SITE NEWS: We are moving all of our site and company news into a single blog for We'll tag all PFR content, so you can quickly and easily find the content you want.

Also, our existing PFR blog rss feed will be redirected to the new site's feed. ยป Sports Reference

For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.

Archive for April, 2008

Classing Up the Joint

Posted by Sean on April 25, 2008

You may have noticed that most of the SR sites have a shiny new logo in the upper left-hand corner. We had been talking about getting nice logos for years, but the formation of the LLC this past December finally gave us the boost to actually get something done. In fact, when we got together at the Baseball Winter Meetings to sign the documents that offically formed the LLC, we met a graphic designer by the name of John Hartwell. John runs a company named Hartwell Studio Works. He has done a lot of sports-related work, including designing the logo of the Birmingham Barons, a minor league baseball team. We have been extremely impressed with John, not only the finished product, but also with the professionalism he displayed throughout the process.

Our hope is that these new logos further unify the sites. We want people of to think of our web sites as parts of a whole, not distinct entities, and we believe these logos are an important step in that process. Please feel free to let us know what you think (you're never shy about that, are you?), and as always we thank you for your support of our sites.

10 Comments | Posted in P-F-R News

An AV-based draft chart

Posted by Doug on April 25, 2008

This is my current best attempt at an Approximate value-based draft chart.

I looked at all drafts from 1980--1999. I figured 1980 was late enough to be "modern," 1999 was early enough to have allowed most of the players to have finished their careers, and 20 years was enough to get a decent sample size.

Then, for each player, I computed what I call his "draft value":

Draft Value = (Total approx value of first four seasons) + (90% of AV of fifth season) + (80% of AV of sixth season) + ... + (10% of AV of 13th season)

The basic idea is this. When you draft a player, you're getting roughly his first four years, and then some uncertainty. This was just a quick way for me to capture the fact that there is some chance that your draftees will be helping some other team and not yours in five, six, seven years time. Also, teams presumably value 2008 production more than 2015 production.

Then I looked at the draft values of all players selected in a certain slot, and bunged them down into a chart. Well, sort of. Let me take a line and explain it:

7 Comments | Posted in Approximate Value, NFL Draft

Assessing the Jared Allen trade

Posted by Doug on April 24, 2008

As you probably are aware, 26-year-old all-pro defensive end Jared Allen was just traded from the Chiefs to the Vikings in exchange for the 17th, 73rd, and 82nd picks in the upcoming draft. There was also a swap of sixth-rounders, but I'm going to ignore that.

This seems to me like a good deal for both teams. My favorite Viking blogger, Pacifist Viking, likes the deal from the Minnesota side:

If anybody would like to complain about giving up draft picks, I encourage you to look at Grant's Tomb for a list of players the Vikings have drafted since 2000 [Doug note: or you could click this customized query]. The draft is hit or miss: you can score big (Kevin Williams, E.J. Henderson, Adrian Peterson), but you can also come away with no meaningful contributors (see 2000, 2001, and 2005). Draft picks in and of themselves are worthless; draft picks are only meaningful because they can be turned into starting players. The Vikings turned those three draft picks into a 26 year old who had 15.5 sacks in 14 games last season. They turned their draft picks into a very good football player, and that's what draft picks are for.

Very true; with all the draft hype, we have a tendency to forget that picks are just a means to an end. And they are hit-or-miss. But some more so than others. A third-round pick has a better chance of becoming a productive player than a fifth-round pick does, but not as good as a first-round pick. We're dealing with potentials here. Likelihoods. Probabilities. And that was PV's point. But I'd add that Jared Allen is just potential at this point too. We don't know what Allen's future looks like. Granted, we're much more confident about his future than we are about pick #73, but Allen belongs on the probability continuum too, just like a first-round pick, a third-round pick, and a fifth-round pick. Yes, he's very near the good end of it, but he's still on it. Everyone is.

This kind of thing is why I felt the need to develop the approximate value method. If we can put numbers on player seasons, then we can attempt to quantify statements like PV's above. What are these probabilities? What is the probability that the #17 pick in the draft will turn into a solid NFL player? What's the probability that Jaren Allen will continue to be one? What's the probability that the #17 pick will be more productive than Allen over the next year? Two years? Five years? What's the probability that at least one of the three picks the Vikings give up will be more productive than Allen?

Here's what I did.

9 Comments | Posted in Approximate Value, General, NFL Draft

Best draft classes

Posted by Doug on April 23, 2008

Using my still-in-beta approximate value method (which you can read about here, here, here, and here), I decided to identify the best draft classes since 1970.

