Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Futures: Maurice Hurst

A heart condition discovered at the combine has put the Michigan lineman's career in limbo, but Hurst had the best film of any defensive tackle in this year's draft class.

03 Aug 2005

Pro Football Prospectus Mailbag

by Aaron Schatz

We've received a number of questions and comments since the release of Pro Football Prospectus 2005. (On sale now!) We've put the best ones together in this mailbag -- which can also be used as an open thread to discuss the book.

Don't forget that we have a new contact form which you can use to e-mail any of the writers. We'll be doing a regular weekly mailbag during the season, and to be honest asking a question through the contact form is probably a better way to get your question answered than asking it in a discussion thread.

Be aware that I 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.

I should also mention the Pro Football Prospectus 2005 book tour, coming soon to a town near you. All of the cities, times, and locations are listed here but here's the city-by-city rundown:

August 8: Boston
August 9: New York
August 10: Philadelphia
August 11: Pittsburgh
August 16: Indianapolis
August 17: Detroit
August 18: Minneapolis
August 20: Chicago
August 22: Denver
August 23: Dallas
September 6: Washington DC
September 7: Philadelphia, again

There is also a big radio day on Thursday August 4 where I'll be appearing in most NFL markets that aren't on the actual tour. You can find all those times and stations here.

Before we start in with the questions, I should apologize for the fact that there are a few errors and misspellings in the book. I know that some people feel that a book with misspellings automatically must be dismissed as the work of idiots who don't know anything about football. For the rest of you, the mistakes are just annoying. Trust me, I feel the same way.

The worst of the errors is the misspelling of Matt Hasselbeck's name right on the cover, but that one isn't actually our fault -- that comes from the folks who designed the cover. The other errors are all on us, though, and include:

  • I made the same mistake in the Julius Jones comment that is made in this article. Yes, I feel like a moron.
  • Keenan McCardell was signed by the Jaguars as a free agent in 1996, not chosen by them in the expansion draft.
  • Randy Moss does not play all 16 games on grass this year. He does play one game on turf against the Jets. (Although, unless he sets an NFL record with 300 receiving yards in that game, the point about Moss on grass vs. Moss on turf still has merit.)
  • In the Torry Holt comment, we said "Rodney Harrison" when we should have said "Marvin Harrison." I think this was just part of our attempt to give Rodney Harrison as much respect as possible by getting him extra mentions.
  • In the Todd Pinkston comment, we said "Fred Gibson" when we should have said "Reggie Brown."

Now that we have made a mea culpa for our human weaknesses and enjoyed a round of self-flagellation, let's get on to some real questions.

Michael Thompson: Will you be publishing the projected DVOA by team for 2005 in your normal format (offense and defense, rushing and passing)? I didn't see it in the book, maybe I missed it because I haven't read the entire thing yet.

Yes, the week the season starts we'll run:

  • A projected DVOA article like last year with mean projected wins and projected OFF, DEF, and ST for each team. (We don't do separate projections for rushing and passing.)
  • A rundown of predictions from each member of the staff as well as "composite" site predictions.

One of the biggest complaints we've gotten is that the book doesn't give a specific prediction of each team's won-loss record. Instead, we give each team's percentage chance of finishing in five different categories, which looks like this:

Denver Broncos

  • Leinart Land (0-4): 9%
  • Bad Team (5-6): 15%
  • Mediocre (7-8): 23%
  • Playoff Contender (9-10): 24%
  • Super Bowl Contender (11+): 28%

Baseball Prospectus, as many of you know, doesn't include any team predictions in their book. Football magazines always predict exact won-loss records for every team, which is a complete fool's errand given the way NFL teams unexpectedly improve and decline from year to year. We thought this was a good compromise that gave an indication of how likely it was that each team would improve or decline. Well, apparently, the readers disagree, so we'll probably stick specific win predictions in next year's book, with the proper language about how each predicted record is really the mean of a range of projections, not a specific exact projection.

Mark Padden: Any chance of posting your fantasy projections online for book buyers? Just providing one list as an appendix was frankly a bit less than I expected from the book. Thanks.

You aren't the first person to suggest that we offer fantasy projections in a few different formats in the appendix next year. It is definitely on the list of things to consider changing to improve the 2006 book. On the other hand, we don't want to run table after table after table that just represent different league formats and increase the cost of the book to the consumer. (Some people have complained that we don't have tables of the best fantasy players by position, but I figured one big table took care of that -- if you need a running back, just look down until you hit the next listed running back -- while conserving space.)

Baseball Prospectus actually doesn't include a big list of all fantasy players in their book. What they do is offer that as a downloadable spreadsheet as part of a premium subscription package. So we were stuck trying to figure out what to do -- we don't have a premium subscription package. We knew people would want the projections in a list, preferably a spreadsheet they could play with. We can't really post the projections for book buyers because people who bought the book online would have to forward us the receipts, and that would still leave people who buy the book in actual stores out in the cold. What we decided on, as a one-year solution, was to offer the projections with a small $10 donation -- pretty cheap compared to most downloadable projection sheets at other fantasy sites, I think -- and let people know that we were trusting them to buy the book as well. We hope to have something better next year, and we hope people understand that this is our first time doing a book of this type, size, and magnitude so not quite everything was figured out perfectly.

Dan Babbitt: Will you all be issuing revised projections for Travis Henry and Chris Brown? Also, are you aware of anyone doing the type of work that you or Baseball Prospectus does, but with the NBA?

Revised RBs RuYd RuTD RecYd RecTD
Travis Henry 788 5 188 1
Chris Brown 634 5 172 0

Yes, the spreadsheet we're sending out with a donation does include revised projections for Henry and Brown and a few other players. Lest anyone think we want them to send us $10 just for a handful of new projections, we present the most important changes here. They include the Henry and Gardner trades, Johnnie Morton signing in San Francisco, and changes in the expected roles in Seattle and Atlanta.

Revised WRs RecYd RecTD
Bobby Engram 648 4
Jerome Pathon 543 4
Arnaz Battle 344 2
Johnnie Morton 505 2
Rod Gardner 469 3
Michael Jenkins 662 4
Peerless Price 503 3
Roddy White 462 3
Dez White Just say no.

As for your second question, you betcha! Start with Roland Beech, the man behind the NFL site Two Minute Warning, who also has an NBA site called 82games.com. You also have John Hollinger -- unfortuntely, he is no longer associated with the Prospectus family, but he's still writing a great annual book (now called Pro Basketball Forecast) and he writes for ESPN Insider as well. And of course you have to check out the "APBRmetrics" forum and this central links site for basketball analysis, both run by the talented and very friendly Kevin Pelton.

Paul Barrow: Please could you advise the degree to which you cover individual defensive players in the Pro Football Prospectus.

Unfortunately, there are no projections for individual defensive players for fantasy football purposes, if that's what you are looking for. Individual players are discussed in the unit comments that are at the end of each team chapter. We provide individual defensive player stats that are a little different from the standard ones -- plays rather than just tackles, as well as stops (plays stopped short of success), defeats (turnovers, plays stopped on third down before converting, plays stopped for a loss of yards), stop rate (stops per play made), plus good ol' interceptions, fumble recoveries, and sacks. We also give the average yards gained on each play where a defensive player made a play. We're hoping to have these stats on the website for 2005 and then really work on improving our understanding of them for next year's book.

Craig Matthews: Just got the Prospectus in the mail last night and it's become one of those books that you read, put down... and then see how long you can go before that nagging curiosity gets you to pick it up again. I mean, honey, the kids' dinner can wait-- I've got to find out if Clinton Portis's DVOA improved when the Redskins changed their blocking scheme last year. And I mean now!

In looking over the projections, however, I did see a few head scratchers I was hoping you can comment on.

1) Trent Green and Brett Favre ranked 15th and 20th at QB? I understand Green has a tough schedule and Favre is getting older but are you telling me if your life depended on it you would want David Carr (steady but unspectacular, as you note on your site) or Chad Pennington (recovering from a torn rotator cuff) airing it out for your fantasy football team? Ben Roethlisberger and Rex Grossman ranked ahead of Favre? Let's hope you don't run into Peter King in a dark alley (behind a Starbucks of course).

2) Where's the love for Steve Smith (ranked 36), Lee Evans (39) and Donald Driver (51)? Smith is just a year removed from an 88 catch, 1,000 yard season, Evans out produced Moulds in each of the final 7 games of 2004 and Driver had 1,200 yards and 9 TDs last year.

Both Green and Favre may look a little low because what we have here is a mean projection. Both Green and Favre have established a high standard of performance. But the older the quarterback, the higher the chance of injury or rapid decline; therefore, older quarterbacks come out with lower mean projections.

Trent Green's projection is in large part due to projected decline by the Kansas City offense as a whole, due to age and increased chance of injury, combined with Green's age and Kansas City's difficult schedule. On the other hand, I do think that the projection system underestimated how many attempts he'll have this year -- only 486 -- which would mean underestimating all his numbers. (Green has thrown at least 520 passes in three of his four KC seasons.) I wouldn't be surprised if he beats the projection, but I would be surprised if he had a year as good as last year.

Favre's projection had similar issues, with two differences: the projection for the Green Bay offense as a whole is even lower, but the Green Bay schedule is not projected to be difficult. Favre actually projected with the lowest number of games of any starting quarterback other than McNair, but I couldn't bring myself to actually predict Brett Favre would miss a game so I increased his numbers to reflect 16 games of the same quality.

Also remember that there is a difference between the question "which quarterback would you rather have" and "which quarterback would rather have in fantasy football." Go check out Houston's schedule.

On to the receivers in question. Steve Smith's projection simply reflects that he's coming off an injury. Driver is a 30-year-old receiver coming off a career year in an offense projected to decline. (Driver's projection doesn't take Walker's holdout into account, of course.)

I like Evans a lot more than Moulds, too. There is one major subjective component of these projections and that is "expected role," and Moulds ends up with a higher projection than Evans because we projected him to be "1" and Evans to be "2". Evans has fewer projected yards because he has fewer projected catches. That is how Buffalo seems to view them. However, change the expected roles, and you change the projections as so:

Moulds as 1, Evans as 2:
Moulds: 853 yards, 5.0 TD
Evans: 796 yards, 4.2 TD
Both as 1.5:
Moulds: 769 yards, 4.4 TD
Evans: 880 yards, 4.8 TD
Moulds as 2, Evans as 1:
Moulds: 684 yards, 3.8 TD
Evans: 965 yards, 5.4 TD

Yes, projected touchdowns are not whole numbers, reflecting the fact that they are mean projections and not absolute facts.

(Craig then responded with another interesting observation.)

Craig Matthews: It seems there is this rookie running back who struggled early on but then really hit his stride around Week 10 and ended up averaging over 100 yards per game over the last seven games, scoring touchdowns along the way as his team looked like it had finally found the franchise back to fit their system and lift their team out of mediocrity. Of course, most people will recognize this back as PFP-predicted 2005 rushing champ Kevin Jones. Fits the bill, but he's not who I am referring to. Julius Jones? Another good match, but no. To the horror of all of the fantasy football players planning on building their teams around one of the Joneses in the upcoming draft, the guy I am talking about is William Green.

In 2002, Cleveland drafted Green and critics hailed him as the back Butch Davis had been waiting for. After some early struggles he ran for 726 yards and 5 touchdowns in the last seven weeks and helped get the Browns to the playoffs. Of course, ever since life has been a horror show for Green and the Brownies.

