Writers of Pro Football Prospectus 2008

Most Recent FO Features


» The Deep Ball Project

Guest columnist John Kinsley breaks down the tape of every deep pass in the NFL in 2017 and comes away with a shocking conclusion: even without Andrew Luck, the Colts had the best long-ball quarterback in the league.

04 Dec 2005

FO Mailbag

by Aaron Schatz (and the FO staff)

Time for another look at the Football Outsiders mailbag. We get a lot of e-mail, and there are a lot of comments on the discussion threads, so I apologize if your question doesn't get answered. There simply are too many good questions that require well thought out answers. The best way to get your question answered at this point is to use the contact form. If it is a question not related to the DVOA stats, it is more likely to be answered if you send it to one of the other writers, not me.

Be aware that we reference plenty of our innovative FO stats here, not to mention their unfamiliar terminology, so if you are a recent addition to the readership you might want to read this first.

Before we get to some questions, we've had a few e-mails asking if we have an RSS feed. We do, and you'll find it here:


Let's start with a question for Mike Tanier.

Tony Doucet: Mike ... please, please, please ... help settle an argument. Is Drew Bledsoe a Hall of Famer? This is a huge argument among my friends and I. On one hand, its possible he could finish with 50,000 yards, which only Elway, Favre, and Marino have done, and could finish top 10 all time in passing touchdowns, but at the same time, he's never had a year with 30 TDs, he hasn't been to many Pro Bowls, and never really won any awards. Is Drew going to be in the Hall, or is he going to be like the quarterback version of Fred McGriff?

Mike Tanier: This will probably settle no arguments, but here goes. At present, I do not believe that Bledsoe is a Hall of Famer. As you state, he has a chance to be near the top of the NFL all-time lists. If he winds up first or second on the yards or TDs list, then he would have a resume. But if he's fourth or fifth on those lists, and he doesn't do something crazy like lead the Cowboys to three Super Bowls, then he won't be in.

A quick note for any Pro Football HOF argument: Canton is not Cooperstown. The things that make you a shoo-in for the Baseball HOF, like having a long career and climbing the all-time leader boards, are less relevant to the Football HOF voters, for a variety of reasons. Super Bowls and Pro Bowls matter much more than stats, and as you said, Bledsoe comes up short on both.

You may want to track Warren Moon's HOF progress over the next few years to see what becomes of Bledsoe. Moon has tons of stats and took several teams to the playoffs, but he doesn't have a lot of HOF buzz. Bledsoe may find himself in a similar situation, and the fact that the Patriots became great after Bledsoe left doesn't exactly make him look better in the eyes of voters.

Aaron Schatz: I think Bledsoe would get support for the Hall of Fame if he could lead the Cowboys to just one Super Bowl, not three, as long as they won that one Super Bowl. Otherwise, no. A couple other points:

1) Football's Hall of Fame is much more exclusive than baseball's. There's never been an equivalent to the period in the 1970s when Bill Terry and Frankie Frisch ran the Veterans Committee and basically stuck everyone who played for the Giants or Cardinals in the 1930s into the Hall of Fame.

2) Another good comparison for Bledsoe is Boomer Esiason -- always considered a top quarterback, but never considered the best, led his team to one Super Bowl, and had a big season later in his career in his first year with another team (1993 Jets). Esiason is 12th on the all-time passing yardage list. Here's the top ten:

All-Time Top 10 in Passing Yards
Player Passing Yards Comp% Yd/At TD INT
Dan Marino 61,361 59.4 7.3 420 252
Brett Favre 52,448 61.6 7.1 395 245
John Elway 51,475 56.9 7.1 300 226
Warren Moon 49,325 58.4 7.2 291 233
Fran Tarkenton 47,003 57.0 7.3 342 266
Vinny Testaverde 45,189 56.6 6.9 268 261
Dan Fouts 43,040 58.8 7.7 254 242
Drew Bledsoe 42,365 57.4 6.6 237 190
Joe Montana 40,551 63.2 7.5 273 139
Johnny Unitas 40,239 54.6 7.8 290 253

Among the top ten, Bledsoe has the lowest yards per attempt, the fewest touchdowns, and is below average (although not at the bottom) in both completion percentage and touchdown-to-interception ratio.

Richie Wohlers: Have you had fewer/no guest columns submitted for the site this year, or have you chosen not to publish them? It seems like there used to be more guest columns.

Aaron Schatz: Yeah. Almost no guest columns submitted. Weird, huh? Anybody who wants to submit a guest column is welcome to send an idea or even an entire rough draft. The general policy is usually that we either want something with a unique take on things, or a unique new stat, or something really funny. And historical-type stuff is better submitted in the off-season.

Parker Beck: One of my favorite articles appearing in FO was the one about Ricky Williams a couple years ago, the one about how he was "screwed." If I remember correctly, the idea was that Ricky had been overused and that NFL history was littered with running backs that fell off in productivity after years of heavy use, unless their name was Eric. Further, I believe the magic number was 400, as in 400 touches in a year was a pretty good indicator that the RB was about to take a dive. Ricky's touches in:

2001: 313 runs, 60 passes, 373 total
2002: 383 runs, 47 passes, 430 total
2003: 392 runs, 50 passes, 442 total
01-03 1088 runs, 150 passes, 1238 total

Based on this information, I made some pretty bold statements to a couple of friends about how LaDainian Tomlinson was likely to see a big drop in production this year. My evidence:

2002: 372 runs, 79 passes, 451 total
2003: 313 runs, 100 passes, 413 total
2004: 339 runs, 53 passes, 392 total
01-04: 1024 runs, 232 passes, 1256 total

So my question is, what happened? LT2 is having a fine year, his yards per carry are back up and everything appears to be fine.

