Hello BTB. I have been an avid reader for a couple of years now, but this is my first post.
First off, I think this ground may have already been covered, but I haven't been able to find it through searching on BTB - if it is already covered, then I apologize.
I have used the Pro-Football-Reference.com draft database to analyze how well teams have drafted from 1999-2012. A common meme about the Cowboys is that they draft poorly. What the data shows refutes this hypothesis in part, but also, sadly for me as a die-hard fan, corroborates it.
The PFR data set contains a Career Approximate Value and Years as a Starter for each player. These metrics can be used to rank players by their impact and to compare a player's impact with where they were drafted and by which team. The data is not perfect, but using the Career Approximate Value as a ranking tool gives a consistent, objective way tot evaluate whether a draft pick was a success or not.
The first time period I looked at was 2002-2012, chosen because 2002 was the year that Houston entered the league so the data set is "clean" from the perspective of having all 32 teams in it. The ranking may surprise a lot of people:
Rank | Team | Total Career Approximate Value | Total Number of Starter Years |
1 |
DAL |
1198 |
139 |
2 | GNB | 1198 | 125 |
3 | SDG | 1196 | 154 |
4 | NWE | 1176 | 127 |
5 | BAL | 1153 | 145 |
6 | TEN | 1148 | 134 |
7 | CAR | 1145 | 132 |
8 | CHI | 1134 | 127 |
9 | NYG | 1119 | 125 |
10 | SFO | 1117 | 153 |
11 | PHI | 1115 | 131 |
12 | HOU | 1101 | 127 |
13 | IND | 1097 | 133 |
14 | ATL | 1090 | 115 |
15 | PIT | 1087 | 121 |
16 | JAX | 1078 | 145 |
17 | ARI | 1041 | 132 |
18 | CIN | 998 | 119 |
19 | NYJ | 998 | 117 |
20 | MIN | 980 | 112 |
21 | DEN | 976 | 101 |
22 | NOR | 976 | 116 |
23 | BUF | 966 | 113 |
24 | OAK | 958 | 132 |
25 | SEA | 958 | 120 |
26 | CLE | 941 | 123 |
27 | KAN | 907 | 120 |
28 | STL | 854 | 111 |
29 | MIA | 853 | 120 |
30 | DET | 808 | 103 |
31 | TAM | 701 | 90 |
32 | WAS | 653 | 71 |
Dallas ties with Green Bay as the most successful drafting team over the past decade, at least as measured by the contributions of the players drafted by us. It is important to note that a player is linked to whichever team drafted him, not where he plays later. Jared Allen was drafted by KAN, but has played for many years in MIN. He counts as a KAN draft pick for his entire Approximate Value.
There is a pretty obvious "eyeball" correlation between how well teams have fared over the past decade and how well they have drafted. There are outliers for sure - Pittsburgh looks mediocre on this measure of drafting prowess, but has been a top team.
The other, painful, outlier is Dallas. It is difficult to stomach that we have managed to draft so successfully over a relatively long time, but not been able to consistently win. To test a hypothesis that draft consistency is related to win consistency, I sliced the data set over a few periods.