Now, as someone who drafted Mr. Green in the second round in 2003 it is painful for me to even type his name. In 2005, I have been once again looking at the previous year's rookies to build my team around and the two Jones's seem to be the consensus picks to have breakout seasons. What I am looking for is for the Football Outsiders to convince me (please!) that Kevin Jones and Julius Jones are not William Green.

Best arguments I can make are:

1) The character issue. Unless Ricky Williams gets ahold of the Joneses, I think they have a better chance of staying on the straight and narrow.

2) The coaching staff. I'll take a couple of Super Bowl winners (even if they have horrible nicknames like Mooch and the Tuna) over Butcher Davis any day.

What do you say?

There is one number missing from your analysis, and that is carries. Here is an analysis of all three backs split into Weeks 1-10, and Weeks 11-17:

William Green 2002 Runs Yards Avg DVOA DPAR
Weeks 1-10 71 162 2.28 -53.2% -12.6
Weeks 11-17 172 726 4.22 1.3% 10.5
Kevin Jones 2004 Runs Yards Avg DVOA DPAR
Weeks 1-10 88 308 3.50 -10.9% 1.0
Weeks 11-17 153 830 5.42 9.4% 14.4
Julius Jones 2004 Runs Yards Avg DVOA DPAR
Weeks 1-10 5 16 3.20 -139.2% -2.5
Weeks 11-17 192 803 4.18 2.5% 13.8

Kevin Jones's second half explosion went far beyond those of Julius Jones and William Green, so I don't think Kevin Jones and Green are comparable. Julius, however, is comparable. The biggest differences there are:

1) Coaching Staff, as you noted

2) Ganja, as you noted

3) Julius Jones's record only includes the good games at the end of the year. Green had both the bad first half and good second half that could be used to judge him. Green went from bad to good in 2002. Both Joneses went from INJURED to good in 2004.

Also remember, Green wasn't great in yards per carry the following year, but his biggest problem was that he played only seven games, not that he didn't produce when he was playing.

Paul Tarr: I purchased Pro Football Prospectus 2005. My specific reason was to read your updated analysis of the greatest quarterback season ever.

I am a member of SABR and an avid football fan. In general, I was particularly impressed with your analysis at the play-by-play level. You are a step ahead of baseball in this regard. Need to study your techniques more.

I was generally very impressed with your updated analysis on the greatest football season ever:

  • You updated your analysis to include the whole 2004 season.
  • You spoke in more detail about your analytical procedures.
  • You included a detailed analysis of 100 players.
  • Even though you excluded Otto Graham you updated his passing projections in a 2004 environment.
  • You added modifications for rushing, sacks and the quality of the opponent schedule.

That being said I don't agree with your rationale for excluding Otto Graham from your ratings. 1953 was the 8th year in pro football for Graham under his coach, Paul Brown. It was also his 4th in the NFL. Brown's procedures and innovations were no longer new. The original article "It's Otto-Matic: Manning's No. 1" showed great analytical insight on your part in comparing Graham to Ruth. In 1920, Ruth hit 54 home runs. The second highest total in the major leagues was 19. Ruth's slugging percentage was .847 compared to .632 for the second highest in the major leagues. Talk about domination. Ruth, like Graham, played a different game, based on a livelier baseball. Yet Ruth was never excluded from comparisons over time. His dominance was recognized as a level to be achieved and superceded.

Under these circumstances I would love to find out what Graham's PAA was for 1953 based on your procedure. It would seem that you have done enough on Graham's 2004 projections to compute an estimate, even with some limitations. It hasn't been published in your book but I think the data is relevant.

I have to admit that I struggled mightily with trying to figure out where to make the cutoff for the Best Quarterback Seasons article. I always talk about Otto Graham when I am asked about great quarterbacks, and I did want to make sure that I mentioned his great 1953 season. Your reasons for including Graham are certainly valid. The main reason why I felt uncomfortable including him was the low passing touchdown total, so dramatically different from quarterbacks of the past 45 years. Even after adjusting for a 16-game season and the modern offensive environment, Graham would come out with just 16 touchdowns passing, and no matter how much I try to adjust one NFL era to fit another, that doesn't seem like the top quarterback based on what the term "quarterback" means today.

To answer your question, however, Graham works out to 186.3 Points Above Average. (The top quarterback of the past 45 seasons, according to the article in the book, is Bert Jones in 1976, with 163.8 PAA).

As long as we're mentioning Graham, however, we should also mention two other historical quarterbacks with astonishing performances in the early days of professional football. Sammy Baugh's 1945 season, projected to the 2004 offensive environment, is worth 228.6 PAA (passing data only). That includes a translated 92.5% completion percentage and a 27-to-4 TD/INT ratio. Sid Luckman's 1943 season, projected to the 2004 offensive environment, is worth an absurd 268.1 PAA (passing data only) including 6103 yards, or 11.7 yards per pass attempt. Plus, ya know, Sid's one of my peeps.

So, I guess the issue is where to draw the line. Depending on where you draw the line, the best quarterback season of all time belongs to Bert Jones, or Otto Graham, or Sid Luckman. All great quarterbacks.

Reinhard: The top running backs in DVOA last year were LaMont Jordan, Tatum Bell, Dorsey Levens, Larry Johnson, Derrick Blaylock, and T. J. Duckett. Doesn't that look strange, that none of the top 6 DVOA were considered starters really, just good backups? Does their reputation help their performance? Do they only get called on the field in favorable situations? (Teams against which there have been a lot of run plays, tiring out the starter. Or many runs in a row, tiring out the starter. Or when grinding out the clock? Many of the players saw whole games at a time at starter.)

A perfect example from last year would be LaDainian Tomlinson, -5.0% DVOA, compared to Jesse Chatman and Michael Turner, who put up about 18% DVOA between them. Any thoughts on what's going on here?

I think the reason for the phenomenon is pretty simple: sample size. The fewer attempts by a player, the wider the range of DVOA. Here are the five worst backs last year with at least 75 carries, according to DVOA: Quentin Griffin, Anthony Thomas, Maurice Hicks, William Green, and Eddie George. Green with 162 carries is the highest. The wide receivers are an even better example: The DVOA top five includes Brandon Stokely (102 passes), Lee Evans (75 passes), and Plaxico Burress (60 passes) while the bottom five includes five guys with 86 passes or fewer.

This is why I set a minimum for the DVOA/DPAR rankings. I set the minimum to try to include roughly three receivers per team and, for RB, to include both backs when a team has a committee. But the lower I set the minimum, the more of these very high and very low DVOA ratings will make it into the main table for each position. It's basically a subjective decision how many players to rank, and I have no problem with people who refer to our numbers making their own decisions about minimums. (For example, you could write: "According to FO, Curtis Martin has the highest DVOA of any running back with at least 200 carries.")

The other issues you brought up are hopefully already corrected in the DVOA formulas, except for "backup comes in after starter has tired down defense," that one isn't in there at this point. As for Tomlinson...

Tito: I have been doing Fantasy Football since I was 12 years old, having won two Blauvelt Bowls (II and VI). I am also an avid baseball statistician in my free time. Despite this excellent background, I do not understand why LaDainian Tomlinson's DPAR is as low as it is. Is it simply due to his sub-4.0 YPC average? His five fumbles? Enlighten me.

I assume you are talking about 2004, as Tomlinson was second in DPAR for 2003 and sixth for 2002. Tomlinson's rating by our metrics went down last year for the same reason that his conventional ratings (except for TDs) went down last year. He was hurt and didn't play up to his usual standard. It is pretty clear that he was better by mid-season. Check out Tomlinson's regular season numbers before and after San Diego's Week 10 bye:

LaDainian Tomlinson 2004 Yards/G Yd/Carry DPAR DVOA Suc. Rate
Weeks 1-9 77 3.8 -1.4 -15.4% 42%
Weeks 10-16 108 4.0 15.3 7.1% 49%

Yep, he's fine. There are a lot of places where our projections for next year disagree with conventional wisdom, but we have Tomlinson rated the #1 fantasy football player just like everyone else.

Finally, here are two emails I received that have nothing to do with the book. They are requests for information that we don't have, so I figured I would toss them out to the readership to see if anyone can help.

John Grenci: hello, I was wondering if you had the pre-season Las Vegas lines for seasonal wins in past seasons?

Norman Marty: Dear Aaron, we are trying to find a stat that nobody seems to know: How many times on average per game are the ten yard marker chains brought onto the field during a game, and how long does it take to bring them out?

If you know the answer to either of these questions, post it in the discussion. Thanks!

Posted by: Aaron Schatz on 03 Aug 2005

102 comments, Last at 25 Dec 2012, 5:05pm by Geschenke


by pcs (not verified) :: Wed, 08/03/2005 - 2:51pm

First! All I gotta say is:
"Dan Henning likes big backs, and he cannot lie" is the funniest thing I have read this year, anywhere.

by Goldbach (not verified) :: Wed, 08/03/2005 - 3:12pm

I have a question regarding the 2005 projections. For teams such as Philadelphia and Pittsburgh, who rested their regulars in the last game or two, are those games included in the projections for next year? What about for the teams they played against?

Also, I like the fact that you used percentages for the futrue projections, although I would probably have prefered that you broke it down for each win total, ie. Philadelphia has a 4.6% chance of going undefeated, a 5.4% chance of going 15-1, etc.

by geoff (not verified) :: Wed, 08/03/2005 - 3:22pm

Just to pile on to the "no projected records" discussion: it really is the most disappointing part of the book.

The reason I read this site and the book is not the stats, it's the fun that all of us have with the stats. Even if you stole a page from TMQ and said the projections were 0% guaranteed, even if you published the median projection and the low/high projection, whatever, the fact that a baseline is assigned is important, even if we all know it's totally pulled out of your collective statistical butts.

Other than that, I haven't had as much fun reading a sports book since the Bill James abstracts. Thanks.

by Kibbles (not verified) :: Wed, 08/03/2005 - 3:35pm

Aaron, if you really want to get the spreadsheets and data to the end user in a manipulatable form, and don't want to go the subscription route, have you considered putting all the spreadsheets on a CD and including it with the book next year?

by Reinhard (not verified) :: Wed, 08/03/2005 - 3:35pm

I don't have a problem with the projection format, really, just how it ends up breaking down. To use your Denver example:
0-4 : 9%
5-6 : 15%
7-8 : 23%
9-10: 24%
11+ : 28%
A lot of work is put into these numbers, but all it ends up telling me is that Denver has a good chance of improving or dropping slightly, and, like any team, a chance for collapse. A 3% or 4% difference between the range chances is too small to be meaningful to me, and once the % differences become larger, Im taking Denver to Leinart Land...

by Richie (not verified) :: Wed, 08/03/2005 - 3:41pm

Why would anybody want to know about the chains? That's funny.

by Richie (not verified) :: Wed, 08/03/2005 - 3:46pm

I like the projected wins "range" format. It's mostly guessing anyway. "Guessing" a team will win 9-10 games makes more sense than predicting exactly 10.

I think it also acknowledges the fact that NFL teams can swing quite a bit from year to year.