Is the answer as simple as amending the original statement to read "unless you are named Eric or LaDainian" or is there something in the numbers that I am missing? My fellow fantasy football players take great joy in reminding me of my statements about LT2 and then calling me an idiot.

Aaron Schatz: The first thing I should clarify is that the magic number was 370, not 400, and the issue at hand was carries, not touches. There's been some discussion on the comment threads a few times about what the research would look like if we included receptions, but as of yet we've never done that research (I'm hoping to do it in next year's book).

Second, LT did have his breakdown year. It was last year, when he dropped to 3.9 yards per carry. People didn't notice as much because the Chargers were winning and he was scoring touchdowns and thus worth a lot in fantasy football. (I don't know if you bought the book, but while Kevin Jones was on the cover, LaDainian Tomlinson was projected to be the number one player in fantasy football for 2005 in the fantasy projections appendix.)

Third, there's still a good chance that LT's average yards per carry will end up lower than both 2002 (4.5) and 2003 (5.3). He's got a 4.6 yd/carry average now but that was 4.3 yd/carry with four minutes left in the fourth quarter of the Washington game.

Fourth, here are your leaders in carries for 2003:

Ricky Williams, 392
Jamal Lewis, 387
Ahman Green, 355
Deuce McAllister, 351

I think LT is an exception that proves the rule. I'm guessing you passed on Lewis and Green in your draft and they went way too high and their owners are a little bummed out. (I should point out that Rudi Johnson had 360 carries last year and seems fine.)

John Healy: I was wondering. What is the relationship between offensive line adjusted yards and RB yards? The Colts lead the league with 5.16 adjusted yards per carry, but their RBs only have 4.29. What is to be made of this? Is the Edge actually underperforming given his offensive line?

Aaron Schatz: First, this is my chance to mention that the offensive and defensive line tables are updated through Week 12. You are right, sort of. Edge isn't really underperforming, because adjusted line yards still can't completely separate the line from the back. But since his return from the ACL injury, Edge just doesn't break many long runs. He gets a lot of 5-7 yarders, but very few carries that go nowhere and very few that go long. You can see that Indianapolis ranks first in not getting stuffed and 31st in percentage of rushing gains that come more than 10 yards past the line of scrimmage. Edge is also on pace to blow apart the record for RB success rate -- he's at 65% and the highest success rate of any starting running back from 1998-2004 was 61%. I think Edge and his line are equally responsible for this remarkably consistent performance.

James: Your an Ass hole Araon. you should give some respect to Seattle. the win big games and you drop them in the rankings. i hope they make it to the super bowl it will really make you look like the ass hole that you are.

Aaron Schatz: Hmmm. You may be confusing me with this guy. This was one of a zillion e-mails from this week accusing me of anti-Seattle bias. I'm getting used to this each time some teams falls a bit in the ratings for some reason that isn't readily apparent from simple wins and losses. The Seattle people were nice enough to keep their complaints to e-mail instead of infiltrating our discussion threads, but this was interesting because the e-mails just kept coming all week, into Friday and Saturday, almost as if there was a delayed reaction. Were mean things about me posted on a Seahawks message board or something? 

The strangest thing about the accusations of anti-Seattle bias is that I may be the most pro-Seahawks writer on the Internet. When the site started in 2003, our first-ever projection system said that Seattle would be the second-best team in the NFC. I picked Seattle to make the playoffs in 2003 and to win their division in 2004 and 2005. I nicknamed Bobby Engram "The First Down Machine" and I've written about him as the most underrated receiver in football numerous times. I wrote a whole article for the New York Times about how Seattle would be much better than people expected this year. We put Matt Hasselbeck on the cover of our book and I chose him as 2005 MVP in the New York Sun season preview (partly, I will admit, because picking Peyton Manning would have been really boring). Man, if this is how I write about a team that I have a bias against, how would I write about a team I had a bias for?

I hereby promise that we will put a Seattle Seahawk on the cover of next year's book. How about Walter Jones? We all agree he doesn't get enough press. In fact, how about we put nothing but guys named Jones on the cover of next year's book? I can see it now:

  • Walter Jones: The Real MVP
  • Thomas Jones: First-round bust no more
  • Dhani Jones: A man of wealth and taste
  • Pac-Man Jones: Driving Jim Schwartz to drink
  • Donnie Jones: America's Best Punter!
  • Kevin Jones: Um, whoops

Nick: I Just have a question about my home team the St. Louis Rams and how it possible for them to be that low. They win five games and are below teams with two wins and below the Titans, Cardinals, and Saints who they beat. They have beaten a good team in Jacksonville also. They have played Seattle close twice and were beating the Colts until Bulger went down. They have also won without their starting quarterback and wide receivers.

Aaron Schatz: Here's the other reason why I clearly have nothing against the Seahawks. I don't know what's up with the Rams, who are currently 30th in DVOA. They were also ranked very low by DVOA last year (30th despite going 8-8) and the year before (13th despite going 12-4). I wanted to take a closer look, at least at this year's Rams, to see if there was an explanation.