Draft prowess ranking by team 1999-2012:
Rank | Team | Total Career Approximate Value | Total Number of Starter Years |
1 | GNB | 1778 | 211 |
2 | NWE | 1682 | 180 |
3 | CHI | 1680 | 202 |
4 | IND | 1666 | 214 |
5 | PIT | 1652 | 199 |
6 | BAL | 1640 | 211 |
7 | SDG | 1615 | 200 |
8 | CAR | 1609 | 193 |
9 | SFO | 1596 | 214 |
10 | NYJ | 1584 | 198 |
11 | TEN | 1577 | 192 |
12 | ARI | 1567 | 215 |
13 | ATL | 1559 | 189 |
14 | PHI | 1553 | 183 |
15 | BUF | 1508 | 187 |
16 | DEN | 1495 | 172 |
17 | NYG | 1489 | 178 |
18 | JAX | 1478 | 203 |
19 | CIN | 1461 | 189 |
20 |
DAL |
1443 |
166 |
21 | SEA | 1428 | 176 |
22 | MIN | 1345 | 159 |
23 | CLE | 1343 | 172 |
24 | STL | 1320 | 180 |
25 | NOR | 1296 | 157 |
26 | OAK | 1175 | 182 |
27 | DET | 1174 | 157 |
28 | KAN | 1135 | 144 |
29 | HOU | 1101 | 127 |
30 | MIA | 1094 | 152 |
31 | WAS | 1015 | 118 |
32 | TAM | 989 | 133 |
What the above table tells me is that the 1999-2001 period for Dallas must have produced some historically bad drafts to pull us down from a tie for #1 from 2002-2012 to #20 over the 1999-2012 period. Below are the results for 1999-2002 (this contains the whole Campo era, 2000-2002, as well as the last year of Gailey in 1999) period:
Rank | Team | Total Career Approximate Value | Total Number of Starter Years |
1 | PIT | 800 | 108 |
2 | IND | 723 | 103 |
3 | NYJ | 719 | 101 |
4 | GNB | 709 | 98 |
5 | BAL | 697 | 95 |
6 | PHI | 695 | 86 |
7 | CAR | 691 | 85 |
8 | CHI | 690 | 92 |
9 | BUF | 681 | 89 |
10 | DEN | 665 | 84 |
11 | NWE | 653 | 71 |
12 | JAX | 590 | 81 |
13 | TEN | 590 | 80 |
14 | SEA | 581 | 63 |
15 | SDG | 573 | 71 |
16 | SFO | 567 | 77 |
17 | ARI | 566 | 88 |
18 | CIN | 560 | 84 |
19 | CLE | 549 | 70 |
20 | ATL | 536 | 80 |
21 | STL | 528 | 80 |
22 | MIN | 484 | 65 |
23 | DET | 471 | 69 |
24 | NYG | 469 | 65 |
25 | NOR | 464 | 60 |
26 | WAS | 453 | 55 |
27 |
DAL |
419 |
50 |
28 | OAK | 386 | 74 |
29 | TAM | 325 | 49 |
30 | MIA | 308 | 44 |
31 | KAN | 302 | 37 |
32 | HOU | 218 | 32 |
Yikes, that is bad. 27th out of 31 teams (we should ignore HOU as it only has one year in this data set) stinks. We also know, qualitatively, that 2007-2010 was a pretty bad period, and can see this quantitatively below:
Rank | Team | Total Career Approximate Value | Total Number of Starter Years |
1 | ATL | 474 | 45 |
2 | GNB | 415 | 40 |
3 | SFO | 414 | 44 |
4 | MIA | 357 | 52 |
5 | BAL | 353 | 36 |
6 | DET | 350 | 42 |
7 | KAN | 348 | 47 |
8 | PIT | 343 | 35 |
9 | PHI | 339 | 37 |
10 | CAR | 338 | 35 |
11 | TEN | 333 | 35 |
12 | TAM | 322 | 37 |
13 | HOU | 321 | 30 |
14 | NYG | 315 | 33 |
15 | BUF | 314 | 33 |
16 | OAK | 314 | 38 |
17 | CIN | 313 | 30 |
18 | SEA | 313 | 36 |
19 | MIN | 311 | 31 |
20 | JAX | 308 | 43 |
21 | DEN | 307 | 35 |
22 | NWE | 303 | 25 |
23 | ARI | 294 | 32 |
24 | IND | 290 | 31 |
25 | CHI | 285 | 33 |
26 | NYJ | 277 | 27 |
27 | NOR | 265 | 26 |
28 | STL | 254 | 25 |
29 |
DAL |
253 |
26 |
30 | CLE | 248 | 36 |
31 | SDG | 227 | 27 |
32 | WAS | 175 | 20 |
Horrible - even worse than 1999-2002. So the Parcells era 2003-2006 must have been very good indeed to offset these bad bookends, and it was:
Rank | Team | Total Career Approximate Value | Total Number of Starter Years |