The obvious choice is the 1974 Steelers' draft, which produced three Hall of Famers and one other guy who also got elected to the Hall of Fame. Pacifist Viking, however, makes a strong case for the 1986 49ers' class which lacks Canton residents but boasts seven players who started for five or more seasons. Nailing down a third of a team for half a decade in a single draft is pretty impressive.

My method ranks these classes #2 and #4 respectively. In between them is the 1981 Saints class, which is even deeper (but also even more lacking in true star power) than the '86 49ers' class. The Saints landed an astounding 12 players who would be NFL starters for at least two seasons. To give you a sense, consider Russell Gary. He started every game of the first four years of his career, and was even named by the UPI to the all-NFC team one year. He was, according to my method, the seventh-best player drafted by the Saints in 1986.

Before I reveal #1, let me describe the method.

11 Comments | Posted in Approximate Value, History, NFL Draft

Jake Long draft notes

Posted by Chase Stuart on April 22, 2008

The NFL and AFL had their first common draft in 1967, making this year the 42nd draft in modern professional football history.

Jake Long becomes the third offensive lineman selected with the first pick, joining Orlando Pace and Ron Yary. Each made 7 Pro Bowls.

Fifteen lineman have now been drafted first overall, with twelve of them being defensive players.

Exactly two-thirds of the number one draft picks have been offensive players.

Long becomes the eighth player from the Big 10 to be drafted first overall, joining Courtney Brown, Pace, Ki-Jana Carter, Dan Wilkinson, Jeff George, Tom Cousineau and Bubba Smith.

Long becomes the first Michigan Wolverine ever selected with the number one overall pick. USC has the most number one picks, with five.

Long becomes the 37th first draft pick selected from a current BCS school. Alex Smith (Utah), David Carr (Fresno State), Ed "Too Tall" Jones (Tennessee State), John Matuszak (Tampa) and Terry Bradshaw (Louisiana Tech) were all drafted from current non-BCS schools or non-Division 1 schools (Jones, Matuszak).

According to the current breakdown of schools, the ACC and the Big 12 have five number one picks each, the Big 10 has eight, the SEC has seven, and the Pac-10 has eleven. Notre Dame has one (Walt Patulski), and the Big East has zero. Mike Vick was selected first overall when Virginia Tech was in the Big East, but the Hokies are now in the current ACC. The current WAC has two (Carr, Bradshaw) and the current Mountain West has one (Alex Smith).

Long (6'7) joins Bubba Smith and Orlando Pace in a tie for the third tallest player ever picked first overall. Ed "Too Tall" Jones was the tallest first pick ever at 6'9, and John Matuszak ranks second at six feet, eight inches. Ki-Jana Carter, 5'10, was the shortest top selection ever.

Long is also the third heaviest player ever (315 lbs) drafted first overall, behind Dan Wilkinson (335) and Orlando Pace (325). Irving Fryar was the lightest first pick, weighing just 200 pounds.

Long is also the first native Michigander selected at the top of the draft. Washington, Texas and Ohio have each had four top picks, while California leads the country with six sons selected first (Palmer, Aikman, Keyshawn Johnson, Plunkett, David Carr and O.J. Simpson).

Long is the tenth oldest player selected first overall. Billy Simms was the oldest, as he turned 25 in September of his rookie season. Mike Vick was the youngest, and both he and Alex Smith were just 20 years old when drafted.

Long is the 14th player selected by an AFC East team with the first pick. The Bills have four (Bruce Smith, Cousineau, Patulski and Simpson), the Baltimore Colts have one (Elway; the Colts were not in the AFC East when Bubba Smith was selected), the Indianapolis Colts have three (Manning, George and Emtman), the Jets have one (Keyshawn) and the Patriots have four (Drew Bledsoe, Irving Fryar, Kenneth Sims and Jim Plunkett). This is the first number one pick by Miami.

Comments Off on Jake Long draft notes | Posted in NFL Draft

Approximate value, part IV

Posted by Doug on April 22, 2008

Links to part I, part II, part III,

A few weeks ago, I asked the fine folks at the footballguys message board to rank the following list of players according to overall career (not peak) value.

