SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.
Also, our existing PFR blog rss feed will be redirected to the new site's feed.
Pro-Football-Reference.com ยป Sports Reference
For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.
Guest post: Best Draft Classes Revisited
Article by frequent pfr-blog commenter Richie Wohlers. Any transcription errors are the fault of Doug.
About 5 years ago I began a project where I wanted to try and evaluate NFL draft performances over the years. I wanted to come up with a simple method for evaluating all NFL players, regardless of position. I decided that the main goal of an NFL team when it drafts a player is to draft a player who is going to play in NFL games. So I figured I could just rate all players by the number of games they played. I decided to also award bonus points for players who made Pro Bowls or the HOF.
So I began to manually enter this information for all players. I could get some information from NFL.com and some from this website. But all the data was not available for all players. Needless to say, I never got too far in my task. I completed the 1999 draft for all players I could find, and then I pretty much put my project on the backburner. Then, pro-football-reference.com added all this wonderful draft data and games played stats for basically every player since the 1950's, I could finally finish my project. I was just about done with my research when Doug came up with his Approximate Value formula and did a research project that was similar, yet superior, to mine.
Even though my method is more of an estimate than the AV method, and even though we've read a couple of posts on a similar topic, I went ahead and finished my research. I pulled out a few pieces of information which are a little different than what Doug and Chase have already posted in the past few months.
Methodology.
I gave each player a point value for their entire career.
- 1 point for every NFL game played
- 14 points for every All Pro season
- 10 points for every Pro Bowl season
- A bonus of up to 200 points for the percentage of seasons designated as a starter that ended up being All Pro seasons.
I suppose it could be argued that simply playing in an NFL game should not be considered a positive, and in some cases, such as Ryan Leaf, just playing in the game may be a negative to your team. But overall, I think that if a player is finding game time, then he is creating value for his team. Using this formula, the top 20 players who have been drafted since 1970 are:
Bruce Matthews G 918 Anthony Munoz T 792 Mike Webster C 777 John Hannah G 748 Barry Sanders RB 741 Randy White DT 740 Jerry Rice WR 738 Lawrence Taylor LB 726 Mike Munchak G 715 Mike Singletary LB 699 Bruce Smith DE 694 Ronnie Lott DB 689 Dan Marino QB 687 Walter Payton RB 686 Jack Lambert LB 680 Jackie Slater T 677 Jack Youngblood DE 676 Joe DeLamielleure G 656 Randall McDaniel G 656 Jack Ham LB 652Obviously, length of career becomes a pretty big factor in how a player is rated, but for the purposes of evaluating a draft I think this is not unreasonable. What better draft pick is there than to pick a future Hall Of Famer who plays 18 years in the league (like Bruce Matthews)? In my opinion, the goal of a draft is to pick good players who will be able to contribute to your team for a long time.
Over the years, the length of the NFL draft has changed quite a bit. Between 1970 and 1976 the draft was 17 rounds long. Between 1977 and 1992 the length was dropped down to 12 rounds. In 1993 the NFL drafted 8 rounds and between 1994 and 1999 the draft has been cut to just 7 rounds.
Based on this method, the 10 worst HOF inductees were:
John Riggins RB 323 Lynn Swann WR 344 John Stallworth WR 359 Terry Bradshaw QB 371 Jim Kelly QB 385 Michael Irvin WR 404 Troy Aikman QB 410 Kellen Winslow TE 425 Earl Campbell RB 432 Art Monk WR 436A couple of these guys like Lynn Swann and Art Monk are players who are often debated as unworthy HOFers. Then there are guys like Kellen Winslow, Earl Campbell and Michael Irvin who had short careers, but high peaks which were enough to get them into the Hall Of Fame. There's also a batch of guys who were good-to-great players that won multiple Super Bowls (Bradshaw, Aikman). John Riggins is rarely talked about as an undeserving Hall Of Famer, but it could probably be argued that he is undeserving. On the other hand, he was one of the guys who had a nice peak to his career (at an older age) that pushed him over the top.
