≡ Menu

The best drafting teams from 2000 to 2007

In this post I derived the expected value of the contribution of each draft slot based on Pro-Football-Reference.com’s Approximate Value system. You can see the full draft chart here. Once you know what the expected value is for a draft pick, the next step to grading a draft pick is to measure how much actual value was provided. As before, I used the marginal Approximate Value generated by each player in each of his first five years, with the caveat that a player is only credited for his AV after his first two points of AV. Using that formula, Patrick Willis, LaDainian Tomlinson, and Maurice Jones-Drew come in as the three most valuable picks over that time period.

Since 2000, the team with the most amount of draft value in terms of picks was the 2002 Texans. Houston not only received the first pick in each round that year, but the expansion Texans were given several supplemental draft picks as well. The 2007 Raiders and 2000 Browns tied for the second mount amount of value in terms of raw draft picks, but both of those teams wound up with many more whiffs than hits.

The table below lists the best drafting teams from 2000 to 2007. I’ve also broken out each team’s AV above expectation for each year. If you click on any of the values in the columns from 2000 to 2007, you can see the players drafted by the team that year (one side effect of including this information: the columns do not sort correctly).


RankTeamTotal20002001200220032004200520062007
1SDG197-6.358.2-17.1-9.869.65646.40.4
2IND196.68.760.412.940.119.4-18.273.3-7
3JAX179.46.721.433.437.910.19.460.5-2.3
4NWE175.825.520.923.551.9-4.984.2-25.3-8.2
5GNB131.463-39.724.99.3-8.929.353.5-6.1
6PIT120.6-4.215.14320.319.117.69.727
7BAL97.221.42240.14.7-16.1-9.234.310.4
8DAL94.9-4.3-128.338.45.194.9-35.51.6
9CAR87.8-2.646.828.6-4.926.3-25.218.827.5
10PHI80.84.7-18.854.8-17.5-14.253.917.9-1.2
11NYG80.52.6-4.7122.429.631.2-1.631
12CHI79.946.5-23.4-0.835.931.9-25.915.7-11.3
13NOR78.7-5.811.76.6-26.510.90.181.7-1.5
14NYJ65.732.38.3-16.5-4.220.512.213.170.4
15ATL61.8-13.446.8-13.2-1.410.742.8-10.56.9
16TEN57.724.85.34.5-17-13.934.419.6-9.7
17SEA55.833.46.9-22.44.7-0.327.564.7
18BUF47.1-33.444.5-20.146.34.9-14.819.7-1
19DEN33.4-1.4-831.1-22.1-11.1-14.759.6-22.2
20WAS23-15.42.7-13.5-4.220.91.131.4-7.9
21CIN14.9-31.360.1-16.21-4-13.118.44.4
22HOU5.800-9.44.3-17.2-16.744.82.9
23SFO3.61124.6-14.2-5-29.95.911.262.5
24MIA2.63.38.617.5-19.99.72.5-19.1-19.4
25KAN-11.67.2-18.7-23.8245.2-14.48.9-1.5
26MIN-34.8-17.8-15-9.542.9-5.3-5524.919.8
27TAM-49.9-3.1-0.3-0.3-6.4-17.1-16-6.7-10.7
28ARI-51.4-28.3-21.4-34.421.254.5-26.3-16.7-9.6
29STL-60.1-13.1-1.7-18.2-12.424.72.4-41.8-27.2
30CLE-99.2-44.3-7.83-18-6.2-17.6-8.33.6
31OAK-100.9-20.7-17-17.7-4-26.8-7.1-7.6-42.3
32DET-112.3-14.323.4-27.2-17.4-6.9-40.9-29-26.7

The best year — both in overall value and in value above expectation — came from the 2005 Dallas Cowboys, when they drafted DeMarcus Ware, Marcus Spears, Kevin Burnett, Marion Barber, Chris Canty, and Jay Ratliff — giving them an incredible haul of starters, Pro Bowlers, and All-Pros.