Let's get this out of the way first: DVOA is entirely based on stats and therefore does not take into account injuries of any kind. That's part of the reason why the Steelers are so low and why the Jaguars (who keep playing teams with injured quarterbacks) are so high. I don't want to make subjective judgments based on injuries, I prefer to leave that to the readers. For this same reason, the fact that they were beating the Colts until Bulger went down doesn't matter -- the plays in the second half of that game count just as much as the plays in the first half.

Are the Rams just getting away with a lot of close wins? Their Pythagorean win total, based simply on points scored and allowed, is 4.6 wins, and the Rams are 5-6.

DVOA, however, seems to disagree. Here's another fun week-to-week DVOA chart, except this time we'll run both DVOA and VOA, which is not adjusted for strength of opponent. DVOA is the blue line with yellow squares, VOA the darker blue line with little diamonds.

VOA says the Rams' wins have pretty much been all squeakers. The Arizona win was a little bigger, and the Rams were actually outplayed by Jacksonville even though they won -- which is balanced out by the fact that they outplayed San Francisco even though they lost.

Look at DVOA, though, and the Rams have only one positive game all year, the Jacksonville win. All their other wins have been against bad teams, and therefore after adjustment they end up negative.

Without considering strength of schedule, is the system underrating the Rams in some way? Well, the "success points" system that starts the computation of DVOA currently caps the value of all plays at 40 yards (not counting the possible touchdown bonus). People have suggested that this is unfair to teams that depend on big plays. (A few folks have pointed to this as a reason why certain past Indianapolis teams were ranked low, like the 10-6 Colts being 20th in 2002.) In the past rewarding plays over 40 yards has never improved the system's correlation with wins, but that doesn't mean it isn't worth experimenting again. The Rams are still running the offensive style that made them the Greatest Show on Turf, so perhaps this is the problem.

Except the Rams do not lead the league in offensive plays over 40 yards. And the Colts are nowhere near leading the league in offensive plays over 40 yards. When you see which two teams lead the league in offensive plays over 40 yards, one of them will completely blow your mind.

Miami: 11
New York Giants: 11
Denver: 10
(St. Louis has 9, Indianapolis 5)

Somehow, the idea that Miami offense is underrated by our system doesn't quite fly with me. And I also don't buy the idea that the defense allowing the fewest plays over 40 yards is also underrated by our system.

New York Jets: 2
New York Giants: 3
Indianapolis: 3
Cincinnati: 3

Yes, that's right, the powerhouse Miami Dolphins lead the league in plays over 40 yards, and that amazing Jets secondary has done the best job of preventing plays over 40 yards.

The moral of the story is that I have no idea what the heck is going on with the low rankings for the Rams. It's possible that the Rams have just been lucky for three years. Even if DVOA has them a little low, though, I don't think they're a playoff team.

Nick Gourevitch: Any plans for a playoff odds report similar to Baseball Prospectus?

Aaron Schatz: This was one of the things that just fell off the radar this year because I'm overworked, with a lot more responsibilities from the FOX deal, a lot of management to do, plus the two and a half weeks of work time that I lost at the start of the season. It's a great idea and I want to do it but if we do it, we have to do it right, with the right amount of research. A lot of people have volunteered to help out but no matter how many volunteers we have, I still have to manage the project, and I just don't have the time this season. (In other words, ten people shouldn't send me e-mails volunteering, because that's not the problem.) I hope we can do it for next year and have it going from Day 1 of the 2006 season.

Jon Lewallen: Given the discussion over whether it's worth a team's while to sign Terrell Owens next year, I'm hoping FO can confirm one of my suspicions. I know you guys only have data going back a few years, but when was the last time a team won the Super Bowl with the league's best wide receiver, DPAR- or DVOA-wise?

Aaron Schatz: Actually, the answer is never (at least, since 1998). The closest was 2000, when two teams dominated the league. Derrick Mason led the league in DVOA but Tennessee narrowly lost to other top team, Baltimore, in the second round of the playoffs.

I don't really think that's an argument that TO is meaningless, though. We're talking just seven years of data, 1998-2004, and plenty of Super Bowl champions have had great receivers. In 1998, Rod Smith and Ed McCaffrey ranked third and sixth in DPAR (and Shannon Sharpe was the top tight end). Isaac Bruce was fourth in both DPAR and DVOA in 1999. Although they barely made the 50-pass minimum, David Givens was third in DVOA in 2003 and Deion Branch fifth in DVOA in 2004.

It is probably better to have two B+ wide receivers than one A+ wide receiver and one D+ wide receiver. But that's not an argument to get rid of your star receiver, its an argument to get rid of your James Thrash-type receiver

Finally, here's this week's "question I don't know the answer for and therefore we open it up to the readers."

Dennis Hollod: How does a hook and lateral (sometimes called hook and ladder) play get scored. QB Adams throws a 10 yard forward pass to Bell, Bell laterals to Cooper who goes 50 yards for TD. Does QB Adams get credited for a 10 yard pass or a 60 yard TD pass? Does Bell get credited with a 10 yard reception? Does Cooper get credited with a 60 yard TD reception or a 50 yard run?

Whoever knows the answer, go ahead and let us know in the comments. No need to have ten different people post it, of course.