It's interesting to note that out of the 30 #1 overall picks in this study, only 5 have made the HOF - John Elway, Lee Roy Selmon, Earl Campbell, Troy Aikman and Terry Bradshaw. Although Bruce Smith, Orlando Pace and Peyton Manning all seem likely to join them.
I calculated the draft value for every player. This is the actual overall position he was drafted compared to his actual career ranking amongst all players drafted the same year. For instance, the Baltimore Colts selected Stan White with the 438th overall draft pick. Stan White ended up being the 24th-best player from the 1972 draft. That gives him a draft value of 414 (438 minus 24). Here are the 10 best and 10 worst draft values of all time.
10 Best Values
Year Rnd Pick Tm Player PosPts Yr Rank Value College/Univ 1972 17 438 BAL Stan White LB 153 24 414 Ohio State 1974 17 418 SDG Charles DeJurnett NT 118 19 399 San Jose State 1971 17 436 DET Gordon Jolley G 59 49 387 Utah 1970 17 430 HOU Julian Fagan P 56 45 385 Mississippi 1971 17 441 BAL Don Nottingham RB98 56 385 Kent State 1972 17 439 KAN Ted Washington LB 128 54 385 Miss. Valley State 1974 16 404 DEN Darrell Austin G 69 36 368 South Carolina 1976 15 420 STL Lee Nelson DB 135 52 368 Florida State 1976 14 377 TAM Carl Roaches WR 76 10 367 Texas A&M 1975 17 422 NOR Greg Westbrooks LB 66 59 363 ColoradoWorst Values
1997 1 15 MIA Yatil Green WR 8 148 -133 Miami (FL) 1985 1 9 PHI Kevin Allen T 16 143 -134 Indiana 1987 1 6 STL Kelly Stouffer QB 22 142 -136 Colorado State 1988 1 26 DEN Ted Gregory NT 3 164 -138 Syracuse 1987 1 5 CLE Mike Junkin LB 20 144 -139 Duke 1987 1 7 DET Reggie Rogers DE 15 148 -141 Washington 1990 1 7 DET Andre Ware QB 14 149 -142 Houston 1981 1 6 GNB Rich Campbell QB 7 151 -145 California 1983 1 21 PIT Gabe Rivera NT 6 166 -145 Texas Tech 1986 1 1 TAM Bo Jackson RB 0 154 -153 AuburnBo Jackson was drafted first overall by the Tampa Bay Buccanneers in 1986. Tampa Bay was never able to sign him and he returned to the NFL draft the following year. So his value for Tampa Bay was zero. In fact, if you figured the opportunity cost in drafting Bo Jackson, his value is less than zero. The best player of that draft ended up being Pat Swilling. All of the top values above were guys taken in the 14th round or later. Since nobody since 1977 could match them in value, here is the list of top 10 values since the implementation of the 12-round draft in 1977:
Top 10 draft values since 12-round draft:
Year Rnd Pick Tm Player Pos Pts YrRank Value College/Univ 1990 12 329 NYG Matt Stover K 367 7 322 Louisiana Tech 1979 12 328 RAM Drew Hill WR 319 10 318 Georgia Tech 1987 12 334 DEN Tyrone Braxton DB 249 20 314 North Dakota State 1977 12 317 OAK Rod Martin LB 364 8 309 USC 1991 12 326 WAS Keenan McCardellWR 247 19 307 UNLV 1983 12 334 MIA Anthony Carter WR 258 32 302 Michigan 1983 12 310 DEN Karl Mecklenburg LB 419 11 299 Minnesota 1977 12 334 OAK Rolf Benirschke K 189 38 296 California-Davis 1988 12 320 MIA Brian Kinchen TE 233 29 291 LSU 1989 12 316 WAS Jimmie Johnson TE 218 25 291 HowardKickers are not considered to be a good use of high draft picks, but Matt Stover and Rolf Benirschke ended up being very valuable uses of 12th-round draft picks. The list above contains 6 players who actually ended up creating most of their value for teams other than those that drafted them: Stover, Hill, McCardell, Carter, Benirschke and Kinchen.