Chances are that at least one of Chicago, Miami, Cleveland, Washington, Tennessee and San Francisco will be much better in 2005 than anybody is predicting. If you simply gave record predictions for each team, it doesn't help. But instead, you can look at each of those teams chances of winning 9-10 games in 2005. Compare them and maybe get a better grasp as to which of those teams has a better shot at being a surprise team of 2005.

by ABW (not verified) :: Wed, 08/03/2005 - 3:49pm

Well, I haven't gotten my book yet(supposed to be here tomorrow fingers crossed), so I guess I really shouldn't talk, but I'm glad you didn't give exact projected records in the book, IMHO "exact projected" is something of an oxymoron. I would much rather see some numbers that actually mean something rather than a somewhat arbitrary assignment of a record to a team. I've been taught since junior high that EVERY number that is measured, calculated, or derived should have an error bar on it. I guess I think that if you are Peter King writing primarily for entertainment purposes, then predicting just a record is fine, but if you are going to do serious statistical analysis, then you should be giving out percentages when you predict things and standard deviations when you measure things, etc. Of course, I also read all of William Krasker, so I 'm probably not in the majority on this one...

by mistamaxwell (not verified) :: Wed, 08/03/2005 - 3:51pm

One thing I noticed in the individual QB blurbs...Byron Leftwich 'can't run a lick', while Trent Green is a deceptively good scrambler. Straight yds/carry for career are pretty much equal 4.1 to 4.0, and Leftwich runs slightly more frequently (13 passes to run vs.15 passes to run). Yet, the DVOA favors Green by a significant margin (something like 30 to 0 approximately). Is Leftwich coming up a yard short of First Downs continuously or what? I think this is a pretty good example of DVOA vs. conventional metrics.

by Kevin (not verified) :: Wed, 08/03/2005 - 4:28pm

I will continue the discussion . . . its pretty silly to buy a book that gives a thorough, detailed explanation as to why there aren't specific won loss records listed, and yet people STILL complain that they aren't in the book!! If you want specific won loss "guestimates" like the other books, then you should probably go buy one of those books!! The range of projections is what makes this book (among other countless things) unique!!!

by RowdyRoddyPiper (not verified) :: Wed, 08/03/2005 - 4:57pm

Okay, the dead horse, I beat her once again...I love the win range format. I am also more interested in the distribution than the mean in most cases. Variance is the spice of life. I know a lot of people don't think this way, but I also don't expect those people to be reading FO or the Football Prospectus.

The only thing that I think could use a change is more variations on Leinart Land and D'Brickshaw Derby. A team like, let's say the Jets, could be in Crosby Country next year...supposing Nugent doesn't convert on some key fieldgoals.

by Parker (not verified) :: Wed, 08/03/2005 - 5:22pm

If you're truly starved for specific record predictions, I'm happy to supply mine. As a teaser:

Cardinals 9-7

Pretty amazing stuff, huh? I got a whole lotta more where that came from, and it's all gold.

I'll be lurking, so if you have a specific team you need a predicition for, just list it here and I'll get to it as soon as I can.

by Goldbach (not verified) :: Wed, 08/03/2005 - 5:30pm

I think I'm predicting every team will go 8-8. I'll probably have more teams exact right than most publications.

by Aaron (not verified) :: Wed, 08/03/2005 - 5:30pm

Hey, no reason we can't do both the range and the mean projection specific record next year. Give the people what they want. And there will be mean projection records on the site the first week of September. Russell, by the way, gets props for coming up with "D'Brickashaw D'erby."

by Ray (not verified) :: Wed, 08/03/2005 - 5:33pm

I think next year's book should have BOTH the range and specific projection for record. I like to see the range because it does reinforce that anything can happen in this game.

But a hard number would be nice to have as well just to see where the average lies. It's too hard (for me) to get a feel for where a team will probably end up with the broad percentages.

by Ray (not verified) :: Wed, 08/03/2005 - 5:38pm

Sorry for the double post. Above is what happens when you write a post, leave for a while, then just come back and hit submit. Someone else makes your comment for you (thanks a lot, Aaron!).

by Jimmy Two Times (not verified) :: Wed, 08/03/2005 - 6:42pm

What Ron Shandler does is password-protect his spreadsheet and use a word from the book as the password. For example, the password might be the last name of the first player listed on page 385 of the book. That gets you closer to actually getting people to buy the book, though of course people can still cheat around it.

Aaron-- Are you going to be updating the spreadsheet as we get deeper into camp and we get a clearer sense of what people's roles will be? I suspect that any spreadsheet you put together now will be next to useless by September 1, which is when most fantasy drafts are likely to be.

by Richie (not verified) :: Wed, 08/03/2005 - 7:39pm

Is the formula for Best QB Season explained anyplace? I couldn't find it in the book.

by Zac (not verified) :: Wed, 08/03/2005 - 9:51pm

Got a couple.
1. Travis Henry has been traded, so spill the beans on Shaud Williams. Is he really that good? What do you expect out of him?

2. The one thing I was able to see from your record %s is that you guys really expect Philadelphia to go well. 80% that they win at least 11 games?

3. On a related note, you guys are expecting the two conferences to be basically like they were last year, huh? AFC to be very even (the mean projects 8 AFC teams to have at least 9 wins), and the NFC to be more sharply divided (Philly, Seattle, and Tampa each have projected means above 11 wins)? And the NFC North. If my calculations are right, then there's a 25% chance that 8-8 could be good enough to win the division?

4. I caught your little joke. I'm talking about Ricky Williams team last year (in the Running Backs section).

by Mike (not verified) :: Wed, 08/03/2005 - 11:14pm

I realize you have to advertise, but do you have any idea how annoying that flashing neon ad is? Please speak with a reputable web design firm, I didn't think I was headed to a porn site.

by Ron Mexico (not verified) :: Wed, 08/03/2005 - 11:54pm

I realize you have to advertise, but do you have any idea how annoying that flashing neon ad is? Please speak with a reputable web design firm, I didn’t think I was headed to a porn site.

What I want to know is- what happened to the Dusty advertisements? I mean, if you're going to have advertisements to support the site, I'd much prefer the Dusty ones than annoying flashing ones saying that I can get FREE* GAS!!!!!!

Note: I didn't forget the asterisk, which is to tell you that the gas is not, in fact, free, but costs 8 trillion dollars, your beer, your soul, your dog and you have to watch every Lions, Cardinals, Bengals, and Saints game from the last 10 years, in a row with no breaks.

by Mukaikubo (not verified) :: Thu, 08/04/2005 - 12:23am

Just got the book from Amazon today, I ate up the Win Range things. In fact, I put them in Excel and tried to massage the numbers into rough median win counts. I'll be refining that probably until y'all put out your real projected wins, but I have a few comments.

1. Atlanta is going to fall off, but do you guys really believe they're the second worst team in the NFL? Only SF has a higher chance of winning 4 or less games according to the ratings! If someone gave me, right now, even odds bet that said "Atlanta will win 5 or more games this year", I would take it without even thinking twice.

2. Are injuries taken into account in this analysis?

3. I think some bad teams have too-high high end ratings. I love my Titans, but I'm extremely dubious of an analysis which gives them a 40% chance of being a playoff contender.

4. I think some good teams have too-low low end ratings. In what 6% of the time does Indy fail to win even 5 games? Is that the "Peyton Manning drives a minivan containing Edgerrin James and Reggie Wayne off a cliff" scenario?

Those are my specific complaints.

What I did to find median wins was multiply the 11+ range by 12.5, 9-10 by 9.5, 7-8 by 7.5, 5-6 by 5.5, and 4- by 2.5 and add them together for median wins. Every team in the AFC falls between 6.12 (Miami) and 9.79 (Pitt). The NFC has considerably more variance- The Eagles, Seahawks and Bucs (!?!?) all above 10, with the Falcons and 49ers under 6. (The 49ers almost at 5- harsh but sadly probably true) I'm working on 20%, 40%, 60% and 80% percentile win totals for the stats y'all gave (ex: Philly will win *11* or more 80% of the time, and the Jets will win *7* or more 80% of the time) but it's slow, tedious work.

by Glenn (not verified) :: Thu, 08/04/2005 - 2:02am

Cheap bastard that I am, today I went to the Barnes & Noble in Holyoke, Mass, took the Prospectus off the shelf, pulled up an overstuffed comfy chair, then whipped out my digital recorder and whispered into it a whole boatload of nifty fantasy football nuggets for my draft this weekend. Is that so wrong? Is that so wrong?

by James, London (not verified) :: Thu, 08/04/2005 - 6:08am

IT'S HERE!!!!! My prospectus arrived this morning. Unfortunately work prevents me from reading it immediately, but "DVOA gets Crunk, Vol 1" is an instant classic.

by andrew (not verified) :: Thu, 08/04/2005 - 7:41am

Randy Moss's turf game ...

he would need more than 300... 314 or so... to set the record and beat Flipper Anderson....

by Zac (not verified) :: Thu, 08/04/2005 - 8:18am


So I was doing the right thing when I tried to figure out median wins. One question though. Doesn't the 11-16 range have a middle value of 13.5, not 12.5?

I did that, and I also did it another way, and while they are close, they don't totally reconcile. For instance, SF has 4.83 median wins, according to my calculations. But 54% of the time, they win at least 5 games. So the median would have to be above 5, wouldn't it?

by Arkaein (not verified) :: Thu, 08/04/2005 - 9:29am

(note to everyone: sorry if this is excessively long.)

Finally! I've been waiting for this chance to post my beliefs on why I believe that The Rumors of the Packers Demise Are Greatly Exaggerated (I was waiting for the NFC North Four Downs, but this will do, as I just got my PFP copy).

Most people on the board (as well as many other places) seem to believe that a decline in Green Bay's record is nearly a foregone conclusion this year. I decided to try my own method of analysis and break down the differences between the circumstances of the 2004 and 2005 Green Bay teams to see if I could make a strong case for either direction.

My method is as follows. I examine the differences between 2004 and 2005 broken down by category. Personnel is the biggest category, but coaching and schedule are included as well. I estimate which factors improve or worsen the Packers' situation and by how many games the win-loss total is likely to change. This is fairly subjective, and so I'm inviting discussion on these factors. Please try to make an argument for any factors that you think should be changed, if you do make such suggestions.