Posted by: Aaron Schatz on 04 Dec 2005

63 comments, Last at 27 Apr 2007, 12:02am by Ken Roberts


by Todd S. (not verified) :: Sun, 12/04/2005 - 1:31am

Can I just say that I love getting new content at 11:00 PM on a Saturday? Good stuff.

by Bill (not verified) :: Sun, 12/04/2005 - 1:36am

See link for the situation I'm basing this off of.

Adams gets credit for a 60 yard TD pass. Bell gets a 10 yard reception, Cooper gets 50 receiving yards and a TD with *no* reception.

by Bill (not verified) :: Sun, 12/04/2005 - 1:38am

Oops, here's the correct link.

by Dylan (not verified) :: Sun, 12/04/2005 - 1:52am

NO .... go back to the Vikings/Broncos game from two years ago .... End of the first half, Culpepper throws a hail mary to moss for 50 yards, who laterals to moe williams for the last 15 yards .... Moe gets a 65 yard TD reception, daunte gets a 65 yard touchdown pass, moe and daunte lead me to fantasy football victory.

by Travis (not verified) :: Sun, 12/04/2005 - 1:55am

Also, it would be listed in the NFL.com gamebook box score as "Cooper 60 yd. pass play by Adams".

by Travis (not verified) :: Sun, 12/04/2005 - 1:59am

#2, not #4, is correct.

by Dylan (not verified) :: Sun, 12/04/2005 - 2:04am

My bad ... I remember both Culpepper and Williams getting an FFL TD on that one though .... is Yahoo just that bad with stats, or what?

by Catfish (not verified) :: Sun, 12/04/2005 - 2:17am

When did people get the idea that when the analysis and opinion of someone in the media differs from their own, the media must be biased? That seems to be a very arrogant argument to make. It assumes that everyone agrees with you, and anyone who expresses differing opinions must have ulterior motives for disguising 'The Truth'.

by Travis (not verified) :: Sun, 12/04/2005 - 2:20am

Re 7:

Both would get TDs, but Williams would only be credited with 15 receiving yards, not 65. Yahoo might have credited it as it shows up in the box score, so your memory might be correct.

by Kibbles (not verified) :: Sun, 12/04/2005 - 3:08am

Given the discussion over whether it’s worth a team’s while to sign Terrell Owens next year, I’m hoping FO can confirm one of my suspicions. I know you guys only have data going back a few years, but when was the last time a team won the Super Bowl with the league’s best wide receiver, DPAR- or DVOA-wise?

That's a bad question to ask on the T.O. situation. After all, when was the last time Terrell Owens was the league's best wide receiver, DPAR- or DVOA-wise? This season, he's the #27 WR in the league according to DVOA- and I can tell you that there have been tons of teams that have won the Superbowl with the #27 ranked WR or better.

In fact, when do you think was the last time that Terrell Owens was even in the top 16, according to DVOA? The answer's going to surprise you. Wait for it... wait for it... 2001. Seriously. He has far too many drops to really be considered the best WR in the entire NFL, even without the off-the-field issues.

Now, if you're asking whether a team HAS TO HAVE the best WR in the league to win the superbowl... of course they don't. They don't have to have the best QB, the best left tackle, or the best punter, either. They simply have to have the best aggregate talent- and adding Owens to the roster certainly goes a long way towards boosting a team's aggregate talent level.

by John (not verified) :: Sun, 12/04/2005 - 4:18am

Aaron, "The exception that proves the rule" is a semantic misnomer. It is a very old phrase and in this case "prove" means test, like as in the "proving grounds".
I can imagine your analytical mind can figure out that an exception could never prove a rule in the traditional sense. That just doesn't make any sense.
Sorry if this is disjointed, I'm drunk.

by Josh (not verified) :: Sun, 12/04/2005 - 4:51am

"We put Matt Hasselbeck on the cover of our book"

Actually, you put Hasselback on your cover, which clearly shows your anti-Seattle bias.

by NF (not verified) :: Sun, 12/04/2005 - 5:25am

Drunk people provide the etiology of misused phrases on a football statistical analysis website.

"Only in America!!!"

by VarlosZ (not verified) :: Sun, 12/04/2005 - 8:42am

I think LT is an exception that proves the rule.

Pet peeve: This is probably the most widely misused phrase in the English language. As written, it's borderline-gibberish. A lone case in which the rule fails shows the rule to be true . . . because each rule must have one (and only one) exception? What?

In this phrase, "proves" means "tests." As in, for example, "military proving grounds." It does not mean "shows to be true."

Excellent mailbag, as always, and I apologize for nit-picking.

The Grammar Nazi

by James Gibson (not verified) :: Sun, 12/04/2005 - 9:18am

Moon v. Bledsoe:

I don't know if Bledsoe should be following Moon's chances. You mentioned Super Bowls and Pro Bowls, and while Moon is short Super Bowls, he went to 9 Pro Bowls, as I've brought up in some of MDS's HOF articles. Bledsoe has been to 4 Pro Bowls, which is well short. QB Jim Kelly also only made 4 Pro Bowls, but he also made 4 Super Bowls. I don't really like Bledsoe's HOF chances at all, but I like Moon's. And if you look at the raw stats above, it almost seems like Testaverde is close to equal to Bledsoe, and I'd say his chance at Canton is about 0.

