFL history. Let's break it down by decade:
Y/A Att INT% TD% Sk% ANY/A 2000s 6.83 32.7 3.16 4.05 6.47 5.40 1990s 6.79 32.4 3.36 3.93 6.85 5.21 1980s 7.04 31.7 4.18 4.2 7.54 4.98 1970s 6.67 26.2 5.26 4.18 8.20 4.04 1960s 7.26 27.7 5.59 5.26 8.68 4.60 1950s 6.95 26.7 7.01 5.04 8.05 3.78
There's no doubt that Romo and other modern passers enjoy several advantages when it comes to putting up good numbers relative to passers from the '60s and '70s. First, they play in two more games a season. Then, they pass the ball significantly more frequently in each game. And they get sacked less often and throw fewer interceptions -- both due to rule changes in the sport, not because today's quarterbacks are better at what they do. As a result, modern quarterbacks tend to look excellent in terms of gross passing stats and ANY/A relative to their forefathers. Completion percentages, not shown above, are subject to perhaps the largest era bias of all the major stats.
On the other hand, vanilla average yards per pass attempt hasn't changed all that much. For the most part, the league average has always been around 7.0 yards per attempt. And that's why Romo's numbers don't necessarily need to be adjusted for era, at least when you're talking about yards per pass. Adjusting for era and age, Romo's Y/A+ shows that he was only bested by one man -- Otto Graham, who spent all four seasons playing in the AAFC, and not the NFL, during those ages.
If you want to move past the Y/A stats, Romo still is in elite company when it comes to ANY/A. For the 41 years where we have ANY/A data, look at which quarterbacks rank in the top ten during ages 26 to 29: Five Hall of Fame quarterbacks (Montana, Fouts, Aikman, Namath and Marino), two future HOF QBs (Manning and Favre), another guy who could be on the HOF path (Rivers) and Mark Rypien. For quarterbacks, winning the Super Bowl has become inextricably tied with being considered a successful quarterback. But Romo has been about as good as you could ask for if you're a Cowboys fan, and he may be in for his biggest season yet in 2010. In addition to having what looks to be an incredibly stacked offense, Romo's Cowboys would be set up nicely to win the Super Bowl this year, played at Jerry World.
This entry was posted on Monday, August 23rd, 2010 at 6:49 am and is filed under History, Play Index. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
(I say these things as a Saints fan)
I think Tony Romo gets too much credit (from the general media) because:
1. He plays in Dallas--one of the main FOX broadcasters is a former 'Boys QB;
2. His style of play resembles #4 who just unretired again;
3. He has good stats--both basic and advanced.
I think Romo DOESN'T get enough credit (from certain fans/bloggers etc.) because:
1. He plays for the Cowboys--I don't personally know ANYONE who says that their 2nd favorite team is the Cowboys. (For example, mine is the Colts because of Manning's New Orleans' ties.) So, most people are going to dislike/hate him because he dosen't play for their favorite team, not to mention that most folks will dislike the way that the MSM fawns all over him.
2. He only NOW has a playoff victory to his credit. Fairly or not, most folks tend to give the QB too much credit for a win, and too much blame for a loss. Well, considering the Cowboys' collapse in December/January since Romo became the QB, it's easy to see why fans would rate him lower than where he deserves.
3. He has improved. Per FO stats from their 2010 Almanac, he had 19 INT's in '07, then 14 in '08, and just 9 last year. He also cut down on his fumbles, as he had 9, 10, and then only 5 last year. Unfortunately, his sack rate has gone up--but IMO, that means he's just eating the ball instead of trying to make something happen and turning the ball over. That's an improvement--because Dallas has a good enough offense to overcome a sack once in a while.
To sum it up, if he wants to elevate his profile in the eyes of the general public, he needs to have these great stats in January too.
Perceptions of QBs can be downright strange even among supposedly unbiased experts. They often say it's all about winning (especially in the playoffs), but the exact same people give Aaron Rodgers a free pass for having Romo-esque regular season stats but with a worse winning % and 0 playoff wins. Heck, until he imploded in Chicago many people talked up Jay Cutler as being CLEARLY better than Romo...in spite of having worst stats in every area AND a worse winning % AND never even making the playoffs, never mind winning a game. People are only just now seeming to notice that Carson Palmer hasn't been an elite QB since 2006. Tom Brady was called a mediocre system quarterback when he was winning Super Bowls with B-list receivers (as a Pats fan I always loved that one), and now that he's NOT winning titles anymore he's considered super-elite.