Here's my breakdown, by offense, defense, special teams and schedule. Each category is broken into components, and some components are broken into parenthesized sub-components. Brief explanations are given for all modified attributes (sorry if the formatting doesn't work out very well):

-0.5 Offensive Line
(-0.75 loss of Wahle and Rivera)
(+0.25 return of Flanagan and improved shape of Barry)
+0.25 Wide Receivers and Tight Ends
(+0.25 improvement in health and vision of Ferguson)
+0.25 Running Backs
(+0.25 improved health of Green and Davenport)
+0.0 Quarterbacks
+0.0 Coaching

+0.0 Defensive Line
+0.25 Linebackers
(+0.25 improvement of Barnett, possible slight upgrade over Navies)
+0.25 Defensive Backs
(-0.25 loss of Sharper's playmaking ability)
(+0.25 improvement at CB by Thomas passing Carroll)
(+0.25 improved consistency of coverage/tackling at both safeties)
+0.75 Coaching
+0.25 Improved number of turnovers (regression towards mean)

+0.25 Kick Return
(+0.25 improved health of Davenport and Ferguson)
+0.0 Punt Return
+0.0 Kick and Punt Coverage
+0.0 Kickoffs
+0.0 Punts

(-0.5 x 3 for improvements of Lions, Tigers^H^H^H^H^H Vikings, and Bears)

TOTAL = 0.0 + 1.5 + 0.25 - 1.5 = +0.25

Based on my breakdown I predict the Packers to repeat their 10-6 record and return to the playoffs. To elaborate on some of the points:
* Kevin Barry, the extra tackle in the Packers' heavy U-71 formation was out of shape last year, but has lost weight, is reportedly quicker, and could become a starter at right guard or at tackle (in this case Mark Tauscher would move inside to guard)
* Robert Ferguson had his eyes Lasiked in the off-season, and from the sounds of it he had pretty awful vision in at least one eye before surgery but never wore contacts
* Ahman Green was nicked up last year following the large number of carries in 2003, and Najeh Davenport also battled injuries but has been great when healthy. With fewer carries last year and not yet being too old I predict a healthier year for Green
* Nick Barnett was awful last year despite looking pretty good as a rookie. I chalk this up to a horrible offensive scheme (see more below); he'll be in better position on his assignments this year
* Sharper could have been a big asset this year, however his relatively poor play much of last year due to injury means that there won't be too much drop off from then to this season
* Joey Thomas, the other CB the Pack drafted last year, has reportedly improved and likely surpassed Ahmad Carroll at CB, improving the position overall
* The Packers won't have stellar safeties, but the veterans they've brought in (Arturo Freeman and Earl Little) should be more steady than Sharper and Mark Roman were last year
* Yes, Jim Bates is worth nearly a win all by himself. Just look at how much worse the 2004 GB defense was than the average 2003 version with largely the same personnel (except for Mike McKenzie, admittedly a large factor)
* The Packers D forced very few turnovers last year (just 15 total, worst in the league), expect this to improve through sheer chance if for no other reason
* Davenport and Ferguson were both very good kickoff returners in 2003, however in 2004 both were rarely healthy and they got few if any chances to work together
* The division is getting tougher, making this the biggest factor against the Packers

Also, PFP predicted declines in the numbers of Favre, Walker, Driver, and Green, with no improvement in Ferguson or Devanport. This does not make sense. Driver is 30 but healthy (and PFP has his numbers dropping in half!), Walker may not match last year but should be better than 1000 yards easily (unless he splits time or is injured, which should boots Ferguson's numbers). Favre has improved in DVOA and DPAR each of the last two years, and less than 40 sacks in the pasxt two years means he's as healthy as ever. Besides, there's tremendous variation is how QBS age, much more than say RBs.

In a nutshell, the offense is treading water while the defense improving to even average would be a huge improvement. I think few people realize how bad last year's defensive scheme and coaching were. GB has about average talent, personnel did not change much from 2003 but performance was awful. Jim Bates should fix this and make the Pack at least 15th-20th in league defense. This will pretty much cancel out the improvements in the rest of the division.

Two of the biggest factors sited for the Packer imminent decline are the decline of #4 and the loss of Wahle and Rivera. I say Favre may be slowing a bit, but not as much as most people like to think. He threw for 62 TDs over the past two seasons and over 4000 yards last year. He makes some boneheaded decisions but this is no different from the rest of his career. If Ferguson is healthy and improves as much as I think he can and if Walker actually plays this year (admittedly looking like a big if at this point) then Favre will have both one of the best WR corps in the league as well as one of the best RB corps.

Wahle and Rivera are big losses, but I don't think their as big as a lot of people think. For one, they;re both guards who played inside two very good tackles. The Packers also get Mike Flanagan back this year, and while Rugamer filled in pretty well Flanagan is still a step up. Kevin Barry's improved conditioning will help whether he starts or only plays in the U-71 formation leading the Power O running play.

I also have reason to believe that the FootballOutsiders DVOA calculations may not be as good a predictor for future Packer success as it is for other teams in 2005. I invite you to look at the Team Efficiency charts for 2004. Two columns are of import, the Rank associated with Estimated Wins (based on DVOA), and the Rank associated with Pythgorean Wins (based purely on point scored for and against). These two ranks are usually quite close and are good predictors of actual record for the current season. Only three teams had a difference between these ranks larger than four: Detroit, Dallas, and you guessed it, Green Bay. Detroit and Dallas had discrepancies of 6. Green Bay had a whopping discrepancy of 10.

I can't give you a complete explanation of this difference. What I can say is that I believe that Green Bay's DVOA stats are an outlier, and are likely a poor fit for an otherwise excellent model. My conclusion is that DVOA is not a reliable system for prediction the success of the Packers in 2005.

That's all I've got right now, I'd really like to hear some feedback.

by Dan Babbitt (not verified) :: Thu, 08/04/2005 - 10:19am

Hi Aaron,

I have a question about your projections for Jamal Lewis.

Can you explain why the system projects that Lewis will average "only" 16 carries per game?

- Dan

by Tarrant (not verified) :: Thu, 08/04/2005 - 11:36am

I agree with the use of 12.5 rather than 13.5 as the "11+" number, though in using it to project wins using that formula.

I think if we take every team that manages a record of 11-5 or better, that the average number of wins probably isn't between 13 and 14 (many seasons don't even have a 14-game winner, after all), it's likely closer to between 12 and 13 than it is between 13 and 14.


by Mukaikubo (not verified) :: Thu, 08/04/2005 - 12:03pm

Tarrant's got my reasoning right. That's also why I averaged the 0-4 column to 2.5, instead of 2.

I took it a step further this morning and assumed there'd be twice as many 11 wins as 12 wins, twice as many 12 as 13, etc. etc. Same for 0-4. Made the (very wrong, it turned out, but I haven't changed it yet) that there were twice as many 9 wins as 10 wins, 8 wins as 7, 6 as 5 too to try to get a rough estimate of probabilities for each win number (this would be something I'd dearly love to see next year!) Work is still ongoiong, and I think I'd have to do curve fits to get it 'right', which would be a royal pain...

by Jason Powers (not verified) :: Thu, 08/04/2005 - 12:31pm

One of the biggest complaints we’ve gotten is that the book doesn’t give a specific prediction of each team’s won-loss record.

I saw you mention this at the front of the book, too. I don't think there's reason to get self-conscious about it: just post a FP Book FAQ, answer the question, direct anyone who asks to the FAQ, and move on.

You made the right decision. There are plenty of books out there designed to help people gamble. The way you post the likely record projections tells people more about the team than saying 8-8 or whatever. I did note, however, that the team box data at the front of each section has last year's record, a comment on last year's performace and a comment on this year's likely performance. Putting last year's record and something along the lines of "Mean of likely 2005 record" right under it would not be a crime against science, as long as you'd explained that the table (which you should retain) is the projection, not the mean or mode of the numbers that made the table.

In fact, I wouldn't mind seeing an expanded table showing the % of each potential record - the curve would show a nice illustration of how difficult it is to project wins in an absolute way.

"Baseball Prospectus... (offers projections in a) downloadable spreadsheet as part of a premium subscription package"

Beyond the obvious question of why you two don't just merge hosting and subscriptions (then you can pick what's public and not about your site, I don't see why you couldn't keep the stuff on the same webserver, you're both using LAMP sites anyway), I wanted to mention BP's reports are mostly public, and up until recently, they could keep the hosting costs down by posting those as text files, which for the most part are much smaller to store and transmit than binary spreadsheets or even web pages.

It looks like BP has recently changed from pure text to php formatted data, though I figure they could have saved some transfer by leaving the public data as pure text and having the php augmented stuff for subscribers.

As a group of rational science types, both sites could be doing more to optimize their own process, reducing costs and increasing the results for the customers...oh wait, that's what you ask the teams to do, isn't it?

by Nelphonious of Pennefielde (not verified) :: Thu, 08/04/2005 - 1:01pm

Baseball actually hires personnel 'so called' Sabremetricians.
What of FO cutting edge thinktankers self monikering their skills==GRIDOCALCS== OR..............!?

by Mukaikubo (not verified) :: Thu, 08/04/2005 - 2:59pm

Well, unless you believe the Steelers have a 13.4% chance of going undefeated this year, trying to put curvefits on the data's a real blind alley.

With a data set like the Steelers, where every category is higher and higher, you can't do a curve fit that will do anything but insist that a 16 game season is the most likely outcome. I'm thinking of giving my poor, abused spreadsheet 'boundary conditions' and arbitrarily saying that 16 or 0 wins is impossible.

So we're not actually going to know the percent chance of each WL record unless the Authors take pity on we poor sheep.

by Tarrant (not verified) :: Thu, 08/04/2005 - 4:05pm

At least for reference, over the past 10 years (not a huge sample size, but not tiny either), here's the numbers for a given number of wins:

0 wins: 0 (0%)
1 win: 3 (~1%)
2 wins: 4 (~1.3%)
3 wins: 11 (~3.6%)
4 wins: 22 (~7.2%)
5 wins: 26 (~8.5%)
6 wins: 32 (~10.4%)
7 wins: 36 (~11.7%)
8 wins: 38 (~12.4%)
9 wins: 37 (~12.1%)
10 wins: 35 (~11.4%)
11 wins: 21 (~6.8%)
12 wins: 20 (~6.5%)
13 wins: 14 (~4.6%)
14 wins: 6 (~2.0%)
15 wins: 2 (~0.7%)
16 wins: 0 (0%)

There's a nice parallel, actually between 0 and 16 (duh), 1 and 15 wins, 2 and 14 wins, etc. While not exact, it's pretty neat. It does, however, show that 11 wins alone happens almost as much as 13-16 wins combined.

Note that the numbers here don't add up to the total number of games played due to ties, which I threw out.


by Tarrant (not verified) :: Thu, 08/04/2005 - 4:12pm

The averages for wins in each of the categories over the past ten years:

Leinart land (0-4): ~3.3
Bad team (5-6): ~5.55
Mediocre (7-8): ~7.5
Playoffs (9-10): ~9.5
Super Bowl? (11+): ~12.4

So your math isn't too bad, Mukaikubo.


by Trogdor (not verified) :: Thu, 08/04/2005 - 5:02pm

Excellent book. And for the record, I find these projections significantly more interesting than a simple W/L record. To me, it's better to see that a team has a strong chance of the top 2 categories and much less of collapsing, or that it has roughly 20% in all five, or whatever.

Anyway, I had a question about adjusted offense and defense. In the intro it says they are adjusted to a baseline of 23.5 ppg, but for individual teams this doesn't seem to hold. For example, Cincy's AO is 23.5, but their DVOA is +7.6%, indicating they should be above baseline, maybe 25 or 26 or so. Their defense has a positive 1.1 DVOA, yet their AD is 21.8, which is below baseline. There are other examples of this throughout the book.

Maybe I'm not understanding it properly, but it looks to me like the actual baseline used for the team calculations was 21.5 ppg, not 23.5 (Carolina and Houston both have AOs of 21.5 and essentially 0 DVOA). Is it just a typo in the introduction, or am I just misunderstanding it?

by Mukaikubo (not verified) :: Thu, 08/04/2005 - 5:36pm

I've also started work on a rough simplistic method to calculate how many "games" a player's worth. Take the DPAR, subtract by his replacement's DPAR if you can find it (0 if not) and divide by 16 games. Then see how many games margin of victory are that low. It's not intended to be an actual measure of how valuable a player is to the team (Peyton Manning is worth 5 'games', and Brady "only" 3), but an easier way to visualize DPAR for key players.

by Arkaein (not verified) :: Thu, 08/04/2005 - 5:48pm

Hey, I'm not trying to be too annoying here, but has anyone read my previous rant (#27) and have any opinions to share on it. Because I've been reading some more of the Prospectus, and frankly the discussion of the Packers has been driving me nuts. Aaron, I'd really appreciate it you could explain a few things that I just don't understand:

1) Why does PFP claim that Favre is in decline? I understand that the projection model takes age into consideration. However, in places like the OPE chapter the claim is made that Ahman Green might struggle because Favre is in decline (not will likely decline). However, Favre's DPAR and DVOA have both improved significantly last year after holding pretty steady from 2002-2003. Favre is also one of the least sacked QBs around so it's not likely his body will break down all of the sudden this season. So what gives?