For me, Bledsoe's low yards/attempt really sticks out as a negative. IMO, that's the most important stat up there as an individual stat.

by Trogdor (not verified) :: Sun, 12/04/2005 - 11:16am

Kibbles, you fool. Do you not know that only the best QB in the league can win the Super Bowl that year? Have you learned nothing? Stats, making good decisions, accurate throws - meaningless! If the team wins, the QB is the best!

You should be ashamed of yourself for bringing such a sensible approach to that topic. I expect an apology any time now.

by Catholic Samurai (not verified) :: Sun, 12/04/2005 - 12:26pm

RE #15:

Warren Moon is one of the most debated HOF canidates in the NFL, and may in all of pro sports (sans Charles Hustle). Basically it goes like this

"He went to 9 Pro Bowls with 3 different teams!"

"He never went to the Super Bowl!"

"He is in the Top 5 in the most important quarterback statistics!"

"He played in a system that allowed him to throw like crazy!"

"If you count his other league stats, he has the most passing yards of all time!"

"You can't count those!"

"He had to play there because the world wasn't ready for a black QB (debateable)!"

"Doug Williams!"

And so on and so forth. As to what side I'm on, I'll let my autographed Moon Houston, Seattle and Minnesota helmets and my throwback Oilers jersey do the talking.

by krugerindustrialsmoothing (not verified) :: Sun, 12/04/2005 - 12:38pm

Moon also won 3 or 4 Championships in Canada (the Grey Cup), and well, it is called the "Pro Football" Hall of fame. So it doesn't seem to be unreasonable to include a factor for his time up here. Even if it does come at somewhat of a discount to his NFL time.

by Walt Pohl (not verified) :: Sun, 12/04/2005 - 12:57pm

Catfish: The fact that you are trying to cover up media bias against me demonstrates that you trying to disguise the Truth. The only question that remains is: what are your ulterior motives?

by admin :: Sun, 12/04/2005 - 1:55pm

If I remember correctly, the rules of the Hall of Fame specifically state that only performance in American leagues is to be considered. (I'm not saying I think this is fair, I'm just saying that I think this is the rule.)

As for "the exception that proves the rule," wow, that has got to be the most nitpicky nitpicking in this history of nitpickory. Which is also borderline gibberish, and, like the phrase "the exception that proves the rule," gets the point across pretty well.

by Falco (not verified) :: Sun, 12/04/2005 - 2:04pm

Regarding the Rams, in looking at the chart, it appears that the Rams ranking of #30 is based on mean DVOA, and not median DVOA performance. The Rams have 7 mediocre to below average games, and 4 absolutely horrid games. The median performance looks to be closer to -10.0 DVOA. That would put the Rams closer to #20 than #30, and the truth is probably somewhere between the two and closer to #20. If you offered me a best of 7 between Rams and the 7 teams rated in front of them, I would take the Rams over all 7. Would they lay a stinker in one or two games of each series? Probably.

The Rams may have a tendency to do certain bad things in some game situations that drive the overall DVOA down, oh, like abandoning the run too early and going for broke in an attempt to get back in the game. Maybe the coordinators panic with their play calling, and things snowball. I don't watch them enough to know the exact reason.

However, these really bad performances may be less indicative of the real average Rams team (though they must be considered), than the other 7 games.

If you consider median, I think you get a more accurate picture of the Rams. A team that is below average, but capable of playing with most teams, while occasionally inclined to lay a stinker.

I also think the fact that the Rams played the 49ers in week 1, rather than in one of the weeks Alex Smith was QB, is pulling that SF game DVOA down too much.

by Vince (not verified) :: Sun, 12/04/2005 - 2:43pm

Nick Gourevitch: Any plans for a playoff odds report similar to Baseball Prospectus?

You can find something similar here:


#2 is correct on the scoring question. Michael Vick once threw a pass that was tipped and caught by a lineman. The linemen turned around and (wisely) pitched it back to Vick, who scampered for about 19 yards. The linemen was "credited" with 1 catch for about -7 yards. If you check Vick's career receiving stats, you'll see 19 yards gained on 0 catches.

by calig23 (not verified) :: Sun, 12/04/2005 - 2:50pm


Wouldn't that be a penalty, though? You can't throw to an ineligible receiver, which, I assume, the lineman was on the play in question.

by admin :: Sun, 12/04/2005 - 3:08pm

What kind of analysis/projection system is that coolstandings.com based on? It's interesting -- I'm guessing that the FO fans are asking for something additional based on FO stats, of course...

by Todd S. (not verified) :: Sun, 12/04/2005 - 3:19pm

#23 He mentioned that the pass was tipped. Once a defensive player touches the ball, anyone on the offense is eligible to catch it.

by Todd S. (not verified) :: Sun, 12/04/2005 - 3:24pm

#24 Looks like PF vs. PA. They mention Bill James and Daryl Morey's "Pythagorean Theorem."

by dedkrikit (not verified) :: Sun, 12/04/2005 - 5:29pm

Aaron -
RE to HoF & American Leagues
Didn't the CFL have a handful of US teams for a while? I remember a Shreveport, LA Pirates team being around for a couple years...

by Hector (not verified) :: Sun, 12/04/2005 - 8:35pm

Without looking it up, I'd think that Moon compares very favorably to Fouts, who I don't believe had a lot of trouble making it into the HOF.