Re: comment #2, Aaron Rodgers put up 45 points in his playoff start and on the play he fumbled in overtime, the officials missed a blatant facemask penalty. Now compare that to the Cowboys-Vikings playoff game last year and let's figure it out.
People began thinking Brady was elite because he had 50 TD passes in one season and the team won 18 games in a row, so that's a fair assumption by the general populace.
sean, I don't think that's quite true. On Football Outsiders, they created an Irrational Brady-Manning thread between Brady's 2nd and 3rd Super Bowl win. I think that through 2002 Brady was mostly considered to be a system QB. 2003 he began to shed that tag, and it was probably completely gone by the 3rd Super Bowl. Of course, his reputation attained sky-high levels during the 2007 season.
I know he's still ahead of him for career, but a lot of those Otto Graham rankings would be invalid since AAFC stats do not count for the NFL. In fact that reminds me, a way to search for all leagues but the AAFC would be very nice to have.
I don't like Romo, but I like his style of play and I think he gets less credit than he deserves. He's consistent and he's been improving every year.
Chase, I really have a problem with your statement at the end about Romo being about as good as you could ask for if you are a Cowboys fan. That statement is perfectly accurate if you would have been talking about Tom Brady in his 1st 4 years as a starter, as Brady was OK in the Reg. season-an avg. Passer Rating of 87.6, but then in the PO's he turned it up to 91.5 and led his Team to a lot of success. But Romo was just the opposite-he has a Passer Rating of 95.6 in the Reg. season and then drops off to 81.3 in the PO's and has very little success. I would think Cowboy fans would violently disagree with your assessment. I would bet Romo himself would disagree with you. I doubt if his goal is to simply lead the 'Boys to enough wins to qualify for the PO's. Most, if not all, QB's, owners, coaches, and fans have higher goals than that.
Sorry about that, but I copied the number down wrong. Romo's PO Passer Rating is an even worse 80.8
Every team's goal should be to make the playoffs every year. Because anything can happen once you get there. Doesn't matter anymore if you weren't the best team in any month of the season prior to the playoffs. Just be the best at the end and you can end up with a ring.
If not for his injury in 2008, Romo would be 4/4 in making the playoffs for Dallas. Does he need to play better when he gets there? Absolutelly. But if he keeps getting Dallas in that position, then he's going to increase his chances of being able to succeed.
Yes, I'm sure that making the PO's is every team's "secondary" goal. Must you qualify for the PO's in order to go on and win a Champ.-of course you must-it goes without saying to any knowledgeable football fan. But the bigger goal is to have success in the PO's, and Romo has yet to do that. I would bet Cowboy fans would tell you that Tony has to play better in the PO's in the future.
He does have to play better. I have his playoff drive stats, and 1.49 points/drive is pretty poor stuff.
But you know who else has to play better in the playoffs? Philip Rivers. He's thrown at least one pick in every playoff game, and has a losing record with stats that look way different than his superb regular season totals. Yet you don't hear about this nearly as much as you do with Romo, because of the crazy Dallas media frenzy.
It was already mentioned above how Aaron Rodgers is getting the media bandwagon all geared up, but he has issues too with a barely .500 record and playoff blunder moment.
I find the media coverage of these certain R-named QBs very strange.
Good points. I guess the ultimate example would be Favre. I guess we both give him a lot of credit for leading his team to the PO's more times than any other QB (12 times), but what has he really accomplished? He led his team to only 2 Conf. Champ. in 5 tries, and is only 1 and 1 in his Super Bowl games. My main point though is this-a QB only has to play good enough to lead his team to 10 or so wins in the Reg. season (saying it another way-he can have 4, 5, or 6 bad games) and his team can still qualify for the PO's and go on to win a S.B. (Giants). But, in the PO's you don't have that luxury. One bad game by your QB and your team could be done for the year.
Dear Chase,
I nearly fell off my seat when I heard you using the words "propaganda" which is "incredibly misleading" in reference to the various stat headings related to the real-life Yards Per Attempt ratio. One of these variations is Ajusted Yards Per Attempt which you used as your primary building block to create your "GOAT" series!
Although I realize that I'm taking your words somewhat out of context, the fact is that I couldn't agree more with your casual comment and the sentiments it portrays. I've been saying the same thing for several years, but I've been even harsher in my assessment of these contrived stats. I refer to these type stats as "Bizarro World" stats.