Dale Carter
Cris Carter
Kevin Carter
Dexter Coakley
Warrick Dunn
Dan Hampton
Seth Joyner
John Lynch
Clay Matthews
Eric Moulds
Anthony Munoz
Roman Phifer
Bruce Smith
Neil Smith
Aeneas Williams

The idea was to take players from a variety of positions, decades, and overall quality levels, and see if my method put them in roughly the same order as the fan consensus. After posting the list, I realized that I should have included a few more offensive linemen. Munoz is the only one, and he's easy to rate. The omission of quarterbacks was intentional; I'll talk a little about that later.

I realize this isn't a tightly controlled scientific survey, but here is the consensus of the 13 sets of rankings that were offered at the time I tallied them (a few more sets of rankings trickled in later; they're not included here):

10 Comments | Posted in Approximate Value, General


Posted by Doug on April 15, 2008

The sports-reference family of sites is pleased to bring you Please enjoy the site, and tell all your hockey-stat-loving friends.

Justin K, of fame, is heading up this project. If you want to offer general or specific suggestions or criticisms, or just congratulate him on a job well done, please do so at the h-r blog.

Comments Off on Announcing | Posted in P-F-R News

All-pro information upgraded

Posted by Doug on April 13, 2008

When the site re-launched in December, a player was declared an an all-pro if he made first- or second-team on any of the various all-pro teams that were named in a given year. So a second-team all-NFC player and a unanimous first-team all-NFL guy were both simply called "all-pro." You may have noticed that, a few days ago, we essentially switched the meaning of "all-pro" to "first team AP all-pro." We've left that definition intact, but have now added a great deal more info.

The all-pro pages --- like this one, for example --- now provide you with a lot more information on which players were named to which teams. Also, the player pages have a list of all-pro honors, right below their leaderboard appearances. Jerry Rice made a lot of all-pro teams.

And a bit of unrelated news....

Prepare yourself for a site announcement tomorrow (Monday) night that some percentage of you will be very, very excited about. I really don't know what that percentage is, but if you're in it, you won't want to miss this announcement. UPDATE: announcement postponed until Tuesday morning. Sorry for the delay.

9 Comments | Posted in P-F-R News

Approximate value, part III

Posted by Doug on April 10, 2008

Here are the links to part I and part II.

Getting approximate values for defensive players proved to be quite a bit tougher than I had anticipated. What I've ended up with is a method that seems pretty sloppy, with lots of arbitrary "add three points for this" and "multiply this times 1.6" sort of rules. But I generally like the results it gives. And since the point of this is not to be theoretically correct, but to build an objective method that matches peoples' subjective opinions, I'm OK with it.

I'm going to do here the opposite of what I did in parts I and II. That is, I'm going to post some lists before posting the complete methodology. The basic idea is the same on the defensive side of the ball as on the offensive: while we don't know the contributions of each individual player, we do know what they should add up to, because we know how good the team was. So I've given each team defense a certain number of points, based on its points allowed per (estimated) possession compared to the league average. And then I've done my best to divvy up those points among those players according to: sacks (post-1982), interceptions, defensive touchdowns, tackles (post-1994), and all-pro status.

As I mentioned above, the exact mix of those ingredients is something I arrived at by trial and error. I've spent a lot of time asking, why does that guy rank so high? Oh, it's because of the way such-and-such a stat gets figured in. That needs fixing. OK, so now why is this other guy so low? Ah, I see, let's tweak that a little.

The main "breakthrough" here, which I've also applied to the offensive linemen, is that, instead of having a binary was-he-or-wasn't-he an all-pro system, I am now using three different degrees of all-pro-ness: first-team AP all-pro, second-team AP all-pro, and pro bowler who didn't make the first- or second-team all-pro list. This helps to separate the Kevin Williamses from the Trent Coles. Not that there's anything wrong with Trent Cole, but he's no Kevin Williams.

Enough chit-chat. I'm just going to post my approximate value estimates for a handful of famous teams and a few '07 teams. The number immediately after each player is my approximate value estimate. The second number, which is either a 1, 2, or 3, signifies the level of all-pro-ness of that player for that year. 1=first team, 2=second team, 3=pro bowler not on first or second team. But note that this information is only figured into the system at the non-skill positions, where I feel like we don't have the stats necessary to give them reasonable ratings without it. In other words, I'm comfortable (approximately) rating Derek Anderson and Franco Harris and Wesley Walker without knowing whether or not they made the pro bowl in a given year, but I'm not comfortable rating Jeff Saturday or Dave Duerson or Terrence Newman.

Peruse the lists and let me know what sticks out. Who's way off? Do you see anything that appears to you to be a systematic error?

15 Comments | Posted in Approximate Value, General