It’s not hard to see why the Chargers fare so well here — they drafted LaDainian Tomlinson and Drew Brees (who had two big years in SD) in 2001, and then were outstanding in 2004 (Philip Rivers1, Igor Olshansky, Nick Hardwick, Shaun Phillips, Michael Turner – and that doesn’t even include Nate Kaeding), 2005 (Shawne Merriman, Luis Castillo, Vincent Jackson, Darren Sproles) and 2006 (Antonio Cromartie, Marcus McNeill, Jeromey Clary). And that doesn’t include picking up Antonio Gates as an undrafted free agent. Had they been able to keep all that talent (and had better coaching), they probably would have won a Super Bowl, but they did post the 5th most wins from 2002 to 2010.

The Lions have had years of bad drafts, but the 2005 one stands out as their worst. They grabbed Mike Williams with the 10th overall pick, and he had 449 yards and 2 touchdowns in two years. With the 37th pick they went back to USC and selected Shaun Cody, who started just 12 games in four years for Detroit. With a top-ten pick in the third, they drafted CB Stanley Wilson from Stanford two spots ahead of where the Giants grabbed Justin Tuck. To cap things off, they selected the star of the 0-16 season, quarterback Dan Orlovsky, in the fifth round.

Finally, one final note on this. You might notice that most of the teams seem to be doing pretty well in this study. That’s because the data used to derive these numbers go back to 1980; as it turns out, the years from 2000 to 2007 happen to be a very efficient period in NFL drafting history. I also confirmed this with friend of the site Danny Tuccitto, who has his own excellent draft article out today.