Why am I so negative? Because, once you get beyond Yards Per Attempt, you enter into the rhelm of team efficiency stats. Adjusted Yards Per Attempt has very little to do with the merits of the QB and very much to do with the merits of the offensive "team" as a whole. Even the innocent looking stat Yards Per Attempt is, in essence, a team stat---as are all football stats for that matter. There are so many circumstances completely outside the control of the QB that it makes no sense to reward or punish him alone. Yet, that is exactly what happens whenever we make references to this type of "propaganda".
Every once in awhile, one of the bloggers on this PFR web site, makes some brilliant and astute comment on some subject---comments that I like to jot down. One such comment came from "MattieShoes" in the 10-5-09 blog of Jason Lisk concerning QBs who change teams and how their stats are affected: He wrote, "What I draw from this post is that a QB's stats are largely controlled by outside influences. So, drawing conclusions on the quality of a QB based on his stats is dicey at best."
Tony Romo is an excellent QB. Time will tell just how excellent he is.
One bad game by your QB and you can still advance too.
Is there any QB in the last decade that won a SB with an utterly dominant run and needed little help or luck to get by?
2000 - Dilfer was crap in Tennessee, and they won because of a blocked FG returned for a TD and a pass off Eddie George's hands was returned for a TD.
2001 - Brady needed the tuck rule followed by the greatest clutch kick ever to just get past the first game.
2002 - Brad Johnson didn't need anything particularly special, other than that dominant TB defense.
2003 - Brady needed Vinatieri to again hit a tough 45 yard FG on a frozen night to beat the Titans, and Drew Bennett dropped a big pass on 4th down to end the game.
2004 - Can't really say anything about Brady's performance here. McNabb's 3 picks sure helped win a close one though.
2005 - Nick Harper just needs to take that Bettis fumble down the sideline and the Steelers don't get past the 2nd round.
2006 - Vinatieri kicked 5 FGs in Baltimore, a game without any TDs. The defense had to stop Brady on a 3rd & 4 to get Manning the ball back for the winning TD. They had to pick off Brady to end the game.
2007 - Eli's run was very good. A last second TD by Romo would have ended it though. Favre's INT helped in OT to set up the winning FG (Tynes should have made one earlier though). Look at all the close calls on the SB drive. 4th and 1 run. Samuel drops an INT. Tyree catches a pass with his helmet. Steve Smith gets just enough yardage on 3rd & 11.
2008 - Flacco had a chance to get a GW FG and knock the Steelers out. Polamalu picked him off for a TD. James Harrison had a huge INT TD in the SB. Francisco falls covering Holmes on the 40 yard gain to the 6 yard line.
2009 - Favre's insane decision to throw that INT helped a game reach OT that may have never got there without it. Tracy Porter was as much of a playoff MVP as Brees for NO.
Favre has a lot of recent seasons that are hard to defend, but if you go back to 2003. Without that 4th & 26 conversion allowed by his defense, we aren't talking about any Brian Dawkins INT in OT. Favre's playing the Panthers in the NFC-C and he'd definitely have a shot at the SB that year. Then in 1998, he's at least getting past the 49ers if the refs would have called that Jerry Rice fumble. No great TD by Young to Owens to win the game. No loss for Favre.
I've watched a few of these recent SB seasons by teams from start to finish, and it takes a lot of things to fall right to get that far. It's not just about how good you are. Especially not these days.
You guys know I'm a major Pats fan, a big Brady supporter, etc. But here's a strange Brady stat -- YPA leaders, playoffs only, since 2001, minimum 150 total attempts:
http://www.pro-football-reference.com/play-index/tiny/vLG42
In fact, no matter where I set the minimum attempts, he still comes out at or near the bottom in YPA. I know the excuses by heart: wintry Foxborough weather, crappy receivers (pre-07), etc. But it's still surprising. In the absence of net YPA, ordinary YPA is probably the best indicator of ability (INT are random year to year; TD passes are often merely a function of playcalling), and the QB for the winningest playoff team in that span has very mediocre postseason YPA numbers relative to his peers.
As a long-time Brady hater, I am well aware of those YPA numbers. If you look at the 62 QBs with 150+ playoff attempts, he comes in 45th in YPA, right behind McNabb and Kerry Collins.
He also comes in 60th out of 62 in yards/completion (10.4), only above notoriously bad playoff QBs Drew Bledsoe and Steve McNair.