And here are the top 10 and bottom 10 values since 1993:
Top 10 since 1993
Year Rnd Pick Tm Player Pos Pts YrRank Value College/Univ 1995 7 243 MIN Jason Fisk DT 207 23 220 Stanford 1995 7 230 GNB Adam Timmerman G 243 15 215 South Dakota State 1994 7 218 DEN Tom Nalen C 353 6 212 Boston College 1997 7 231 GNB Jerald Sowell FB 197 29 202 Tulane 1993 8 207 NYG Jessie Armstead LB 393 6 201 Miami (FL) 1999 7 213 GNB Donald Driver WR 230 12 201 Alcorn State 1993 8 214 HOU Blaine Bishop DB 309 14 200 Ball State 1994 7 222 NWE Marty Moore LB 212 23 199 Kentucky 1994 7 220 HOU Lemanski Hall LB 201 30 190 Alabama 1996 6 208 GNB Marco Rivera G 273 18 190 Penn State 1997 7 230 NWE Scott Rehberg G 179 40 190 Central MichiganWorst
1995 1 1 CIN Ki-Jana Carter RB 59 116 -115 Penn State 1996 2 38 HOU Bryant Mix DE 7 153 -115 Alcorn State 1996 2 45 MIN James Manley DT 0 160 -115 Vanderbilt 1998 2 31 OAK Leon Bender DT 0 146 -115 Washington State 1999 1 24 SFO Reggie McGrew DT 24 140 -116 Florida 1999 1 29 MIN Dimitrius Underwood DT 19 145 -116 Michigan State 1994 1 5 IND Trev Alberts LB 29 123 -118 Nebraska 1997 2 32 ATL Nathan Davis DE 6 150 -118 Indiana 1997 1 26 SFO Jim Druckenmiller QB 6 150 -124 Virginia Tech 1996 1 30 WAS Andre T. Johnson T 3 157 -127 Penn State 1999 2 33 CIN Charles Fisher DB 1 161 -128 West Virginia 1993 1 24 PHI Leonard Renfro DT 23 153 -129 Colorado 1997 1 15 MIA Yatil Green WR 8 148 -133 Miami (FL)The best team draft of all time belongs to the Dallas Cowboys. That draft was anchored by Randy White who accumulated 740 value points over his career. The Cowboys picked 10 players with career values of 100 or higher including Pat Donovan, Herbert Scott, Bob Breunig and Thomas Henderson. The only other team to pick 10 players with career values of 100 was the 1981 Saints. That draft was led by Rickey Jackson and included George Rogers Frank Warren and Hoby Brenner. It ranked as the 4th-best total draft value of all time.
Out of the top 10 drafts, five teams went on to Super Bowl appearances shortly after making those drafts: Dallas, Pittsburgh (on the list twice), Buffalo and Miami.
The worst draft of all time belonged to the 1975 Kansas City Chiefs. Interesting that it happened the same year that the Cowboys had the greatest draft of all time. The Chiefs picked only one player who ever played in an NFL game. Morris Lagrand played 13 games in his career. The 1989 Los Angeles Raiders had the 2nd-worst draft of all time, but that ranking is mitigated by the fact that they had no picks in the first 5 rounds of the draft due to trades.
Top Team Drafts
Year Team Pts 1975 DAL 3171 1971 PIT 2480 1974 PIT 2472 1981 NOR 2234 1977 NYJ 2167 1985 BUF 2116 1983 HOU 2081 1985 HOU 2071 1975 SDG 2054 1983 MIA 2048Worst Drafts
1988 DEN 188 1971 MIN 181 1975 STL 177 1999 NOR 174 1980 SDG 142 1978 CHI 80 1977 WAS 57 1976 WAS 50 1989 RAI 36 1975 KAN 13Here are the teams with the best and worst average player value for their drafts any given year. Since so many late round picks provided little value in the days of the 17-round draft, no teams from that era make this list. The 1994 Seahawks had only 5 draft picks, but they made the best of it. Every one of their picks had career value of 155 or better. The draft included Kevin Mawae, Sam Adams, Carlester Crumpler, Larry Whigham and Lamar Smith. The 1996 Ravens picked Jonathan Ogden and Ray Lewis, whose career values were high enough to make up for the picks of Dexter Daniels and Jon Stark.