1 | SDG | 782 | 100 |
2 |
DAL |
717 |
86 |
3 | NYG | 679 | 79 |
4 | CHI | 674 | 73 |
5 | NWE | 660 | 80 |
6 | ARI | 637 | 89 |
7 | GNB | 599 | 73 |
8 | IND | 599 | 75 |
9 | TEN | 581 | 70 |
10 | SFO | 567 | 91 |
11 | JAX | 543 | 75 |
12 | NYJ | 542 | 67 |
13 | BAL | 540 | 77 |
14 | NOR | 538 | 69 |
15 | CIN | 518 | 69 |
16 | CAR | 514 | 67 |
17 | ATL | 500 | 61 |
18 | HOU | 495 | 60 |
19 | STL | 489 | 69 |
20 | MIN | 487 | 57 |
21 | PIT | 481 | 55 |
22 | PHI | 464 | 56 |
23 | BUF | 450 | 56 |
24 | CLE | 449 | 56 |
25 | KAN | 449 | 56 |
26 | DEN | 448 | 46 |
27 | SEA | 436 | 70 |
28 | OAK | 434 | 65 |
29 | MIA | 371 | 50 |
30 | DET | 323 | 43 |
31 | WAS | 297 | 38 |
32 | TAM | 280 | 39 |
And the two most recent drafts, though it is still very early days and so these numbers must be taken with a huge grain of salt, look ok (but not great, yet):
Rank | Team | Total Career Approximate Value | Total Number of Starter Years |
1 | SEA | 98 | 7 |
2 | CLE | 97 | 10 |
3 | WAS | 90 | 5 |
4 | DEN | 75 | 7 |
5 | TEN | 73 | 7 |
6 | ARI | 70 | 6 |
7 | CIN | 70 | 6 |
8 | HOU | 67 | 5 |
9 | CAR | 66 | 6 |
10 | NWE | 66 | 4 |
11 | BUF | 63 | 9 |
12 | MIN | 63 | 6 |
13 | TAM | 62 | 8 |
14 | MIA | 58 | 6 |
15 | GNB | 55 | 0 |
16 | PHI | 55 | 4 |
17 |
DAL |
54 |
4 |
18 | IND | 54 | 5 |
19 | BAL | 50 | 3 |
20 | ATL | 49 | 3 |
21 | STL | 49 | 6 |
22 | SFO | 48 | 2 |
23 | NYJ | 46 | 3 |
24 | OAK | 41 | 5 |
25 | JAX | 37 | 4 |
26 | KAN | 36 | 4 |
27 | SDG | 33 | 2 |
28 | CHI | 31 | 4 |
29 | DET | 30 | 3 |
30 | NOR | 29 | 2 |
31 | PIT | 28 | 1 |
32 | NYG | 26 | 1 |
So is our inconsistent performance on the field highly correlated with previous inconsistency in draft classes? I think this must have a lot to do with it. Swinging from high to low like this cannot be the recipe for success. If you look at the above and find some teams that draft well (e.g. GNB, BAL), they seem to do so no matter what period we look at - they also seem to perform pretty consistently on the field. Without the Parcells era, we would have likely been consigned to the fate of some of the other teams that seem to consistently fall at the bottom of the above tables as well as the W/L column: OAK, TAM, WAS, MIA, KAN to name a few.
It seems we need some patience in a coach and personnel evaluation team for them to work for a few years to get and keep things on the right path. I think we may have that with Garrett & Co., but we won't know for sure for another 3-4 years as the career arcs of the 2011-12-13 draft classes become known.
The above data also seems to rule out clearly that some teams are just "lucky" in the draft. Because teams draft in reverse order of success, one might expect that teams that do well will subsequently have worse drafts, but this doesn't seem to be borne out. Teams that do well, draft well and seemingly do so no matter if they are picking 1 or 32.
I hope you found the above interesting. Depending on feedback I will post more in the future, particularly on the interplay between Combine / pro day performance and player success. I believe that measurables matter a whole lot more than many casual fans and media types would like to believe.