I think he relies on the dink and dunk a lot, and plays conservative while relying on the defense to do the job in many of these games. Then sometimes he's just plain awful like the Baltimore game in January.
Starting with the Carolina Super Bowl and ending with the win over Jacksonville in the 2005 Wild Card round, Brady had a great playoff run of 5 games. But the games before that and after that, he's been average at best. Reminds me of what I said about Montana/Bradshaw/Aikman, but usually your hot streak comes at the end or beginning of your career, not in the middle (of your playoff career).
Neil, the Foxoboro weather argument doesn't work. And those receivers did a pretty good job of not dropping the ball unlike some bigger-named guys on other teams that blow it.
In 2001, Brady only had the one bad weather game (the Tuck Rule game). The weather in Pittsburgh was fine, and the SB was in a dome.
In 2003, it was a very cold night against the Titans, but the field was fine. Steve McNair, playing against the tougher #1 ranked NE defense, had a 8.08 YPA compared to 4.90 for Brady. For the AFC-C, the weather was not good, and having recently watched that game, I couldn't believe how many passes both QBs were throwing near the defense. Brady could have had 4 INTs himself that day. The SB was again played indoors in Houston.
In 2004, they got another bad weather day against the Colts, and both QBs played dink & dunk football. The Steeler game was cold, but no snow or anything. Roethlisberger had a 9.42 YPA in that game. The SB was played in Jacksonville, weather-free.
In 2005, both games were played in clear, cool weather.
In 2006, it was nearly 50 degrees when they played the Jets at home. San Diego had nice (January) weather. Indy was a dome.
In 2007, the weather was unusually nice and calm for both games in NE. The SB was played indoors.
Last year, it was cold, but the field was in great shape for the Baltimore game.
So you're looking at 3 games out of 18 where the weather was a factor, and the common link is snow. You can subtract those games and he'd still be at 6.55 in YPA.
what has [Favre] really accomplished? He led his team to only 2 Conf. Champ. in 5 tries, and is only 1 and 1 in his Super Bowl games.
Two of the 3 conference game losses were in overtime. Three of five conference games were on the road.
Going .500 in big games is perfectly respectable, and the longer a QB's career, the more likely he is to regress to that mean: Joe Montana lost his last two championship games, for instance.
Others have noted that YPA and similar statistics are team passing statistics not QB statistics, and that Romo has a strong supporting cast. This is of course true, but it's fair to say that most Hall of Fame quarterbacks spent their primes playing on pretty good teams, so it may not be hugely germane.
What I would like to focus on is the apparent discrepancy between Romo's performance in YPA-based rate stats and the most sophisticated passing rate metric I am aware of, DVOA.
Here are Romo's DVOA rankings and percentages for his four years as a starter:
2006: 5th, 18.9%
2007: 6th, 25.5%
2008: 10th, 18.5%
2009: 7th, 32.3%
Now, clearly that represents strong performance; I don't think anyone is denying that Romo is a very good quarterback. But that does suggest that he has reached 30 without playing at close to an All Pro level, and ought not to be an automatic Pro Bowl selection (in fact, if he played in the other conference he might never have been to a pro bowl).
Peyton Manning finished ahead of him in DVOA in all four of those seasons, as did Brady in the three for which he was healthy. Ok, those guys are first ballot Hall of Famers in their primes - no shame in finishing behind them. But in 2008, he was also behind a rookie (Matt Ryan), a guy in his first year as a starter (Aaron Rodgers), a couple of ageing good-but-not-great players with worse supporting talent than his (McNabb and Pennington) and a good but not great player with worse supporting talent who is younger than him (Eli Manning). So, maybe it was just a down season. What about his best year, 2009? That 32.3% DVOA at age 29 looks mighty impressive, but it's still not as good as Matt Schaub's the same year at age 28, and Matt Schaub is good, but he's not a superstar and I don't think he's ever going to be (with good reason).
So what's going on here? Is Romo deficient in some way which DVOA captures and YPA etc. miss? Or brilliant in some way which YPA is picking up and DVOA isn't? It's not a strength of schedule thing: an eyeball of the VOA-DVOA differentials for the years in question suggests that he's faced overall a pretty average slate of pass defenses. It could be that he's somehow particularly bad at accruing first downs, as opposed to yards, which DVOA values very highly. I have no way of verifying whether that is in fact the case, but I can't see any terribly intuitive reason why it should be. Is there some kind of era-wierdness that your adjustments are somehow missing (like a new era of even easier passing having started just in the last couple of years)? But if I'm understanding the methodology behind A/Y+ correctly, that shouldn't matter in that case at least. Or is it simply his consistency between ages 26 and 29 that is remarkable, giving him a better performance mean while others have higher peaks and lower troughs? I really don't know, and I'm genuinely intrigued.
http://www.pro-football-reference.com/play-index/tiny/fmeR4
Romo is 1st over the last 4 years in Y/A, and tied for 2nd in AY/A, but only 5th in ANY/A. The problem for him is that it's a golden era of passing with Manning, Brady, Brees and Rivers having just enormous years. There's certainly a big drop-off after those five.