Top
1994 SEA 296 1996 BAL 227 1983 MIA 205 1993 NYG 199 1993 NOR 193 1977 OAK 180 1997 SEA 177 1995 TAM 175 1999 NOR 174 1993 NWE 171Worst
1970 KAN 19 1973 SFO 16 1980 SDG 14 1971 MIN 13 1975 STL 12 1977 WAS 10 1978 CHI 9 1989 RAI 7 1976 WAS 5 1975 KAN 1What colleges have provided the most NFL value? I don't think it will come as any surprise to see USC at the top of this list, and by a wide margin. The only school that surprises me on this list is Texas A&M at #9. A&M's top contributors were Ray Childress, Lester Hayes and Richmond Webb.
Top Colleges
College/Univ Pts USC 24,463 Penn State 18,453 Notre Dame 16,943 UCLA 15,557 Ohio State 15,143 Miami (FL) 15,066 Nebraska 14,584 Michigan 13,952 Texas A&M 13,590 Florida 12,714 Pittsburgh 12,663 Tennessee 12,383 Colorado 12,117 Florida State 12,107 Alabama 12,043 Arizona State 12,001 Washington 11,580 Oklahoma 11,358 Auburn 10,632 Stanford 10,489The Baltimore Ravens only had 4 drafts in this study (1996-1999), but they made the most of it. Their average drafted player has provided over 117 points of career value. Of teams that have drafted every year of this study, the Rams have picked up the most value with 83 points per player. On the other hand, over a 30 year span, for every point of career value the Rams drafted, Kansas City only drafted 3/4 of a point. Most of the teams at the bottom of this list are not surprising. Lots of
teams with lots of bad seasons (Cleveland, New York Jets, Cardinals, etc.). But the one team that stands out is the Washington Redskins. The Redskins have played in 5 Super Bowls despite being one of the worst drafting teams in the league.Average Player Value
Team Average BAL 118 OTI 88 TAM 85 CAR 84 RAM 83 PIT 83 JAX 82 RAI 81 NWE 81 SFO 81 SEA 80 MIA 79 DAL 77 PHI 76 BUF 76 NOR 75 CIN 75 DEN 75 CHI 74 GNB 73 NYG 73 DET 72 CLT 71 MIN 70 ATL 70 SDG 70 CRD 70 NYJ 69 CLE 68 WAS 67 KAN 61This list is the average value of each draft for each team. The Pittsburgh Steelers have averaged 1,184 points of value from their drafts over the years. No surprise there. The biggest surprise is that the Houston Oilers have the 3rd-highest average draft value, despite having quite a few bad teams over the span, but they did manage their first Super Bowl appearance from the last batch of drafts.