by MdM (not verified) :: Sun, 12/04/2005 - 8:53pm

I'm glad someone explained to me the exception testing the rule thing. It kind of makes my skin crawl when people misuse it, too. Sorry, Aaron..I love your work, man! :)

by Ron Mexico (not verified) :: Sun, 12/04/2005 - 10:33pm

Yep! Click my name and you'll see 16 receiving yards with no catches.

by JonL (not verified) :: Sun, 12/04/2005 - 10:50pm

RE: #10

I didn't mean to imply with my question that Terrell Owens was the best wide receiver, it was more of a hypothetical. And I do think it's interesting that, at least going back to 1998, no team that's won the Super Bowl has had the best wide receiver for that year. It's most likely a statistical anomaly, but it's interesting, at least to me.

by Larry (not verified) :: Sun, 12/04/2005 - 11:07pm

Re #31:
The best WR is 1 player, and is thus on exactly one team. So, the chance of such a player being on the Super Bowl winner is nominally 1-in-{# of teams in the NFL}. It is probably a little better than that, given that a team with a good player is a better team than a team with no good players, but its only 1 player. The better question is to compare the answer to this question with the same question for RB and QB. I'm guessing the answer is likely no more than 1 or 2 teams in either of those categories have won the SB since '98.

by JonL (not verified) :: Sun, 12/04/2005 - 11:14pm

RE: 32

...which was the next thing I was thinking about. It is one player, true, but given how much importance sports journalists and analysts place on that one person, I thought it was interesting to note.

to take it a bit further, it might be interesting to look the Super Bowl teams from the last seven years (for now) and see where each player ranks in DVOA to see if there is any consistency in the rankings of certain positions for each team.

my hunch is that the old bromide "defense and a solid running game wins championships" is in fact true (at least in 9 out of 10 cases), but (showing my ignorance) has anyone looked at why this might be? Might it be because running plays involve fewer variables that could hurt the offense than passing plays?

by andrew (not verified) :: Mon, 12/05/2005 - 12:47am

I used to read Arion, Mage of Atlantis. You look nothing like him, Aaron.

by Larry (not verified) :: Mon, 12/05/2005 - 6:45am


It is certainly worth looking at whether good teams tend to be at the top of any particular categories, but at the end of the day, good teams score points (Kyle Orton led teams excepted), which means somebody has to be making some plays sometime, and prevent points, which means the defense has to do something. I suspect you're wrong, though, the QBs will most likely show up at the top, since they're assigned the result of over 50% of the plays, and winning teams score points, therefore their QBs have to at least rack up DPAR, and you need DVOA to move the ball down the field. So, my guess of the moment is you'll find "Highly rated QBs win SBs," but correlation does not imply causality.

by Anson (not verified) :: Mon, 12/05/2005 - 10:34am

Aaron wrote: "...Boomer Esiason — always considered a top quarterback, but never considered the best." ... Well, Esiason did win an MVP award, which means he was considered the league's best for at least one season.

by Rick (not verified) :: Mon, 12/05/2005 - 10:43am

Best WR on a Super Bowl-winning team?

Last player like that who really stands out was Jerry Rice.

Aside from Rice, I can't think of any WR who was clearly the best WR and was also on a Super Bowl winning team.

If Carolina meets Indy in the Super Bowl this year, this debate could become very interesting.

by masocc (not verified) :: Mon, 12/05/2005 - 11:00am

16 yards on 0 catches? Did he make the first down?

What's that make the DPAR?

by zlionsfan (not verified) :: Mon, 12/05/2005 - 12:10pm

Re CFL: Yes, there were some American teams for a while, including the 1995 Grey Cup champion Baltimore We-can't-use-the-name-Colts. (see link)

by James Gibson (not verified) :: Mon, 12/05/2005 - 12:29pm

My instinct would be that the #1 QB or a top QB would show up more than the other positions, because yards/pass has been shown to be a more determining statistic than other single statistics (as opposed to compliation statistics like DVOA), such as yards/carry. Of course, a stout defense like Baltimore's in 2000, will couteract a bad offense. But I'm pretty sure they led in fewest yards/pass against.

WRs, I think, would be a little less important overall because as individuals they are involved in a lot fewer plays than the QB being that there are multiple WRs per team. While I think the passing game gives us more of a look into a solid team than a running game, an individual RB would probably be more involved in total # of plays than an individual WR. A QB would be involved in the most.

by Jason (not verified) :: Mon, 12/05/2005 - 1:59pm

Would the lateral really be counted as a pass play in the stats and box score? I thought a lateral that is not caught was counted as a fumble (and open to be picked up by the defense), therefore wouldn't the yards gained on a lateral be a run rather than a pass? If a busted lateral was counted as an incomplete pass I'd say yeah, it should be passing yards. But maybe I'm wrong on the lateral/fumble thing.

by Andrew (not verified) :: Mon, 12/05/2005 - 3:10pm

Anson #36:

Esiason won a split MVP in 1988 with Roger Craig and Randall Cunningham also winning. Not impressive, considering Cunningham went on to win two more MVP's (1990 and 1998), while Esiason didn't. Esiason's stats really aren't far different from Cunningham once you adjust for Cunningham's 1000 fewer pass attempts, except that Cunningham was better at avoiding INT's, and could run. Esiason had 3 winning seasons in a 12 years starting. Cunningham had 5 in 8 years starting. Is Cunningham HOF material?

by Greg F (not verified) :: Mon, 12/05/2005 - 3:25pm

Warren Moon should make the HOF in my opinion.