Reply to Postings # 6 & 7 "Bob S."
Re: Brady & Romo
Dear Bob,
You corrected the passer ratings for Romo, but still have Brady's incorrect. Brady's passer ratings are 93.3 regular season and 85.5 post-season. Of the top 20 QBs with highest post-season passer ratings (minimum 150 attempts), six have lower passer ratings in post-season games than in regular season games---Manning, Roethlisberger, Favre, Young, Brady and Gannon. That's pretty good company for Romo to be in---eight Super Bowl wins between them. Romo still needs improvement in post-season, but look at what Brees did last year---with an unbelievable 3-game run, he jumped up to just behind Bart Starr in post-season passer rating (I believe that the NFL requires a minimum of either 120 or 125 passes to qualify which is why Jeff Hostetler doesn't make the list).
Clark, you missed the point a little on Brady. I was talking about Brady's 1st 4 years as a starter as I stated in the post. In those 1st 4 years Brady stepped up his performance level in the PO's and it really paid off for him and his team. As for the others, you're right about Favre but c'mon, his Rating goes from 86.6 all the way down to 86.3--doesn't that qualify as negligible? My point, again, is basically this: a QB usually had to play good to very good to have a lot of success in the PO's and some were able to do that, and some were not.
Reply to Posting 21 "Bob S."
Dear Bob,
The only game Joe Montana ever got pulled out of was in the playoffs against the Vikings in 1987 when he could do no right and the Vikings Wade Wilson could do no wrong. This was the 3rd consecutive dreadful game that Montana had in the playoffs. In these three games, Montana threw no TD passes and 4 INTs! Should we petition the HOF to throw Montana out on his ear?
Fortunately, Montana redeemed himself, but if we just looked at those three games, we'd have to judge him the worst of all-time.
Few QBs play enough post-season games to conclude very much. Even those who have played a lot of post-season games---those games form only a minor portion of their careers, and given the element of luck, most end up with about a 50% winning percentage. Only Starr and Brady stand out above the crowd---and Brady had extraordinary luck. Romo has only 4 playoff games under his belt. Give him another 8-10 playoff games and maybe we can draw a few conclusions.
Bob, you're crazy if you think a QB can just willingly "step up their performance" in the playoffs. Why would a QB on a good team ever NOT play to their maximum ability - regular season or otherwise? It's also safe to say that you face better teams in the playoffs and so you'd expect everyone's stats to decrease compared to the regular season.
Additionally, performance in playoffs is all about luck, particularly record. For one thing playoff sample size is tiny. In Neil's chart above, those guys averaged just 9 playoff games over a 9 year period - which is a LOT more than most QBs. For another, a QB's last game each season is more likely to be his worst game and thus over-represented simply because of the single-elimination format. If a QB was unlucky and happened to have two bad games to start the playoffs consecutive years, that's a huge blow.
Judging a QB by his playoff record is a ridiculous practice. Again, sample size is small so it's hard to make any reasonable comparison. A fluke play, an untimely flag, a bad spot, or just a funny bounce can be the difference between a first round exit and a Superbowl win. Referencing Neil's chart again, those QBs average only 1.3 games over 500 and those are some of the best QBs of all time. Brady alone accounts for half of that edge; without his 14-4 record the others average 0.7 wins over 500. Just 0.7 wins! Why? To start, 50% of playoff teams exit with a losing record and another 25% leave 500, meaning only 25% of teams each year can have a winning record. A QB is essentially punished by making the playoffs more often because he's much more likely than not to worsen than improve his record. Roethlisberger looks like he has a great record now, but wait a few years and see how it drops (if he ever makes it back).