Average Draft Value
Team Yearly Average PIT 1185 RAM 1061 OTI 1057 MIA 1056 DAL 1054 CIN 1045 NWE 1028 BUF 994 TAM 939 NYJ 916 CRD 907 NOR 905 GNB 905 SFO 898 CHI 891 CLT 890 RAI 887 NYG 887 BAL 882 SDG 866 DET 860 ATL 857 PHI 856 DEN 832 SEA 824 CLE 811 MIN 804 JAX 754 KAN 734 CAR 668 WAS 638Here is how much total career value each NFL team has drafted over the 30-year study:
Pittsburgh 35,542 LA/St Louis 31,831 Houston/Tennessee 31,719 Miami 31,671 Dallas 31,628 Cincinnati 31,341 New England 30,846 Buffalo 29,826 New Jork Jets 27,470 St Louis/Arizona 27,201 New Orleans 27,154 Green Bay 27,143 San Francisco 26,934 Chicago 26,720 Baltimore/Indianapolis 26,685 Oakland/Los Angeles 26,619 New York Giants 26,596 San Diego 25,966 Detroit 25,812 Atlanta 25,704 Philadelphia 25,676 Denver 24,950 Minnesota 24,130 Tampa Bay 22,528 Kansas City 22,022 Cleveland 21,884 Seattle 19,781 Washington 19,135 Jacksonville 3,772 Baltimore 3,526 Carolina 3,341What positions are the safest to draft? Here is the average career value from each drafted position. Fullback has been the most productive position by far. Lorenzo Neal and Sam Gash come out as the most valuable fullbacks of all time. There have only been 40 fullbacks drafted. I assume this is because many players who eventually play fullback in the NFL were classified as a RB or some other position in college and that's what they are listed at in their draft classification. Out of the 40
fullbacks listed, 30 of them have career values over 100. The nose tackle position is another that has a pretty high average value, but it is also caused by terminology. The last time a drafted player was designated as a nose tackle was 1991. The interesting part is that QB actually comes out as the least valuable position. I think part of it is because only 1 QB plays at a time in football, but teams are continuously trying to shore up the position. So they draft QB's, many of whom will rarely play. It's also interesting that the other skill positions - WR and RB are at the bottom of the list with QB.Avg by position FB 133 P 107 NT 100 C 88 K 86 DE 85 G 81 DB 78 TE 78 LB 78 T 72 DT 68 RB 66 WR 66 QB 63 WB 50 KR 0Here are the positions that have provided the most total value over the study.
By Position DB 132,544 LB 109,740 RB 86,667 WR 78,845 DE 66,271 G 56,721 T 52,495 TE 44,351 DT 39,014 QB 29,361 C 26,641 P 12,863 NT 12,431 K 11,779 FB 5,332 WB 100 KR 0Since there are typically 4 DB's on the field at any time, and only 1 QB on the field, it's unfair to say that DB provides more value than QB. In fact, we would expect that DB would provide 4 times the total value as QB because of this. So, I normalized the total value by position to look at total value per player who typically sees the field in a game. RB comes out as the most valuable position.
Normalized RB 86,667 TE 44,351 WR 39,423 DB 33,136 DE 33,136 LB 31,354 QB 29,361 G 28,361 C 26,641 T 26,247 DT 19,507 P 12,863 NT 12,431 K 11,779 FB 5,332
This entry was posted on Wednesday, September 17th, 2008 at 5:18 pm and is filed under History, NFL Draft. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

I'm a bit confused as to what the bottom is measuring... Most starters in the NFL is drafted at some point , so if you just went by games started, these numbers should be roughly equal, right? I'm missing something...
Any methodology that does not end up grading Ryan Leaf as one of the ten biggest busts since 1993 has to be fatally flawed...
I've always wondered why Gale Sayers is in the Hall of Fame. Even giving him the full 200-point bonus (he was All-Pro his for all five of his significant seasons), he only rates out with 378 points. That would make him the fifth-worst HOF-er on your list.
Mattieshoes, I'm not quite sure what your question is asking.
I'm not even sure what I'm asking
I think it's a methodology thing. Something seems hinky.
If I'm following this correctly, ALL first overall picks have an "expected" value of 1, so they maximum actual value of zero since they can't do better than first in their draft class. Conversely, just about any 7th rounder who manages to PLAY in the NFL would have a huge positive value simply because some of the hundreds in front of him never got into a game.
Would it make more sense to figure out an average value for a given draft position, then compare the player's value to his slot's average value? Otherwise, you just seem to be penalizing early picks and... err, what's the opposite of penalizing? Righto, I'll make one up. ... and bonusing late picks?