2) Shouldn't Ahman Green have a chance for a rebound? Yes, his blocking will be a bit worse (though I don't quite believe that losing Wahle and Rivera together is the end of the world, as PFP would seem to indicate in multiple places). However last year he wore down after 355 carries in 2003. Last year he only had 259 carries. Davenport will take some of his carries, but as Davenport isn't projected to gain more yards shouldn't Green have a chance to rebound? He's only 28, a couple of years from the RB wall, and he didn't take much pounding on the bench his first 2 NFL seasons on the Seahawks bench.

by Zac (not verified) :: Thu, 08/04/2005 - 6:46pm

DRAT! I got home and was all set to listen to Aaron on ESPN 1510 AM out of Milwaukee, when it turns out I can't get it in. Why? Because I live in Oshkosh, where WOSH is broadcasting at 1490 AM. They block out anything for 20 or 30 in either direction.

by Soulless Merchant of Fear (not verified) :: Thu, 08/04/2005 - 7:28pm

Re: #38

I can't speak for the guys who crunched the numbers, but I've mentally penciled in Favre for a not-insignificant decline this season due to the loss of offensive linemen. Favre himself might be as sharp as ever, but he's gonna be hurried and smacked around a lot more this year than he's used to.

I love the guy and root for the Packers, so man, I hope I'm wrong.

by Sean (not verified) :: Thu, 08/04/2005 - 8:21pm

38 -

1. Favre is in decline because old people do not have to be hit to decline. It's just what happens. This is a very common sports-projection assumption. Like many such assumptions, it is based on the observed history of many players, and like many assumptions, it will be wrong sometimes. Perhaps Favre is the guy who will continue to perform at a high level at age 43, but it sure isn't likely.

2 - If you search around, you will see a lot of research that shows that backs who are badly overused by their teams do not tend to recover particularly well. Plus, I don't know if you have ever played football, but there is nothing "only" about taking 259 hits in a season from extremely fast, extremely strong players intent on separating your head (and the ball) from your body. And, completely subjectively, Green does take a lot of hard hits. If you want to think that Green will be strong/healthy/KurtWarnerized enough to crack 1800 combined yards, you are entitled to believe it. But I would assume (there it is again) that most non-Packer-fans attempting to perform logical, realistic projections will have Green at, to you, sub-acceptable performance levels. You ask "Shouldn’t Ahman Green have a chance for a rebound? " The answer is that he does get a chance to rebound. He's still playing in the NFL, right? Does FO/FP think that it is likely? Evidently not, but Green is in no way, shape or form dependent on the opinions of FO/FP to keep the kids in Cheerios, and I'm pretty sure that any money FO/FP are making on Green is very, very indirect.

by Sean (not verified) :: Thu, 08/04/2005 - 9:40pm

First off, great book- I ended up dropping $70 this year between Scientific Football and Pro Football Prospectus instead of my usual $20-30 on a combination of preseason mags, fantasy football previews and scouting reports, and the money was very well spent. Both will be mandatory reference materials throughout the season (along with the FO website, of course). Also, I'll weigh in and say that I like the win projection system used here.

Some additional comments and/or constructive criticism:

-I would have liked to have seen compilation charts included for all of the sections covered, so that the offensive line, defensive front seven, secondary statistics, individual DPAR and DVOA information could be accessed all in one place.

-The essay in the San Diego section on the domino effect that an increase in third down efficiency can have on overall team performance was really interesting. I do think that to a certain extent, the conclusion that teams that have higher DVOAs on first and second downs should see their third down DVOA inevitably regress to the mean needs to be taken with a grain of salt. More specifically, it needs to take personnel issues into greater account. For instance, the Texans defense was a bit better than average on first and second downs (-3.1% and -8.1%, respectively), but then was terrible on third down, to the tune of 9.8%. Part of that may be flukely, but it seems to me that it largely highlights the fact that the Texans had a terrible pass rush. They were able to put teams into passing situations, but their personnel deficiencies meant that they were unable to take advantage of that. Without the team specifically addressing the personnel problem, how much can reasonably be expected to improve simply through regression to the mean?

-The individual defensive statistics are a good start, but I think they still have a ways to go. What's most noticeable when cross-referencing between Scientific Football and the PFP is that Joyner accounts not only for plays that a defensive player was involved in, but also plays that they were targeted to be involved in (basically, the equivalent of judging DB performance by yards per attempt rather than yards per completion). It's possible that there is simply no statistical substitution for breaking down game tape if you want to see if particular defenders are being targeted. It also seems to me that there are some disparities between your own numbers and the comments made. For instance, when breaking down the Denver secondary, your notes suggest that Champ Bailey had a good year but was negated by the fact that teams simply threw at other players and left him alone. You note that he was assigned to cover top tight ends like Tony Gonzalez and Antonio Gates, which is good to know, but presumably he spent most of the year matched up on #1 wideouts. And when you get to Denver's stats, two things pop out- 1) that the DVOA vs. #1 WRs is a not very good 7.4% and 2) that Bailey was involved in 94 plays, the same number as Kelley Herndon on the other side, and that Bailey had a much lower stop rate percentage and a much high average yards than Herndon. I bring that up because Joyner does an extended breakdown of Bailey's season and concludes that he pretty much got toasted all year long and that he graded out as the most overrated defensive player in the NFL (which, from my own observations, I would be inclined to agree with).

by Arkaein (not verified) :: Fri, 08/05/2005 - 12:19am

Sean, thanks for your thoughts. About Ahman Green, I find it interesting that PFP does not predict a similar decline for backs like Shaun Alexander (also age 28, 974 carries over the past 3 seasons, increasing each year), Edgerrin James (age 27, but about 300 more career carries than Green and 920 over the past 3 years with a serious knee injury in his past). Compare to Green, who has had 900 carries over the past 3 years. While nicked up about every other year, he has never had a serious injury.

Jamal Lewis is not expected to rebound either, this is a bit puzzling to me considering he's still young and only had one year of overuse. I will agree that both Lewis and Green take some hard hits, but I also doubt this was accounted for in the projections (if this is wrong someone let me know). Fred Taylor is also expected to start declining at age 29, but Priest Holmes is projected to recover to 1200 yards at age 32 and just a few more career carries than Green and a greater history of injury.

My prediction: 1200-1300 yards with 9 TDs for Green behind a still above average O-line.

As for Favre, I can't really see a reasonable career comparison that says Favre will drop off. Many QBs have been productive late into their 30s (some like Flutie and Warren Moon to 40+), Favre is healthy and has actually improved slightly over the past few years. I just don't think the standard aging projections have much bearing on Favre, at least on a single year basis.

My projection: 31 TDs, 16 INTs, 63% completions, 3700 yards, sacked 24 times.

by Sean (not verified) :: Fri, 08/05/2005 - 12:33am

Arkaein- I think the difference in projections for Alexander and Green are easily justified on several counts:

1) According to DVOA and DPAR, Alexander was much better than Green last year.

2) Alexander runs behind arguably the best line in football, certainly the best left side, while Green's line is going to be impacted by the loss of the two guards in all liklihood.

3) Alexander looks to have an easier schedule, littered with poor run defending teams, while the Packers are going to have a tougher time of it, particularly inside the division, where the Lions, Bears and Vikings all figure to field pretty good run defense units.

4) Green has one of the better backups in the league available to take carries from him.

Seems pretty straightforward to me.

by Arkaein (not verified) :: Fri, 08/05/2005 - 1:27am

Good points Sean, though I do have a few minor quibbles. Alexander was better than Green last year, but that doesn't explain to me why Green whould decline another 100 yards and 0.3 YPC. The O-line is probably a bigger factor, and I hadn't accounted for run defenses faced.

As far as Davenport, it would seem logical that Davenport could steal a few carries from Green, although PFP also predicts Davenport's numbers to decline, and sharply (100 less yards and 1.2 fewer YPC!)

We'll just have to wait and see I guess. I'll be hoping that I'm right in predicting that Flanagan's return and improved conditioning in extra tackle Kevin Barry minimizes the loss of the two guards.

by Zac (not verified) :: Fri, 08/05/2005 - 8:22am

I just remembered my favorite anecdote from the book. From the RBs section, under Shaun Alexander, when it mentions Barry Sanders in the final game of his rookie season. I found an old SI that covers it. Click my name to view it.

As you say, Shaun Alexander could learn a lot from Barry Sanders.

by Aaron (not verified) :: Fri, 08/05/2005 - 11:19am

Sean wrote: "It's possible that there is simply no statistical substitution for breaking down game tape if you want to see if particular defenders are being targeted."

Sean, it's not just possible, it's definitely a fact. It's an unfortunate fact that we have always acknowledged as our weakness -- it is mentioned in the introduction to the book, and has been mentioned on the website numerous times. Which is why you'll be interested in the note on the top of Extra Points that says "Game Charters Needed." We're going to try to take that next step. Please drop us a line and take part.

by Nelphonious of Pennefield (not verified) :: Fri, 08/05/2005 - 12:46pm

Aaron/salute your constant honing and reflecting of system models effectiveness to get closer to "IT."
One other NFL factor difficult to quantify are the semi secret schisms between owner/ GM/head coach(or two headed Shanahan+ Walrus syndrome) that undermine roster assembly and 16 game use.So I pick a six pack of 2005 teams most subject to this UNDERFORMING grease;PACKERS,BRONCOS,GIANTS,SEAHAWKS,RAMS,COWBOYS.

by Aaron (not verified) :: Fri, 08/05/2005 - 1:08pm

Just a note -- I don't want Sean to think my comment there (#47) is meant to be negative, rather it is meant to be constructive, towards pointing Sean and other people to the Game Charting project. By the way, Sean's comment (#41) on the difference between scientific projection formulas and subjective predictions is right on the money. I couldn't have said it better myself. Once we put in the one subjective variable (expected role), KUBIAK spit out a projection, and I didn't change it, even if emotionally I disagreed.

by Richie (not verified) :: Fri, 08/05/2005 - 2:48pm

1. Favre is in decline because old people do not have to be hit to decline. It’s just what happens. This is a very common sports-projection assumption. Like many such assumptions, it is based on the observed history of many players, and like many assumptions, it will be wrong sometimes. Perhaps Favre is the guy who will continue to perform at a high level at age 43, but it sure isn’t likely.