***Update: Pursuant to Topher’s request in the comments

TMYearValExp AVAct AVPksVAL/PkAct AV/Pk
nyj200770.429.6100417.625
nwe200584.229.811471216.3
dal200594.955.1150811.918.8
clt200673.330.7104710.514.9
nor200681.753.3135810.216.9
jax200660.531.592610.115.3
clt200160.438.69978.614.1
cin200160.156.911778.616.7
den200659.642.410278.514.6
sdg200556481047814.9
nyg200531.223.85547.813.8
crd200454.559.511477.816.3
sdg200158.256.811587.314.4
sfo200762.561.512496.913.8
phi200254.844.29986.912.4
htx200644.870.211576.416.4
sdg200469.668.4138116.312.5
min200342.947.19076.112.9
car200146.847.29485.911.8
sdg200646.437.68485.810.5
buf200346.342.78985.811.1
dal200338.451.69075.512.9
pit200243347785.49.6
atl200542.837.28085.410
was200631.417.64965.28.2
was200420.935.15645.214
nwe200351.952.1104105.210.4
chi200046.553.510095.211.1
clt200340.139.9808510
phi200553.954.1108114.99.8
gnb200653.574.5128114.911.6
gnb20006355118134.89.1
atl200146.863.2110104.711
car200426.330.75764.49.5
nyg200429.658.48874.212.6
jax200337.954.19294.210.2
gnb200224.930.15564.29.2
pit200320.331.75254.110.4
nyj200032.378.71118413.9
rav200240.137.9781047.8
chi200431.950.1828410.3
nwe200223.531.55563.99.2
det200123.441.66563.910.8
den200231.138.97083.98.8
nyg200731407183.98.9
sea200033.459.69393.710.3
jax200233.448.68293.79.1
buf200144.558.5103123.78.6
rav200021.456.67863.613
min200624.952.17773.611
ram200424.727.35273.57.4
mia200217.514.53253.56.4
car200727.548.57683.49.5
rav200634.348.783103.48.3
pit200727477483.49.3
car200228.663.49293.210.2
rav200122345673.18
oti200534.467.6102113.19.3
sfo200124.644.46983.18.6
sea200527.544.57293.18
crd200321.254.8767310.9
kan2003243862837.8
chi200335.972.11081239
gnb200529.348.778112.77.1
nwe200025.529.555102.65.5
oti200024.841.266102.56.6
min200719.850.27082.58.8
pit200419.144.96482.48
car200618.841.26082.47.5
cin200618.437.65682.37
chi200615.735.35172.27.3
phi200617.952.17082.28.8
pit200517.633.45182.26.4
buf200619.756.37692.28.4
pit200115.137.95372.27.6
clt200419.440.66092.26.7
jax200121.447.669102.16.9
nwe200120.954.175102.17.5
nyj200420.539.560102.16
nyg200322.437.6601125.5
nor200410.940.15161.88.5
oti200619.659.479111.87.2
nor200111.737.34971.77
mia20049.725.33561.65.8
clt200212.946.15981.67.4
atl200410.750.36171.58.7
nyj200512.233.84681.55.8
rav200710.435.64671.56.6
nyj20018.339.74861.48
nyj200613.174.988101.38.8
kan20068.936.14571.36.4
sfo200611.256.86891.27.6
clt20008.735.34471.26.3
jax20059.439.64981.26.1
pit20069.738.34891.15.3
gnb20039.325.735913.9
jax200410.163.9741017.4
sfo20001179901118.2
sea200662733615.5
mia20018.644.453915.9
dal20028.356.76590.97.2
kan20007.241.84980.96.1
buf20044.946.15160.88.5
oti20015.322.72870.84
kan20045.235.84170.75.9
nor20026.658.46590.77.2
phi20004.753.35870.78.3
dal20045.134.94080.65
cin20074.434.63970.65.6
atl20076.963.170110.66.4
jax20006.735.342110.63.8
sea20074.727.33280.64
sea20016.973.180120.66.7
mia20003.324.72860.64.7
was20012.736.33950.57.8
sfo20055.967.173110.56.6
sea20034.749.35490.56
cle20073.656.46070.58.6
oti20024.551.556100.55.6
htx20034.366.771100.47.1
rav20034.757.362110.45.6
mia20052.555.55860.49.7
htx20072.938.14170.45.9
cle20023515480.46.8
nyg20002.647.45070.47.1
ram20052.452.655110.25
dal20071.634.43680.24.5
was20051.142.94460.27.3
nyg20021394070.15.7
cin20031707190.17.9
sdg20070.435.63660.16
nor20050.143.944706.3