Playoff record is also highly determined by whether or not you win a Superbowl and if you're a 1 or 2 seed. Manning has a 9-7 record in the playoffs, but 4 of his 9 wins come from the 2006 Superbowl. Had he instead lost the Superbowl and been a 2 seed, his overall record would 7-8, a three game swing for only losing one more game! Bye teams are punished because not only do they play one less game (and thus one less opportunity for a win) but they play harder competition AND winning the Superbowl has a huge impact on their record. Bye teams have only 4 options for their record: 0-1 (-1), 1-1 (+0), 2-1 (+1) and 3-0 (+3). See the jump from winning the Superbowl compared to losing it? That difference is enough to change the perspective on your entire career. You telling me you wouldn't think differently of Manning if Hank Baskett covered that onside kick and his playoff record was 10-6?
In essence, playoff records are highly contingent on luck, number of playoff appearances, seeding advantages weighted towards weaker teams, and Superbowl wins. Considering these factors against small career playoff sample sizes, it's virtually impossible to determine the difference between QBs solely based on record. It's a good descriptive stat like time of possession, but it's a terrible prescriptive stat.
James, you make great points. We have had similar discussions here regarding coaching records in the playoffs (specifically Marty Schottenheimer). I always think it's unfair to be too hard on Schottenheimer for his bad playoff record, because he almost always makes the playoffs and many times had a first-round bye. These factors created a difficult challenge to build a great playoff record. If we counted first round byes as a win, and missing the playoffs entirely as a loss, I think Schottenheimer's playoff record compares much better than the 5-13 mark he actually has.
Re: The Manning Index
One of the most interesting formulas Doug ever came up with was "The Manning Index" which ranked post-season QBs by "an approximation of the probability that an average QB would compile the same (or better) record by random chance". Originally published in 2005, Doug updated "The Manning Index" in 2006 and 2007 (1-24-07 blog).
Anyway, from the 1972 through 2006 seasons, only 9 NFL QBs exceeded the number of actual wins over the number of expected wins in post-season play. Doug included Terry Bradshaw and Chase figured up Bart Starr's totals. The parameters of the list are that each QB must have appeared in at least 8 games and thrown at least 10 passes. There is some overlap here---i.e., Montana and Young are both penalized for the Vikings loss while Brady and Bledsoe both get credit for the win over the Steelers in a game that Bledsoe pulled out in relief.
Through 2006: Brady 4.9; Starr 4.1; Bradshaw 3.1; Rypien 1.6; Montana 2.3; Aikman 1.8; Elway 1.8; Favre .8; Brunnell .5; Gannon .1. Rypien ranks ahead of Montana because of some quirk in the formula. As far as Manning is concerned, he comes out at -.3.
Hopefully, Doug can update this for us.
Yes, the Playoffs-what a waste of time. Why do they even have the Playoffs? The really good QB's should definitely boycott the Playoffs. Oh, by the way, I was talking to a lot of sports nuts today, and all of the Marino, Fouts, and Moon fans totally agree with all of you. Now, if I could only determine what all of those QB's had in common.
Re: Addendum to Posting # 25
Sorry, I overlooked Phil Simms who had a differential of .6 on the above list.
Re: The Manning Index---The Flip Side
The beauty and fun of football stats is that you can always find one set of stats to contradict another set. And, wouldn't you know, someone found a set of stats which, in part, contradict The Manning Index. And, to add irony to the situation, the stats come from the same point of origin---Doug!
Bryn Swartz, a sports writer for "BleacherReports", wrote an article on 2-18-10 entitled "How Important is Defensive Support to a QB in Post Season?". A loyal supporter of Doug, Swartz used Doug's formula in his "Wins Above Average" series, which emphacized defensive points allowed, to figure out how many wins above average were attained by each post-season QB through the 2009 season.
These are the results: Brady 5.0; Elway 4.4; Warner 4.1; Aikman 3.3; Roethlisberger 3.2; Montana 3.1; Favre 2.8; Young 2.6; Marino 2.4; Bradshaw 2.2. Other QBs of note: Starr 1.5; Manning .5; Simms -.4; Rypien -.2; Brees 1.6; Brunnell .7; Fouts 1.3; Moon zero; Romo -.6.
On a "extra win per game" basis, Roethlisberger is tops at one extra win per 3.1 games; Warner is second and Brady third.
Swartz draws many other interesting conclusions. One thing he points out---for you Marino bashers---is that Marine had the worst defensive support of any post-season QB---followed by Fouts and Young.
He concludes his article, "So, the next time you pass judgment on a QB's post-season play, make sure that you have looked at more than just his numbers and his won-loss record. Check his defense."