Click the name to see a discussion on "the exception that proves the rule"

by JonL (not verified) :: Mon, 12/05/2005 - 3:43pm

RE: #42

I don't really disagree with your overall point, but three winning seasons in 12 years can't be put entirely on Esiason. He did spend many of those years with the Bengals and the Coslet/Carroll/Kotite Jets. And let's not forget the Arizona Cardinals.

by Scott de B. (not verified) :: Mon, 12/05/2005 - 3:55pm

Without looking it up, I’d think that Moon compares very favorably to Fouts, who I don’t believe had a lot of trouble making it into the HOF.

Without taking a position on whether or not Moon belongs in the Hall, I don't think you can directly compare their stats. Football changed a lot between the late 70s/early 80s and the late 80s/90s.

by Tecmo Bo (not verified) :: Mon, 12/05/2005 - 4:33pm

I don't know about the Fouts/Moon debate. I'd take a bearded QB over a mustachioed one every time. Though maybe Jeff Hostetler is the exception that proves the rule.

by mshray63 (not verified) :: Mon, 12/05/2005 - 4:36pm

Re Catfish in #8: I believe the answer to your question (rhetorical though it may be) is Inauguration Day 2001.

by Murr (not verified) :: Mon, 12/05/2005 - 5:31pm

Considering the karma boomerang that has hit the last couple of teams to rant and rave about their teams' low ranking (Falcons and Redskins, I'm looking at you), I'm quite happy to hear about the Seattle fan over-reactions. Should bode well for my boys tonight.

Denver, of course, is... the exception that proves the rule. (HA!! Suck on it and like it, etiology weenies!)

by Tormanator (not verified) :: Mon, 12/05/2005 - 5:56pm

Why is no one mentioning the obvious example of how stats at QB do not a HOFer make? I refer, of course, to Vinny Testaverde, who ranks obscenely high in almost every career category, yet will never even sniff the outskirts of Canton.

by Tormanator (not verified) :: Mon, 12/05/2005 - 6:06pm

Re: mshray63 in #47, I believe you may be right.

by JG (not verified) :: Mon, 12/05/2005 - 6:55pm

If I remember correctly, the rules of the Hall of Fame specifically state that only performance in American leagues is to be considered. (I’m not saying I think this is fair, I’m just saying that I think this is the rule.)

Does it specifically say 'American' or does it say USA leagues? I know more than a few Canadians and Latin Americans who would claim that they are just as much "American" as someone from the United States.

by C (not verified) :: Mon, 12/05/2005 - 7:17pm

Okay, I looked some stuff up, for both the Super Bowl champions and runners up, based on DVOA rank among the players who met the FO minimum to be ranked.

Note that I've done two things that might be controversial: one is to ignore running backs receiving DVOA rank (and WR rushing DVOA, but the effects there are less significant). Since I don't think there is a clean way to combine both RB rushing and pass catching, I'm going to ignore the out-of-the-backfield stuff and pretend we're still in 1980 for this work.

The other is more of a design choice: I chose DVOA over DPAR in terms of how to rate a player. I did this because DPAR is more dependent on the parameters of the offense, so dependent on number of carries, that it somewhat masks individual talent in a way that DVOA doesn't. DVOA is still affected by one's job in the offense, certainly, but not as much because it is normalized.

Note also that though I rate everything as "2001 NE QB" or "2004 PHI WR1" that actually means the end-of-season stats from just Tom Brady or just TO. I'm not taking weighted averages or per-game numbers or anything like that. This is another reason to go with DVOA rather than DPAR: injuries can seriously affect DPAR but are less problamatic for DVOA (obviously nagging injuries that players attempt to play through will have an effect, but I can't correct for that with the data at hand).

2004 NE:
QB: 2
RB1: 8
RB2: N/A
WR1: 27
WR2: 30
WR3: 5

2004 PHI:
QB: 5
RB1: 15
RB2: 3
WR1: 17
WR2: 34
WR3: N/A

2003 NE:
QB: 14
RB1: 34
RB2: 47
WR1: 3
WR2: 40
WR3: 46

2003 CAR:
QB: 22
RB1: 20
RB2: 51
WR1: 48
WR2: 52
WR3: 47

2002 TB:
QB: 4
RB1: 34
RB2: 37
WR1: 36
WR2: 8
WR3: 61

2002 OAK:
QB: 3
RB1: 4
RB2: 36
WR1: 2
WR2: 24
WR3: 41

2001 NE:
QB: 17
RB1: 33
RB2: N/A
WR1: 15
WR2: 44
WR3: N/A

2001 STL:
QB: 1
RB1: 1
RB2: 2
WR1: 8
WR2: 6
WR3: 1

2000 BAL:
QB: 38
RB1: 13
RB2: 4
WR1: 43
WR2: 57
WR3: N/A

2000 NYG:
QB: 6
RB1: 24
RB2: 35
WR1: 3
WR2: 16
WR3: N/A

Analysis: There doesn't seem to be any hard and fast rule here, though the QB DVOA does seem to normally track pretty well. The 2003 game and the 2000 Ravens are the only time a QB that wasn't in the top ten in the league in DVOA made the Super Bowl. Six different teams in the study brough a wideout in the top ten in the league to a Super Bowl, going 3-3. Teams without a wideout in the top ten went 2-2. Five brought someone in the top ten in rushing DVOA to the Super Bowl, going 2-3. Teams without went 3-2, obviously.