tam2001-0.335.335903.9
tam2002-0.314.314801.8
sea2004-0.338.338705.4
chi2002-0.838.8389-0.14.2
den2000-1.460.45910-0.15.9
buf2007-143427-0.16
phi2007-1.238.2378-0.24.6
ram2001-1.776.7759-0.28.3
jax2007-2.349.34711-0.24.3
nor2007-1.539.5387-0.25.4
kan2007-1.536.5357-0.25
nyg2006-1.638.6377-0.25.3
atl2003-1.417.4166-0.22.7
cin2004-4625811-0.45.3
car2000-2.638.6367-0.45.1
rai2003-4524810-0.44.8
pit2000-4.260.2569-0.56.2
gnb2007-6.148.14211-0.63.8
nyj2003-4.248.2447-0.66.3
car2003-4.951.9478-0.65.9
nwe2004-4.951.9478-0.65.9
tam2000-3.119.1165-0.63.2
nor2000-5.829.8249-0.62.7
min2004-5.344.3398-0.74.9
tam2006-6.737.73110-0.73.1
nyg2001-4.734.7307-0.74.3
sfo2003-536317-0.74.4
clt2007-745389-0.84.2
htx2002-9.485.47612-0.86.3
sdg2000-6.329.3238-0.82.9
cle2006-8.355.34710-0.84.7
dal2000-4.319.3155-0.93
nwe2007-8.224.2169-0.91.8
oti2007-9.745.73610-13.6
cle2001-7.859.8528-16.5
rai2005-7.143.1367-15.1
cle2004-6.241.2356-15.8
tam2003-6.423.4176-1.12.8
oti2004-13.959.94613-1.13.5
tam2007-10.763.75310-1.15.3
rai2006-7.648.6417-1.15.9
den2004-11.149.13810-1.13.8
ram2003-12.454.44211-1.13.8
det2004-6.958.9526-1.28.7
sdg2003-9.846.8378-1.24.6
chi2007-11.340.3299-1.33.2
rav2005-9.239.2307-1.34.3
den2001-840326-1.35.3
tam2005-16624612-1.33.8
dal2001-1234229-1.32.4
was2002-13.540.52710-1.42.7
min2002-9.551.5427-1.46
was2003-4.217.2133-1.44.3
phi2004-14.235.22110-1.42.1
sfo2002-14.235.22110-1.42.1
gnb2004-8.936.9286-1.54.7
was2007-7.928.9215-1.64.2
det2003-17.464.44711-1.64.3
kan2005-14.435.4219-1.62.3
atl2002-13.234.2218-1.72.6
atl2006-10.523.5136-1.82.2
min2000-17.849.83210-1.83.2
clt2005-18.242.22410-1.82.4
cin2005-13.141.1287-1.94
ram2000-13.136.1237-1.93.3
min2001-1541268-1.93.3
htx2004-17.243.2269-1.92.9
crd2007-9.647.6385-1.97.6
was2000-15.473.4588-1.97.3
mia2007-19.455.43610-1.93.6
buf2002-20.157.13710-23.7
kan2001-18.731.7139-2.11.4
sdg2002-17.163.1468-2.15.8
tam2004-17.134.1178-2.12.1
crd2001-21.475.45410-2.15.4
cle2005-17.658.6418-2.25.1
den2003-22.144.12210-2.22.2
mia2003-19.929.9109-2.21.1
rai2002-17.754.7378-2.24.6
atl2000-13.430.4176-2.22.8
sea2002-22.449.42710-2.22.7
ram2002-18.239.2218-2.32.6
rav2004-16.121.157-2.30.7
det2000-14.336.3226-2.43.7
rai2001-1732157-2.42.1
den2005-14.727.7136-2.52.2
buf2005-14.822.886-2.51.3
car2005-25.252.22710-2.52.7
nwe2006-25.348.32310-2.52.3
cle2003-1842247-2.63.4
cin2002-16.245.2296-2.74.8
crd2006-16.747.7316-2.85.2
htx2005-16.732.7166-2.82.7
oti2003-1734176-2.82.8
phi2003-17.536.5196-2.93.2
rai2004-26.860.8349-33.8
sfo2004-29.949.92010-32
det2002-27.256.2299-33.2
phi2001-18.840.8226-3.13.7
nyj2002-16.536.5205-3.34
det2007-26.769.7438-3.35.4
ram2007-27.241.2148-3.41.8
cle2000-44.382.33813-3.42.9
rai2000-20.734.7146-3.52.3
crd2000-28.353.3258-3.53.1
crd2005-26.351.3257-3.83.6
nor2003-26.549.5237-3.83.3
mia2006-19.124.155-3.81
rai2007-42.382.34011-3.83.6
chi2001-23.449.4266-3.94.3
det2006-2944157-4.12.1
buf2000-33.436.438-4.20.4
ram2006-41.855.81410-4.21.4
crd2002-34.449.4158-4.31.9
chi2005-25.946.9216-4.33.5
dal2006-35.540.558-4.40.6
cin2000-31.357.3267-4.53.7
kan2002-23.841.8185-4.83.6
den2007-22.237.2154-5.63.8
gnb2001-39.751.7126-6.62
det2005-40.943.936-6.80.5
min2005-556167-7.90.9
  1. I considered him drafted by the Chargers and Eli Manning drafted by the Giants for this study. []
{ 39 comments }
  • pmpmpm March 26, 2013, 7:12 am