I think Favre could be fairly compared to Dan Marino. Marino started declining in his 14th season. It was a slow decline. It was also the first year of Jimmy Johnson, which may have had something to do with it. Favre is entering his 15th season (14th as a starter). I would expect a slight dropoff from Favre this year, but not much.

by Tarrant (not verified) :: Fri, 08/05/2005 - 3:11pm

And for those of us happily waiting not only for the NFL, but for the start of the college season, the first Coaches' poll for the 2005 season came out today...I'm gearing up for 7th Day Adventure and the half-college-related Confessions discussions...


by Sean (not verified) :: Fri, 08/05/2005 - 3:14pm


I appreciate the link. (I'd love to take credit for the comments in 41, but alas, that was a different Sean.)

by Led (not verified) :: Fri, 08/05/2005 - 3:15pm

Question about the Dolphins' secondary and Surtain. According to the PFP, Miami was 9th in DVOA against #1 receivers but 29th in DVOA against #2 receivers (and almost twice as bad as the team ranked 28th!). That suggests that the #2 CB played pretty darn poorly. But the PFP also says that Madison put in a good year, so can we conclude that the CB pooping the bed against #2 receivers was Pat Surtain? Not a good sign for the Chefs. On the flip side, if Madison was covering the #2's that's not a good sign for the Phins. Inquiring minds want to know who the stinky CB in Miami was.

by John (not verified) :: Fri, 08/05/2005 - 4:40pm

I'd like to see each team's schedule in the book next year.

by John (not verified) :: Fri, 08/05/2005 - 5:00pm

I just caught another error.
This one's a little too annoying. I was reading the Larry Croom comment. John Navarre did not play against the Jets. Shaun King started that game and was replaced by Josh McCown. So, of course, Herman Edwards "neglected to mention any of this in his Christmas card to Denny Green."
It's easy for me to spot the McCardell, R. Moss, and Croom type errors, I'm wondering if you fellows have spotted any statistical errors.
The last line in the Jerome Mathis write-up is quite good.

Any chance at getting some photos into the book next year?

by Hook (not verified) :: Fri, 08/05/2005 - 5:25pm

Not directly related to this, but in traditional stats does anyone know if a sack or interception counts as a pass attempt?

by Vince (not verified) :: Fri, 08/05/2005 - 5:35pm

An interception is absolutely counted as a pass attempt. Sacks are absolutely not counted as pass attempts.

At least, that's in NFL stats. I don't know about FO stats.

by Aaron (not verified) :: Fri, 08/05/2005 - 6:37pm

Let me get this out of the way, since I know a lot of people have asked in emails -- and will really wonder after today's Pinkston injury. I won't be sending out a new projections spreadsheet to those who donated, but I will figure out a way, through mailbag articles or otherwise, to make new projections available on the site for players with major injuries. Not every little possible depth chart change, but the big stuff. Since they will be free, everyone will get them, even those who bought the book but didn't pay for the spreadsheet, but since I will only be giving players who change, it won't be useful to those who don't buy the book.

by pm (not verified) :: Sat, 08/06/2005 - 9:26am

I have a question about the book:

Is the book available in a regular bookstore like Borders or Barnes and Nobles

by Aaron (not verified) :: Sat, 08/06/2005 - 11:02am

Absolutely, it should be there now and if it isn't, ask the manager why not! Most of my bookstore appearances are actually at Borders.

by Ryan (not verified) :: Sat, 08/06/2005 - 1:59pm

I bought mine at Borders because I couldn't wait for shipping.

by glr (not verified) :: Sat, 08/06/2005 - 9:24pm

re: favre

John Elway had arguably his best seasons after 35.

re: champ bailey

I watched every Denver game last year, and Bailey was burnt only a handful of times, and almost always it was by the other team's number one. A couple of these were 'flukes.' One of the times that Chad Johnson beat him, he was blatantly interfered with. When Michael Clayton beat him he slipped. Look, big name WR's are going to make big plays sometimes. He only had 3 INT's, yes, but two of those were amazing. He also made a ton of big tackles (he rarely misses tackles). I think his INT totals were definitely hurt because teams avoided him.

It isn't fair to look at his 'sub-par' numbers, see a couple of big plays against him on the highlights, and assume he had a bad year. He absolutely didn't.

by Sean (not verified) :: Sat, 08/06/2005 - 9:56pm

Here are the numbers that KC Joyner has on Bailey in Scientific Football- note that they don't include the Jacksonville game, when Jimmy Smith went off.

Bailey got thrown at 90 times. For comparison's sake, Kelly Herndon got thrown at 94 times, so teams were absolutely not avoiding Bailey; in fact, they were targeting him, especially on deep balls. Teams went deep on Bailey 24 times, completing 50% of those passes for 425 yards and 4 TDs. Only one cornerback in the league gave up more yardage on the deep ball- Lito Sheppard, and the 4 deep touchdowns was among the worst in the league- only Nick Harper, Jason David and Al Harris gave up more. Bailey's total completion percentage against was 63.3%, 67th best in the league. He gave up 778 yards and 8.6 yards per attempt, 65th best in the league. He gave up 7 TDs in total- only Terence Newman, Jason David, Brian Williams, Marcus Trufant, Al Harris and Will Allen gave up more.

Yes, Bailey was usually matched up against the other team's number one receivers, but his numbers aren't just not great, they're subpar, and at every depth, too (but especially deep).

by David (not verified) :: Sat, 08/06/2005 - 10:02pm

Absolutely a great book! Both baseball and football prospectus bring me back to the days of reading Bill James' Abstract. Wish I'd heeded my father's advice and kept them! It feels like you're in some secret society, unlocking the keys to a (really small) part of the universe.

Anyway, I had one serious fantasy football related question. Dan Lewis' article, "OPE: How Do I Explain It?" has two sections that I want to highlight:

"4. Tight ends should be drafted late. Make exceptions only if Tony Gonzalez or Antonio Gates falls further than expected. Don't reach to get Jeremy Shockey, Jason Witten, or Alge Crumpler: you'll be fine with a Bubba Franks or Daniel Graham." (398)

It would seem that you didn't follow your own advice in the fantasy projections! These projections, remember, "give projected draft order based on marginal value of each player, the idea that you draft based on how many more points a player will score compared to the worst starting player at that position, not how many points a player scores overall." (426) Unfortunately, your projections are advising your readers to draft those gentlemen with the 12th (Gates), 13th (Witten), 16th (Gonzo), 34th (Crumpler), and 37th picks (McMichael). Surely Gates at the end of the first round in a twelve team league is not "dropping further than expected"?

Similarly, Lewis recommends last round or waiver wire pickups for kickers. But your projections recommend kickers starting in the fifth round with Vanderjagt. In fact, your projections recommend that six kickers should go in the first 70 picks.

If the projections incorporate the marginal value of the position, it should be taking into account the ideas Lewis expressed in the "Seven Habits of Highly Effective Fantasy Footballers," right? I feel like I'm missing a crucial piece here. Enlighten me?

Anyway, we'll see how drafting "the Lewis way" and drafting "the fantasy projections way" works out, as I'm trying one team based on each approach. I betting Lewis and the KUBIAK projections do better than KUBIAK projections alone, however.

by pm (not verified) :: Sat, 08/06/2005 - 11:23pm

A little off-topic:

Aaron will you be a permanent guest on the hotlist or was it a one time thing.

by David (not verified) :: Sat, 08/06/2005 - 11:41pm

How so? This section's to discuss the book right? I'm asking a question about how two sections of the book relate: the Lewis article and a part of the appendix.

by Zac (not verified) :: Sat, 08/06/2005 - 11:44pm


I think we are a ways away from being able to have a cheatsheet that you can follow 100%, just picking the highest player on it. Drafting is partly a science, and partly an art.

The players in the appendix are adjusted by the worst starter baseline. So basically, each player on the list is ranked based upon how many more points they will score than the worst starter (12th QB, 24th RB, 36th WR, 12th TE, 12th K in the 12-3WR league they list). This is called Value Based Drafting. For more info on VBD, go to http://www.footballguys.com/05vbdrevisited.htm.

It's also good to have average draft information telling you when players will normally go. I recommend http://www.antsports.com/adp.aspx for that. In it, you will see that in Normal TE-required 12-team leagues, Tony Gonzalez is usually the 1st TE drafted, and he gets picked, on average, at #30.

The #16 ranking in the book tells where the calculations say he should go (this is the "science" part). But since Gonzalez never goes that early, there's no point in drafting him that high. You might as well wait until the 3rd round (this is the "art" part).

by Zac (not verified) :: Sun, 08/07/2005 - 12:16am

I'm almost positive pm wasn't talking to you. He was saying in advance that his post was off-topic.

I posted a comment in response to you that must have gotten held up somewhere (or maybe I forgot to hit send on it). If my comment doesn't show up by tomorrow I'll rewrite it.

by Kaveman (not verified) :: Sun, 08/07/2005 - 4:37pm


Bailey was often put in man coverage on the opposing team's #1 receiver, with safety help funneled towards Kelly Herndon's side. Given that quarterbacks will avoid double coverage when possible, it isn't a surprise that Bailey was thrown at as much as he was. A Denver paper actually says that he was thrown at by opposing passers more often than any other Broncos DB in 2004. From that article:

"That's just the life of a man-to-man cornerback," Indianapolis Colts coach Tony Dungy said earlier this off-season. "We looked at it before the playoff game against Denver and he had hundreds of passes where he was in man coverage. You give up 10, 15 catches, it isn't the 485 other plays people remember, it's those 10, 15 plays.

Bailey highlights through the season:

- 9/12 vs KC: 1 on 1 against Tony Gonzalez, holding him to 2 receptions for 17 yards. Also had an INT.

- 9/19 vs JAC: 1 on 1 against Jimmy Smith, held him to 2 receptions. Leftwich had 120 yards passing that day.

- 10/10 vs CAR: 1 on 1 on Muhsin Muhammad, holding him to 1 reception for 9 yards, his lowest receiving yardage in 23 games.

- 11/7 vs HOU: 1 on 1 on Andre Johnson who had 3 receptions for 28 yards.

- 12/25 vs TEN: 1 on 1 on Drew Bennett, holding him to 2 catches for 26 yards. Two weeks before, Bennett had 12 for 233 yards and 3 TDs. The previous week, Bennett had 12 for 160 with 2 TDs. Granted though, that those games were against KC and OAK.

Bailey had other strong games, but we all know that he also had a couple of bad games, especially against Chad Johnson on Monday Night Football.

The trouble with statistics, as most readers of this site should know, is that it is very easy to misinterpret them if you aren't given sufficient context. I think this is the case in Joyner's analysis, since he seems unaware of how Larry Coyer uses Bailey in his defensive schemes.

by Kaveman (not verified) :: Sun, 08/07/2005 - 4:41pm

I forgot to say, got my PFP a few days ago, and am now happily digesting it. Great work, Aaron and everyone.

by Sean (not verified) :: Sun, 08/07/2005 - 5:36pm

Actually, Joyner knows exactly how Bailey was used, and he writes about it at length in his book. All his stats come from his own breakdown of game tape, not from any league source. I'm going to take the liberty of quoting extensively from his Bailey write-up:

"Bailey is widely known as one of the best CBs in the entire league, as evidenced by his 2004 All-Pro nomination. Bailey may have been a dominant CB in previous years, but the nomination doesn't jibe at all with his 2004 performance. Bailey gave up 7 TDs in coverage this past year, placing him tied for 71st in the league. He ranked tied for 71st in good/tight coverage. Bailey also tied for 73rd in burns against him, equating him with such luminaries as Fernando Bryant and Nick Harper. He also ranked 67th in completion percentage, ranking him near Lance Frazier and Nick Harper. He was 67th in total yards and tied for 65th in yards per attempt.

His rankings against deep passes were even worse. Bailey was 79th in deep yards, tied for 73rd in deep completion percentage, tied for 16th most in deep attempts, tied for 69th in deeo TDs, 73rd in deep tight/good coverage percentage, and had the 19th highest deep pass percentage of attempts. Bailey wasn't just beaten deep; he was targeted for deep passes quite frequently.