All data from the FO archives. All mistakes are my own.

by thad (not verified) :: Mon, 12/05/2005 - 8:24pm

re james gibson
the Titans were first in 2000 with 4.65 net yards per pass
the Ravens were 7th with 5.32.

by JonL (not verified) :: Mon, 12/05/2005 - 10:04pm

RE: #52

From a cursory glance it appears that teams would do well to have either a top-10 running back or a top-10 wide receiver, but not necessarily both

I'd be interested to see the defensive figures

by JonL (not verified) :: Mon, 12/05/2005 - 10:18pm

...so here they are. these are the unweighted team defense ranks, in the order of total/pass/rush

NE 6/11/4
PHI 13/10/20

NE 3/3/14
CAR 14/15/12

TB 1/1/9
OAK 5/10/3

NE 19/12/27
STL 3/4/5

BAL 1/7/1
NYG 14/18/4

Next I think it might be interesting to look at drive stats, but I can't get to that right now.

by Lou in Cincy (not verified) :: Tue, 12/06/2005 - 12:11pm

"I think LT is an exception that proves the rule. I’m guessing you passed on Lewis and Green in your draft and they went way too high and their owners are a little bummed out. (I should point out that Rudi Johnson had 360 carries last year and seems fine.)"

Actually, I was a little worried about Rudi after the workload he carried last year. And in fact he's had a little publicised knee injury nagging him all year. Luckily, Marvin Lewis has almost preternatural foresight when it comes to his roster. With Chris Perry healthy and spelling him, plus a more balanced attack, Rudi is finally at full strength for the (ulp) PLAYOFF RUN!!!!

by Parker (not verified) :: Tue, 12/06/2005 - 12:37pm

RE: #44

This has always been a sticky point for me as well. Football is very much a team sport, so how do you apply the teams success to an individual, whether it's positive or negative? On the other hand, when discussing the best of the best, how can you ignore it?

And why is it that the W/L record and # of SB's is almost exclusively applied to QB's? Is that simply because everyone believe that the QB is so much more responsible for the overall outcome of a team than any other position or even groups of positions?

I just think it's a bit suspicious to ignore the fact the Boomer only had 3 winning seasons but give Kelly credit for 4 SB appearances. Of course, no one here has done that, so it's sort of a moot point.

By the way, thank you Aaron for answering my question. It begs a followup: Why not include receptions as part of the RB's workload, and is the punishment recieved by a RB considerably less on a reception than a rush? Additionally, should any consideration be given to RB's that pick up blitzes more often, as I imagine that is additional strain.

by Andrew (not verified) :: Tue, 12/06/2005 - 12:54pm

Parker #57:

If you watched the Eagles in MNF last night or any Jets games without Chad Pennington, you saw the difference between a team with and without a good QB. Similarly, watching Brett Farve and the 21 dwarfs with Green Bay, you can see how Farve has carried this bunch of replacements as far as they have gone (not many wins, but very competitive in most games)

Obviously Farve, Pennington and McNabb deserve a ton of credit for the wins and losses their teams gain or endure.

If Boomer was really a HOF caliber QB, those teams post 1990 he was on would have done much better. Reality was, they really were not competitive.

by Dave (not verified) :: Tue, 12/06/2005 - 3:12pm

RE: #41 - I believe you might be confused as to the situation involved. The assumption is that a legal forward pass has been completed. THEN a lateral occurs which is a continuation of the passing play. I think that's why it counts as receiving yards, and can't be considered rushing yards. Also, it can't be incomplete, since it's either got to be a fumble or illegal second forward pass.

RE: #51 - I have heard before that the term "American" is used in other countries (and by those hailing from other countries in the US) to refer to any or all people from both North or South America.

However, I think you'd be hard pressed to find very many people in the US who would ever use the term "American" to refer to anything other than being from the United States of America. It's not any type of slight - just a typical usage, since most people refer to their country of origin, not region or continent (as would be the case with other countries).

(Sorry for the one large paragraph. I don't think the line breaks are working.)

by Dave (not verified) :: Tue, 12/06/2005 - 3:22pm

RE #59 (my previous post) - I guess the line breaks just don't work in the preview, but work in the actual post...

by hrudey (not verified) :: Thu, 12/08/2005 - 3:50pm

When scoring a lateral TD play, it's also important to not forget to add the following:

(PAT Failed. John Carney kick wide right.)

by Anson (not verified) :: Sat, 12/10/2005 - 6:29pm

Andrew #42 wrote "Esiason won a split MVP in 1988 with Roger Craig and Randall Cunningham also winning." That just ain't true. Esiason won the AP's MVP award outright. Craig was the AP's Offensive Player of the Year and Cunningham won the Bert Bell Trophy from the Maxwell club. But regardless, your assertion that Esiason's MVP award is "not impressive" because Cunningham went on to win two more MVPs just doesn't make any sense. But my original point stands: Esiason was clearly the best QB in football in 1988, and from 1985-1989 (excluding the 1987 strike season) he was second only to Marino

by Ken Roberts (not verified) :: Fri, 04/27/2007 - 12:02am

Nick Gourevitch: Any plans for a playoff odds report similar to Baseball Prospectus?

You can find something similar here:


and here:

Sports Club Stats
although the "sophisticated algorithm" it uses is flipping a coin.