    The Vikings had a brutal draft in 2005, headed by Troy Williamson at #7. As far as I can tell it’s your worst-ranked single year at -55, or am I missing something even worse?

    Reply
    • Chase Stuart March 26, 2013, 1:31 pm

      Yep, that would be the “winner” with CLE 2000 coming in second.

      Reply
    • Richie March 26, 2013, 1:38 pm

      The cool thing about Miami’s -19.7 in 2007 is that I think EVERY Dolphins fan knew it was a terrible draft as soon as they took Ginn. (Of course, many of us wanted Brady Quinn, and that might not have turned out any better.)

      Reply
  • Danish March 26, 2013, 8:12 am

    Teams get credit for only the years the draftee spent with the team? Ie. Brandon Marshall isn’t running up Denvers 2006 score?

    Speaking of 2006 Broncos draft: If Marshall and Cutler had stayed with the team, this could be a candidate to beating Dallas 2005, right?

    Reply
    • Richie March 26, 2013, 1:22 pm

      I think the team gets credit for the first 5 years of a player’s career – regardless of where they played.

      Reply
    • Chase Stuart March 26, 2013, 1:31 pm

      That is correct – the team is only given credit for the player’s time with the team.

      Reply
  • Danish March 26, 2013, 8:21 am

    Oh and as a bit of feedback, considering that the tables can’t be sorted, you could’ve included a best/worst table. Or is there a nifty trick I don’t know of with the search function?

    Reply
    • Chase Stuart March 26, 2013, 1:32 pm

      True. I suppose I could have just reproduced the table without the links, too. FWIW, the table *is* sortable but it sorts alphanumerically instead of treating the cells like numbers. There’s still some use to the sort feature.

      Reply
  • Richie March 26, 2013, 1:29 pm

    I think that limiting to the first 5 years of a player’s career ends up discounting teams that did well in drafting QBs. Most QBs (especially during this period) don’t provide much value in their first season or two. So where Jacksonville grabs MJD and gets excellent value from him as a rookie, New England drafts possibly the best player of the decade, but gets 0 value in year 1 (which would be ~20% of his total production).

    Of the top QB’s drafted from 2000-2007, almost all of them had limited or no production as rookies. I think Roethlisberger was the only one with 10+ AV as a rookie.

    Reply
    • Chase Stuart March 26, 2013, 1:34 pm

      That’s true, although I don’t think it’s enough to really swing too much. Even the ’04 Chargers look good and Rivers helps despite missing two years. But I hear you on the bigger point.

      Reply
  • Richie March 26, 2013, 1:46 pm

    Have you ever given each draft year a “rating” based on total value? You say 2000-2007 was a strong period. But what are the best years (and worst years) since 1980? How much variance is there between the best and worst years?

    Reply
  • Topher Doll March 27, 2013, 9:46 am

    I was wondering since you have this database if you have an average AV per pick. This is a very good method of judging a team since it takes into account ability to maneuver on draft day and such to get more picks, but I was interested to see which team had the best AV per pick. By looking at it this way I’d be interested to see which teams got the most per pick rather than teams that may have only drafted average by had a large number of picks.

    Reply
    • DB March 27, 2013, 10:51 am

      What Topher suggested would be cool. I’d like to add to that idea – how about some kind of metric that takes AV vs what would normally be expected at the particular draft position. That is, what kind of value should you expect out of the #1 spot and compare AV for the #1 spot for each player drafted there, and so on for the other draft slots.

      Reply
      • Chase Stuart March 27, 2013, 10:53 am

        I’m not following (unless the sarcasm is going over my head). That’s what this post does.

        Reply
        • DB March 27, 2013, 12:22 pm

          Nope – no sarcasm here, I like your blog too much. I think I poorly worded my previous comment. I’m re-reading your article, so I might be misunderstanding, correct me if I’m wrong. It seems like the metric in the table is a counting stat, so the greater the number of picks a team has the greater the opportunity to pad the score. I think you touched on this in your second paragraph about the Texans.

          Also, teams that have a lot of depth and no room on their rosters for rookies would be penalized. This years 49er’s come to mind, who I believe have the most picks in this upcoming draft. The Rivers/Brees situation also comes to mind here. Rivers initially didn’t have the opportunity, and later when Brees left for NO, he accumululated AV but not for the team that draft him. It seems to me that SDG would get penalized by the way the metric is calculated, even though by all accounts they drafted well here. I don’t know how to account for Rivers being on the bench, but perhaps allowing AV to be accrued to the drafting team regardless of whether draftee plays for his original team would make it more truly reflective of the title of this article.

          Reply
          • Chase Stuart March 27, 2013, 12:25 pm

            It’s my fault if the article isn’t clear: the chart is *not* a counting stat. It’s the measure of production *above* expectation. I just noted the Texans saying that some teams have a lot of draft value, hence the need to measure production relative to expectations. But I’m sorry if I wasn’t clear.

            If you draft a player and he does nothing for you but plays well for another team, that’s not a good draft pick, IMO.

            Reply
            • Richie March 27, 2013, 1:09 pm

              If you draft a player and he does nothing for you but plays well for another team, that’s not a good draft pick, IMO.