The Denver coaches tried explaining all of this by saying that the deep safeties blew some coverages and left Bailey in one-on-one situations. My only reply to this is look at the big plays (25+ yards) that Bailey gave up last year:
Week 4 at Tampa: Michael Clayton beats Bailey on a slant/go route for a 51 yd TD.
Week 7 at Cincy: Chad Johnson beats Bailey on a go route for a 50 yd TD, then beats him later for a 50 yd completion on a post route.
Week 11 at NO: Joe Horn beats Bailey for 27 yards on a hitch/go route.
Week 12 vs. Oak: Jerry Porter beats Bailey for a 42 yd TD on a go route, then beats him later for 52 yds on a post route.
Week 13 at SD: Eric Parker (yes, Eric Parker) beats Bailey for 30 yds on a slant/go route.
Week 16 at Tenn: Derrick Mason beats Bailey for 37 yds on a slant/go route.

It's not just these plays Bailey was beaten on. Here are the other TDS:
Week 12 vs. Oak: Jerry Porter again beats Bailey, this time on a seam route for a 14 yd td.
Week 15 at KC: Eddie Kennison beats Bailey for a 7 yd TD on a slant route, then beats him for an 18 yd TD on a post route.
Week 17 vs. Indy: Marvin Harrison beats Bailey for a 7 yd TD on a slant route. Keep in mind that this game was meaningless to the Colts and was for a playoff berth for the Broncos.
I don't have the week 2 Denver-Jacksonville game, but Jimmy Smith had a big game that day and his matchup was due to be against Bailey.

You simply cannot tell me that all of these plays were due to other people in the Broncos secondary blowing coverages, especially since John Lynch was voted to the Pro Bowl. Even if they did blow some coverages, wouldn't you expect Bailey to do better than this in man-to-man coverage? What's most surprising to me is that many of these big plays happened on nationally televised games, yet Bailey's reputation somhow remained intact.

His reputation was even enhanced when he didn't do anything. In the Wild Card playoff game the Colts targeted the Reggie Wayne-Roc Alexander matchup instead of the Marvin Harrison-Champ Bailey matchup, and the press reported that Bailey shut out Harrison. What a bunch of malarky! I covered this a bit in the defensive intro, but there's more to the story. Roc Alexander is a 3rd string nickel back and the Broncos decided to leave him in man-to-man coverage vs. Reggie Wayne. The Broncos also blitzed their safeties quite frequently, so Alexander didn't have the benefit of deep safety help on many of the plays. The Colts saw this total mismatch in their favor and targeted it very successfully all day (Wayne had over 200 receiving yards). Harrison's low numbers had nothing to do with Bailey's coverage and everything to do with the Colts going elsewhere, yet Bailey was still getting kudos. In fact, when you think about it, the Broncos had 2nd and 3rd string CBs in their secondary late in the season, and the other teams were still going deep at Bailey, and therein may lie the explanation for Bailey's performance.

I have 5 of the first 6 Broncos games from 2004, and Bailey was only thrown against deep on 4 passes in those 5 games. Over the course of the next 11 games (including the playoffs) Bailey was thrown at deep 20 times, and one of those 11 games was against Miami, with maybe the worst passing offense in the league.

I think it's quite possible that as the season wore on and the Broncos secondary issues started piling up, the Broncos coaches left Bailey in one-on-one coverage with his receiver. In fact, since I track whether or not a defender is playing man or zone coverage, I know Bailey was in man coverage.

As soon as other teams realized this, they started trying to pass deep on Bailey, if just to try going against man-up coverage. As Bailey started getting beaten more and more, teams started trying to throw deep against him more often. In the one game late in the season when Bailey got deep help, against the Titans and Drew Bennett, he and John Lynch were able to shut Bennett completely down.

It's entirely possible that Bailey needs to give himself more cushion at the snap (he had the 55th lowest soft coverage percentage). It wouldn't hurt his short pass completion percentage, as it is already 63rd in the league right now, but it could sure help him stop those deep passes."

That's a long quote, but consider it a plug for the guy's book, which is excellent and which makes a perfect companion piece to Pro Football Prospectus.

by Kaveman (not verified) :: Sun, 08/07/2005 - 6:40pm


Hard to argue with such specific numbers. Would you know, Sean, if Mr. Joyner gets team video, or if he's compiling his stats from recordings of the regular telecast? One would assume the former, as I don't think the regular telecast shows enough of the field to analyze defensive schemes and coverages.

In any case, I'm somewhat unconvinced (and this might well be because I'm a Broncos fan ;->) . Denver's pass defense DVOA in 2004 was -18.9%, ranked 4th in the league. If Champ Bailey, the #1 CB, was 60th to 70th in so many statistical categories, who was responsible for the Broncos' stellar pass defense?

by Aaron (not verified) :: Sun, 08/07/2005 - 7:05pm

The linebackers and safeties. :) If you see in the book, Denver was amazing against TE and RB, not so much against receivers. One of the downsides of KC's astonishingly in-depth research on the secondary and receivers is that the book has very little about linebackers or running backs, not just in the running game but in the receiving game.

by Led (not verified) :: Sun, 08/07/2005 - 7:31pm

It seems to me that in order to fully assess Bailey's play you need to look at the number of times he was in single coverage against number one receivers and compare it to some league baseline. The raw number of times he got beat, while troubling and certainly reason not to accept his reputation unquestioningly, doesn't tell us that enough without the context in which the events occurred.

by charles (not verified) :: Sun, 08/07/2005 - 8:52pm

Redskins fan here. i'll make this quick.
IMO, Bailey's weakness is that he stares in the backfield too long no matter whether he is in zone or man. He also lets too many receivers eat his cushion up believing he he is the fastest player in the nfl and can make it up. Any washington fan can tell you about the times he got beat by freaking amani toomer and james thrash.

I'm not mad at champ i'm mad at his coaches. This is the nfl if you leave a cornerback in straight man to man with no safety help for too many plays in a row he's going to get beat and beat alot.

by charles (not verified) :: Sun, 08/07/2005 - 9:27pm

Basically what i'm saying is that champ bailey was no worse in 2004 than he was any other year he was in the league, but the broncos had a better record and more exposure on tv so it just seemed like champ bailey struggled last year. it was a typical year, imo, for bailey.

by Sean (not verified) :: Sun, 08/07/2005 - 9:54pm

Aaron is absolutey right. Denver has one of the best collection of pass defending linebackers in the league, and John Lynch had a legitimately Pro Bowl caliber season. And obviously Bailey was effective at times, just not nearly often enough to justify being called one of the top corners in the game.

Joyner's numbers focus heavily on the quality of a player's coverage, whether or not a pass was actually completed against him. The Pro Football Prospectus defensive numbers concentrate on the results of plays that a player was actually involved in. And yet both measurements come up with a very similar set of numbers for Bailey. Joyner has Bailey thrown at 90 times, PFP has Bailey involved in 94 plays, and they both show Bailey giving up a high number of yards per play that he was involved in. That's the deep ball weakness rearing its head.

by SteelerBill (not verified) :: Sun, 08/07/2005 - 11:24pm

Each year the question is asked about which NFL Preview publication you purchase - the answer should be simple - Pro Football Prospectus...amazing...great stuff!!

by pcs (not verified) :: Mon, 08/08/2005 - 2:53am

For the next edition, it'd be good to include each team's schedule, or at least a list of their opponents. The NFL sked is set in April; I don't know whether that's early enough for the PFP lead time, but I think the next year's opponents can be determined as soon as the season ends.

by Sean (not verified) :: Mon, 08/08/2005 - 7:11am

Just to follow up on my earlier exchange with Aaron, I decided to hit the books and look at the different statistical breakdowns for the starting corners in both Scientific Football and Pro Football Prospectus. There may be no way to accurately determine who is being targeted without resorting to game tape, but I was curious to see if information coming from two different sources and with two different agendas would end up painting similar pictures about DB performance. Would a DVOA approach that emphasizes defenders holding down receivers to acceptable yardage totals on each down bring to the forefront the same players that a scouting approach with an emphasis on the quality of coverage on any given play and no particular emphasis on game situation would? So I looked at some key figures side-by-side. From PFP, I looked at the DVOA numbers- plays, stops, defeats, average yards per play, and stop rate percentage. From SF, I looked at total attempts against, completion percentage, yards per attempt, percentage of tight or good coverage, as well as missed passes and total missed pass yardage (passes directed at the corner that were incomplete due to a poor throw instead of good pass defense).

The five corners involved in the most plays:

T. McGee (126)
J. Butler (122)
D. Barrett (118)
M. Trufant (116)
R. Barber (109), S. Brown (109)

The highest number of stops:

K. Herndon (57)
S. Brown (54)
R. Barber (52)
D. Robinson (45)
C. Bailey (45)

Highest number of defeats:

K. Herndon (27)
D. Robinson (25)
R. Barber (24)
N. Clements (24)
C. Bailey (24)

Lowest average yards per play:

K. Herndon (4.7)
S. Brown (5.3)
R. Barber (5.4)
B. Kelly (5.9)
D. Townsend (6.3)

Highest stop rate percentage:

K. Herndon (61%)
F. Smoot (51%)
J. Azumah (50%)
P. Surtain (50%)
S. Brown (50%)
The first thing that’s notable is that Kelly Herndon’s stat line just jumps off the page. According to these measurements, he had by far the best performance of any corner in the NFL last year, so long as you don’t consider the quality of receivers he was matched up against. You look at his performance and wonder how in the world the Broncos decided to let him walk and go with Lenny Walls. The other players who appear regularly are Sheldon Brown, whose stop rate percentage is particularly impressive considering that he was in on 109 plays. The two Tampa corners stand out as having had exceptional years, and Champ Bailey’s numbers are actually very good in every area except average yards per play, where he is undone by his struggles defending the deep ball. As far as players being involved in a high number of plays, not surprisingly you see a lot of number two corners or guys on bad pass defenses or both. What is a bit of a surprise is how much action both Barber and Brown saw, which makes their numbers all the more impressive.

So how do those players fare when being broken down on film? Brown, Barber and Kelly fare pretty well, the other players less so. Let’s look at how the SF rankings break down:

Here are the top five in attempts:

M. Trufant (141)
A. Harris (133)
B. Williams (127)
T. McGee (126)
J. Butler (122)

A lot of overlap between attempts and total plays, suggesting that while it’s nice to have the additional information, you can get a good feel for which players were targeted in the secondary strictly through plays they were involved with. Indeed, when you look at the two lists, there are only a few instances where the starting corner that made more plays had fewer attempts thrown at him, and generally the numbers reflect similar proportions- Gary Baxter saw 104 plays to Chris McAlister’s 56 and 108 attempts to McAlister’s 65, for instance. The most striking instance of number reversal is in Oakland, where Charles Woodson made 90 plays to Philip Buchanon’s 68, yet Buchanon saw 76 attempts while Woodson had only 58 balls thrown his way. The implication is that Woodson saw heavy duty in run support, while Buchanon was useless in that regard.

Top five in completion percentage:

D. Bly (36.5%)
M. McKenzie (40.8%)
E. Warfield (44.0%)
S. Springs (44.6%)
D. O’Neal (44.8%)

Many of the leaders in low completion percentage played in man intensive schemes, and were helped by heavy blitzing to boot. Dre Bly is a player whose numbers will jump off the board in much the same way Kelly Herndon’s did- by the reckoning of these stats, it was Bly who was the best corner in the league last year, something that doesn’t come out in his DVOA numbers. I’ll touch on this more in the next section.