I like what James hinted at with seeding and who your first opponent is. You really can be penalized for making the playoffs practically every year. Drew Brees hardly ever makes the playoffs, has been on several teams that disappointed in not making the playoffs (02 & 05 Chargers, 07 & 08 Saints), but last year's title run helps him with a 4-2 record overall.
Maybe it'd be a worthwhile study to breakdown the first playoff game opponent for, let's say, the 25 QBs with at least 10 playoff starts. I'm sure some guys have had easier slates than others, even as averages. It's easier to win a Wild Card game at home against a 9-7 team than it is to open up against a 13-3 team in the Divisional round that just won a playoff game while you were sitting on your arse the past week.
Re: The Element of Luck
Many of us like using the term "luck" with respect to football outcomes. To a fan, like myself, luck is when the ball bounced off of Frenchy Fuqua and none of the officials saw what happened---thus creating the Immaculate Reception---or when the most gosh-awful call in football history stole a Super Bowl appearance away from the Raiders---the "Tuck" call.
In the math world, luck is defined as "a random process in which the outcome cannot be controlled". In the above two cases, the outcome could be controlled---if the officials made the correct call.
So, how much does luck play in the outcome of football games? Brian Burke addressed this question in a three-part article beginning July 31, 2007. I won't even attempt to go through the math, but his conclusion utterly surprised me. He concluded that around 52.5% of the games played between 2002-06 were decided by luck. Just to make sure, he went through the previous five years worth of games and came up with a figure of 51%.
Yes, I know that this sounds impossible. I can hear the howls of protest. I don't believe it myself. Read the articles and see what you think.
Clark - even if the Tuck Rule went the Raiders' way, they still would have had to beat the Steelers the following week.
Bob, let me ask you this - if a QB's ability controls his playoff success, what magical change did Drew Brees make last summer? I say he's the same guy from '08... and '06... and '04.
He threw for 700 less yards in '09 than '08, had the same number of passing TDs, 34, one fewer turnover (6 fewer picks, but 5 more fumbles lost) and seven more sacks. Yet one team missed the playoffs at 8-8 and the other started 13-0.
Now let's look at '06: 30 more passing yards, same number of interceptions, two fewer sacks, three fewer fumbles lost. Brees and the Saints went 10-6, beating the Eagles in the Wild Card game before losing to the Bears in the Divisional round.
Yes, obviously Brees had a statistically better season in '09: he had 8 more passing TDs in '09 than '06, and his y/p was 8.5 compared to 8.0 in both '08, '06 (and 7.9 in '04). But you know what? To me that's just a career season, one he likely won't reach again. The point is Brees is an elite QB and has been for years - last year was just the one time it paid off. Using the PFR database, he has four of the top 40 passing seasons from 2000-2009 (>125 pass attempts) no matter whether you use Y/P, AY/P, or ANY/P (4 of the top 20 ANY/P!).
What made last year special? Obviously his career year made a difference, but his defense popping off 8 turnovers for TDs certainly helped win a couple extra games compared to its norm. I say the defense, and primarily the lucky turnover scores, was the majority of the difference between previous seasons and earning the bye, easing Brees and the Saints' path to the Superbowl (the Vikings did all the could in the conference game too).
This isn't to say Brees didn't deserve it or isn't a good QB, but that last year's playoff success wasn't determined by anything Brees did last year he wasn't doing before.
Actually Brees did do some things differently last year that led to more success.
- He cut back on sub-par performances. When he threw for 5000 yards in 2008, he had 6 games with a passer rating in the 60's (5 of them on the road), and they lost all 6 of those games. Last season, he only went under 70 one time. That was the Miami game where he did start terrible, turned it over 4 times, but did lead the offense to a good amount of points, 298 yards passing, 1 TD, 2 rushing TDs. He was much more consistent in 2009.
- He also had a career year in leading his team to victory in the 4th quarter. Prior to last season he had very few comebacks and GWD's in his career. Last season he led 6 game-winning drives and 3 comebacks, including the SB.
Bob S, Brady is hardly GOAT status statistically in the playoffs, but he did have the best post-season defense two years in a row in 03 and 04. And in 2001, well you know what happened.
About Brees though, yes he gets a pass for failing to make the playoffs and being inconsistent on an regular basis. He had a career year in 2009, but Philip Rivers (you can make a case for Peyton too) was even better last season if you go by Football outsiders. What it comes down to is you need an entire team to win, and luck. Not having interceptions bounce off feet, Kaeding blowing a game, terrible running game, mega drops in the Superbowl by Garcon/Wayne. It comes down to various factors.