              Interesting. In a vacuum I would say that a situation where you traded a player (within his first 5 seasons) and he did well for another team shouldn’t really count against you (I guess depending on how much trade value you got). But then, in the NFL the chances of a team ever trading away a good player within his first 5 years is pretty slim. Usually what happens is a team cuts a guy within those first 5 years and then he does well someplace else. So in that case, he was a bad draft pick.

              But how man players are there who accumulated significant AV in his first 5 seasons with a different team? A few examples that come to mind:

              Jay Cutler (I think Denver probably got good value in trade, so I wouldn’t call this a bad pick by Denver)
              Brett Favre (why did they trade him? They thought they were set with Chris Miller?)
              Clinton Portis (Broncos got good value)
              Brandon Marshall (hmm…I’m detecting a theme)
              Wes Welker

              Reply
    • Chase Stuart March 27, 2013, 10:54 am Reply
      • Topher Doll March 27, 2013, 12:37 pm

        Not what I was thinking, though that and it’s sequel article are both fantastic. I was merely referring to taking a look at the total draft AV value for each team, creating a total for each team based on this metric than dividing it based on the number of draft picks that team had. So, for example, if a fictional Seahawks team had a total draft pick AV value of 150 with a total of 70 picks for an average AV of 2.14. Then compare that to a fictional 49ers team with a total draft AV value of 150 with a total of 80 picks for an average AV of 1.88. So while the two teams would be similar in terms of total AV it would be clear the Seahawks got more production PER PICK.

        Sorry if that wasn’t clear and thanks for the great work.

        Reply
        • Chase Stuart March 27, 2013, 12:40 pm

          Gotcha. I could probably do that. But would you want to know how much value over expectation per pick the team had or simply how much value per pick? I would think the former.

          Reply
          • Topher Doll March 27, 2013, 12:47 pm

            I think both would have their use, I do think the value over expectation would have more use. I have been doing most of this on my own but when you wrote this article I couldn’t help but ask since I assume that your database is more extensive than mine, which is pretty much PFR’s draft finder. If it’s too much work, I understand, the amount of content your provide is extensive.

            Thanks again.

            Reply
            • Chase Stuart March 27, 2013, 12:50 pm

              Thanks – I can’t promise I’ll get to it tonight, but it’s on the to-do list.

              Reply
              • Topher Doll March 27, 2013, 6:59 pm

                No worries at all, I was expecting to see it in a week or so, anytime sooner would be more than I could ask for, thanks.

                Reply
                • George March 27, 2013, 7:26 pm

                  Sorry just a random thought on this one – would there be any value in coming up with a league average AV per pick (as well as looking at the individual teams obviously) and that way you could put a number on who was constantly drafting better than average (so to speak) – and effectively was the amount that they were above (or below) average statistically significant so to speak (or could it just be accounted for by a blip e.g. a Jamarcus Russell – Oakland seem to average between 0 and -20, with a sharp downward spike on 07 which could be enough of a spike to move them from below average to significantly below average etc.).

                  Reply
                  • Chase Stuart March 27, 2013, 9:51 pm

                    Can you unpack that a little bit more?

                    Reply
                    • George March 28, 2013, 4:57 pm

                      When breaking it down by picks it kind of put a slight hole in what I was getting at (with regard to the Raiders anyhow). What I was getting at was what was the league average AV per pick? (I’m figuring it isn’t zero – but I am figuring I may not have 100% understood the metric). Are any teams consistently above the league average AV per pick? Are they statistically a significant amount above the league average AV e.g by at least a couple of points (conversely is there anyone consistently below the league average AV – e.g. Tampa Bay looks like it is zero or worse per pick every year – why? & why are Detroit even worse every year except 2001 which clearly looks like a blip – what was the blip in 2001 caused by).

                      When just looking at the team VAL without breaking it down per pick, it looked like 2007 was an even worse year than normal for the Raiders so I just figured the source of the blip was probably Jamarcus Russell. When breaking it down per pick you can see it wasn’t substantially different to 2004 or 2000 so not being aware of the Raiders management over that period but you’ve got to look at the fact they have been consistently below average in getting production from picks (if I’m making sense) and 2007 wasn’t an even worse than normal year.

            • Chase Stuart March 27, 2013, 10:21 pm

              Topher – I have updated the post.

              Reply

Leave a Comment