Top five in yds/attempt:

S. Springs (4.2)
D. Bly (4.4)
S. Brown (4.9)
W. Peterson (5.0)
B. Kelly (5.5)

Rhonde Barber just missed the list, continuing the strong showing by the two Tampa corners. Shaw Springs and Dre Bly are right up at the top of the league in yards per attempt, and in each case their numbers are tremendously different than their yards per play, where Springs averaged 6.9 and Bly a pedestrian 7.3. Bly’s numbers are understandably lower here, as the Lions played a lot of soft coverage, so Bly was making tackles in run support further downfield than cover 2 corners or guys who played close to the line in pressure schemes. The other player who is going to stick out here despite nothing DVOA numbers is Will Peterson. I think the reason for this is that he had a high number of penalties that were probably factored into his DVOA yards but not into Joyner’s ratings, because his overall stat line is lights out: 89 attempts, 44.9% completion percentage, 5.0 yards per attempt, and, as we see in a moment, the highest tight coverage percentage in the league.

Top five in tight/good coverage percentage:

W. Peterson (37.1%)
M. McKenzie (36.6%)
K. Herndon (33.3%)
A. Samuel (33.3%)
A. Harris (32.3%)

As with the completion percentages, some of these guys are helped by playing in an aggressive man scheme. When you look at the full stat lines for McKenzie and Harris, you see that they allowed low completion percentages but gave up a lot of yards, as when they were burned, they were burned for big yardage, an inevitable result of playing without safety help.

What players have the biggest disparities in their stat lines? Well, here are a few that stand out:

K. Herndon: 94 plays, 57 stops, 27 defeats, 4.7 avg. yds, 61% stop rate
K. Herndon: 96 attempts, 54.2% completion, 8.0 yds/att., 33.3% tight/good coverage

Herndon must have been involved in a lot of run support for his yards per play to drop so low. If he was tackling runners instead of defending passes, that would also help explain why his stop numbers are so good.

S. Madison: 59 plays, 19 stops, 9 defeats, 10.1 avg. yds, 32% stop rate
S. Madison: 69 attempts, 50.7% completion, 6.4 yds/att, 17.4% tight/good coverage.

According to DVOA, Patrick Surtain was the better of Miami’s two corners, with a higher number of stops, defeats, stop percentage, and a significantly lower average per play- 6.3 to 10.1. According to Joyner’s numbers, Madison was the better player, with a lower completion percentage and a lower yards per attempt- 6.4 to 8.6. I’m not sure what to make of this, other than to suggest that Madison ended up making some tackles way downfield either against runners or receivers who he hadn’t been covering, and that his numbers get hurt as a result.

by Aaron (not verified) :: Mon, 08/08/2005 - 9:44am

Sean, I've always been impressed by your research-packed message board posts -- and puzzled as to why you don't submit them as guest columns, where they might get more attention.

Anyway, I just wanted to make clear that the numbers from PFP that you quote aren't DVOA, even though they use the same system of declaring success at 40%/60%/100% of yards on 1st/2nd/3rd down. They are numbers we borrowed from Roland Beech and are unadjusted in any way. Plus, as you note, they don't split rush and pass plays (one of the improvements we hope to add next year).

by Sean (not verified) :: Mon, 08/08/2005 - 2:56pm

Thanks. I think I've just been lazy about going the extra mile and putting something in a column format, but I'll take you up on that.

Gotcha on the defensive numbers- I knew they weren't an exact match with DVOA, but that they were working off a similar concept.

by Lasse Johansen (not verified) :: Mon, 08/08/2005 - 3:46pm

I just wanted to tell you that your book has made to Copenhagen, Denmark.

I has been a long wait but it was well worth it, it is a great book and a great site.

A big thank you from Morten Andersen land :-)

by Todd S. (not verified) :: Mon, 08/08/2005 - 6:03pm

"The Ming of Beers."

Well done.

by Mukaikubo (not verified) :: Mon, 08/08/2005 - 7:22pm

I thought I saw a stat in the book about how, of 9 teams that did 1 or more better than Pythagorean projections would indicate, all but the Eagles and Cardinals regressed. Where is it in the book, if I didn't hallucinate it?

by Ned (not verified) :: Tue, 08/09/2005 - 1:33pm

Re: 85
I think it is in the Jacksnoville chapter.

by Sid (not verified) :: Tue, 08/09/2005 - 11:12pm

Just got back from the book signing. :)
It was pretty interesting.

by Sid (not verified) :: Wed, 08/10/2005 - 1:09am

RE: 46

Nice article, Zac. It was interesting to go back and see if some of the predictions in that 1990 article were true.
Barry Sanders was simply incredible, and certainly one of the five best RBs of all-time (and arguably the best, since he would've run away and hid with the all-time rushing record had he not retired early).

by R.J. (not verified) :: Wed, 08/10/2005 - 3:22am

Great to see the discussion here about Champ Bailey, as I have seen references to Joyner's criticism of him in other publications -- but without putting the negative stats in context. I watched and re-watched every Bronco play last year and overall Bailey played very well, nearly great. I won't rehash the great games listed above (post 69), but I do want to put his performance into the proper context that Joyner's criticism of him misses. First, (as already mentioned here) the fact that Bailey was consistently required to play the best receiver on the other team in sole man coverage and MORE IMPORTANTLY, he did that while the ridiculously anemic pass rush of the Denver defensive line was trying to get to the QB. No matter how good a CB is, he can't cover a top wideout all over the field forever -- and many times that is exactly what Bailey was asked to do. (That is also why the numbers show he was "targeted" deep so often.)
Denver's overall defensive numbers were very good -- with my own eyes and judgment I know it was because Bailey was able to do so much on his own. Bailey definitely made a few big mistakes (and I agree it was partly from overconfidence in his ability to close) -- but I can't imagine more than a couple of CBs who could have performed as well under those circumstances.
One last point, Bailey is consistently ranked as the best CB in polls of NFL players . . . are they also caught up in the media hype about him? Or do they possibly understand something about what he is capable of doing and is asked to do on the field?

by Miles (not verified) :: Wed, 08/10/2005 - 3:55am

Re: downloadable projections. I wonder if one future compromise might be to put in a code in the book that would get a buyer access. Certainly the unscrupulous could share the code with others that have not bought the book, find the code while flipping through the book at the local bookstore etc. However, the same sort of abuses can occur with the current system (publish the spreadsheet to a P2P file sharing system, OCR the physical pages like the Harry Potter book etc).

by Nelphonious of Pennefielde (not verified) :: Wed, 08/10/2005 - 11:49am

Aaron/What are your top 10 mythbusting football cliche uncoveries(ie.uncoachable fumbling recovery)your data crunching bases have revealed at FO?

by Ray (not verified) :: Wed, 08/10/2005 - 12:16pm

What sort of things happen at the book signing? How long does it go on? Do I have to be there when it starts, or can I come a half-hour to an hour later and not miss much?

by Richie (not verified) :: Wed, 08/10/2005 - 2:26pm

Does there exist a listing of cumulative opponents DVOA for 2005 broken down into offensive and defensive?

eg, what is the Colts' cumulative opponenents' defensive DVOA for 2005?

by Kaveman (not verified) :: Wed, 08/10/2005 - 8:06pm

#89's last point brought to mind another thing. Someone correct me if I'm wrong, but I believe that who goes to the Pro Bowl is determined not just by fans, but by the players and coaches as well, and each group has equal voting power. Fans can and do get caught up in media hype, but it is hard to believe that players and coaches decide their votes based on what they read on ESPN.com.

Regarding the Broncos defensive line, it might interest people that Gerard Warren, Michael Myers and Courtney Brown (until his elbow dislocation) are playing on the first team with Trevor Pryce. Ekuban has now replaced Brown. Last year's starters (barring Hayward-free agency, and Ellis Johnson-released) are all lower on the depth chart. Engelberger is on the second team behind Pryce.

Then again, this might interest only Broncos fans... :->

The PFP doesn't list Gerard Warren in the Cleveland Browns section. From the Broncos' website, Warren played 13 games (13 starts) at defensive tackle for Cleveland to record 38 tackles (21 solo), four sacks (26.5 yds.) and a team and career-high three forced fumbles. He ranked third on the Browns in sacks and tied for second on the team with 21 quarterback pressures. Seems like he meets the criterion to be listed (involved in 20 plays)?

by Larry (not verified) :: Fri, 08/12/2005 - 4:00pm

Joining the discussion late, wanted to read most of the book before throwing in my 2 cents. I love the probability prediction, rather than a single number. One of FO's lovely Excel graphs with each possible win total would be very cool, and I wouldn't object to listing the median prediction, either, I guess. However, not cowing to the conventional numbers is what makes FO great, so I wouldn't want to see a floating single number out there without a mathematical explanation of what exactly it was and why it means something.

On a different note, I just wanted to say I love the book, and perhaps start a trend by saying that the pieces I found most fascinating were the essays on drive momentum and the Bears research article on the cost of sacks. Absolutely great stuff that really adds to understanding the true impact of things that happen during a game and sheds light on strategic choices coordinators can make.

by Matt (not verified) :: Fri, 08/12/2005 - 4:45pm

I'm enjoying the book, which I purchased primarily for Fantasy-realted reasons.

Aren't I your primary audience? You guys HAVE to figure out a way to include your projections in the purchase price.

Also, given the weight DVOA et al put on stats similar to YPA, YPC, etc., why don't you include "passes", "rushes", "catches", and even "targets" in your projections? Clearly your KUBIAK computer is capable of spitting out more than these basic fantasy-only stats...

Or is that what the extra $10 is for?

by Sid (not verified) :: Mon, 08/15/2005 - 1:45pm

Don't miss Aaron in the Indianapolis area.

He'll be on a radio show today, and then another tomorrow, followed by the book tour stop tomorrow.

WIBC 1070 Indianapolis
Mon. Aug 15 7:00pm

ESPN 950 Indianapolis
Tue. Aug 16 4:05pm

I believe those times are Eastern.

by Sid (not verified) :: Mon, 08/15/2005 - 8:40pm

They've been talking about raising hogs for the past half-hour on WIBC. Doesn't seem like they're big into sports over in Indianapolis, lol.

by Dean (not verified) :: Wed, 08/17/2005 - 1:04pm

And the Prospectus makes it to...Perth, Australia.

Lots of fun reading ahead, thanks to Aaron and everyone at FO.

by Chad Eden (not verified) :: Thu, 08/18/2005 - 6:11am

Do you update your projections after training camps? If so, can we purchase this? Thanks.

Mitch Eden

by jimmo (not verified) :: Fri, 08/19/2005 - 1:30am

pleasantly surprised to see PFP in stock, and prominently displayed, in a cool bookstore in Traverse City MI this past weekend. Good job.
On the negative side, and not to beat a horse too hard, that misspelling of Hasselbeck's name of the cover truly is horrible, no matter who's at fault. Its small, it happens, but I have to admit that I wouldn't have purchased the book simply because of that error if I didn't already know the quality of the FO content inside.
Having said that, PFP delivers and then some. Great work!

by Geschenke (not verified) :: Tue, 12/25/2012 - 5:05pm

Denn auf unserem Geschenke des weiteren Geschenkideen Blog
präsentieren unsereins schöne Dinge und Wohnkamine.
Wir ziehen umher und sehen immer wieder mal was
uns sehr gefällt und das präsentieren wir hier.
Auf jeden fall können Sie viele Produkte auch sogleich kaufen,
jedoch ausschließlich per amazon, damit ihr Einkauf sicher ist natürlich!