For example the 2007 Colts and 2006 Colts. Peyton was much much better in that loss in Indy than in his Superbowl run, but a Kenton Keith chokage at the 2 yard line caused an interception. A Coin Flip decides the 2008 game, whereas Brady gets away with mediocore production in his 2004 and 2001 run in the playoffs. The AFC is also a little more difficult, as Football outsiders has shown with DVOA and such.
He developed his swagger.
Scott,
In that Miami game, New Orleans held a 40-34 lead at the 2-minute warning. Miami was driving for the possible game-winning touchdown. Tracy Porter picked off a Chad Henne pass and returned it for TD, giving New Orleans a 46-34 win. Drew Brees had nothing to do with that. If Porter doesn't get the pick, maybe Miami scores and wins that game. Instead of going to 6-0, the Saints are 5-1. Maybe this changes their confidence, and they don't roll through the rest of the regular season. Or maybe nothing else changes, so the Saints finish 12-4 instead of 13-3. I'm not sure how the tiebreakers worked, but that gives the Saints the same record as the Vikings. Maybe that puts the NFC Championship game in Minnesota.
The point is, There is a heck of a lot that happens on the football field that the QB cannot control. All he can do is play well and put his team in a good position to succeed, but he needs help. Drew Brees didn't really change anything in 2009 to switch from "guy who can't win anything" into Super Bowl champ. The only thing he changed, was winning. Same thing for Manning in 2006. Same thing for Elway in 1997.
#35 - Put your best Gus Johnson voice on: "He developed his Swagger, presented by Old Spice body wash, the greatest SMELL ... in the N.F.L. Wham, bam, thank you for smelling great, man!"
This is an excellent thread and it simply comtinues to show the prevailing train of thought in the media and among some fans that those with the most titles in a team sport are just "better" than those who don't have as many. While it is certainly true that your performance helps your teams win, it's ridiculous to treat one's individual performance as THE reason your team wins, or even the most important reason. An assembly of great talent is always better, and, as someone mentioned above, alot of it is luck - which old school analysts and fans DESPISE using as a reason to explain success.
Anon, EVERY play has some luck involved-if it's your team you have to hope that the luck goes your way, and vice versa. It's only a story if you can come up with a certain QB or a certain team that had more than their share of either good or bad luck. Do you know of any that would fit the category? I can't name any off the top-except maybe from this angle---in the Raider hey-days (Stabler and Casper) it seemed as though they were getting a lot of breaks from the officials and the officials calls and non-calls. That advantage later seemed to pass on to the Dolphins, and ironically enough, I think Don Shula just happened to be the Head of the Rules Committee. I think the Dolphins were top 3 in Least Penalties Called Against for quite a few years. And then add on to that "Penalties that were not called Against "and that's a nice advantage to have. I'm not sure how long they enjoyed that advantage though, or exactly what years were involved.
After re-reading my last post, the end needs some clarification. I meant to say that that the Dolphins were consistently 1 of the least penalized teams under Coach Shula, and then add on to that the times that announcers would say-"I think the Dolphins really got a break there"-as they really didn't think the penalty should have been called on the Dolphins' opponent. A team can benefit in 2 ways when it comes to the officials calls and non-calls.
***Swartz draws many other interesting conclusions. One thing he points out---for you Marino bashers---is that Marine had the worst defensive support of any post-season QB---followed by Fouts and Young.
He concludes his article, "So, the next time you pass judgment on a QB's post-season play, make sure that you have looked at more than just his numbers and his won-loss record. Check his defense."***
I recall vividly how it would be discussed by announcers etc relating to Miami and Marino, just how many resources went into the passing game - of the operating budget how much was allocated to Marino, his fortress of an offensive line, and the receivers. The whole team was skewed toward the passing game and its success. The running game and defense got low priority in time and dollars. So one has to go beyond just having a good offense and blame the defense when the defense was short changed to subsidize the offense. Marino's stats and lack of upper tier success should be viewed through this prism, especially when looking at stats. Stats also have a background to them tracing back to upper level scheming and managing decisions.
All that all this means is
Romo is a good QB with amazing stats and even though people think he's overrated hes actually way, way underrated.
As soon as Romo wins a SB, or plays 5-6 more years with these same stats and possibly a SB appearence at the very least, he'll all a sudden be in that elite group of QBs, either way, he'll be there