Well, except that this is supposed to study teams who drafted well. If teams were 100% accurate at drafting, then the #1 pick would be the most valuable, and the 255th player the 255th most valuable. Basically, it's comparing players to their Platonic ideal value, if I may misuse the word "Platonic" in a hopefully not too confusing manner, instead of their average slot value. I'd agree that it probably is penalizing individual players - after all, it does seem a little harsh to actually expect the #1 draft pick to always be the most valuable player in the draft.
But the methodology for ranking teams seems sound enough, since it doesn't actually include draft position at all, and I agree with the basic premise - you draft people to play football, if they do that at all, even poorly, you got something out of it, especially when you consider how many draft picks never play a game. I definitely would take the individual player rankings with a large grain of salt, and there does seem to be something very very funny going on with the positional rankings at the bottom - I suspect it has to do with players being drafted as RBs playing FB, leading to the RB category getting counted for nearly twice as many players as it should be.
The 200 point career bonus seems like it needs to be tweaked depending on how it was calculated. But the way I read it is that the lower the proportion, the fewer the points. But then given two players, who would you rather have?
Player A) Plays 10 seasons in his career and is All Pro in all of them
Player B) Plays 20 seasons in his career and is All Pro in 10 of them and just misses it in 10 other years. Player B would have a lower proportion and thus have a lower 200 point bonus, but is clearly much more valuable.
"A bonus of up to 200 points for the percentage of seasons designated as a starter that ended up being All Pro seasons."
I know Randy White was pretty good, but are you sure you don't have him mixed up with Reggie?
My theory with that bonus was to help balance out the guys who had short but awesome careers. Basically to differentiate between your player A and B above.
Without the bonus, player B would probably come out significantly ahead of player A. But I think a guy like Barry Sanders who was basically one of the top 3 RB's in the league for every one of his seasons deserved a little extra value. He never had any "hanger on" seasons like Emmitt Smith or Priest Holmes.
Also, remember the purpose of my player rating system was to come up with a quick single value to rate each player in order to compare across positions. I wasn't looking so much to create a list of greatest players (like Doug did with Approximate Value). I was looking for a way to evaluate draft picks.
In your example above, I think drafting either player A or player B is a great draft pick. In my system, one of them might have a better score, but either one will have a score that will probably evaluate them as one of the top 20 draft picks of their class.
In evaluating a team's ability to draft, there should be an end limit on length of career. I dont think anyone can predict a career lasting 20 years; to grade a team's draft better because they chose a guy who lasted 20 years instead of 14 is simply rewarding something that is luck and not skill.
I think it still seems flawed in terms of telling how well a team drafts.
Team A trades away draft rounds 1-3 for players each year. Team B trades away draft rounds 4-6 for players each year. Team A would show up worlds better because early picks are penalized.
Plus there's going to be a whole bunch of players with a value of zero, so anybody in the last round is going to show up with a positive value even if they never play a game because they'll be tied for #154 or some such even though they were drafted #200.
Also, this suggests picking #1 with the 20th pick is worse than picking #80 with the 100th pick. That doesn't feel right...
Although the Ravens have only been in existence since 1996, they are actually the continuation of the Cleveland Browns franchise (owned by Art "rot in hell" Modell) and should be combined with the pre-1996 Browns. The "new" Browns are an expansion franchise beginning in 1999 (and thus don't qualify for your study).
bmoore,
I just used the NFL designation that this website uses. This site considers the Baltimore Ravens and Cleveland Browns to be separate franchises. I believe the NFL officially recognizes them as separate franchises as well. (Which is how I wish all relocated franchises would be handled. The Rams name should still be available to Los Angeles if we ever get a new team. (I know - the Rams originally belonged to Cleveland, but we'll ignore that for now.))
I have a signed "old" black and white photograph of Gregory Westbrooks #52 - Playing on field for Oakland Raiders and NFL football player photo. I don't have a favorite team, but my boyfriend is a Bronco's fan. I was wondering if Gregory Westbrooks still play for the Oakland Raiders.
He is my neighbor and at 57years he is a little to old to play NFL football, however, he is a real nice guy!