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.
ShareThisSome draft data
I'm going to be fairly busy for the next week or so. I will do my best to post something every weekday, but don't be too disappointed if it's just data without much commentary.
I mentioned last week that I didn't have as much draft data as I'd like, and I don't. But I realized that I do have some. What I don't have (in electronic form) is a complete record of every draft. What I do have is the draft round of every player who has played at least for the last few years. For most draft questions, the latter data set doesn't get you exactly what you want. It's generally more useful, to take one example, to know that X% of all 2nd-round running backs have ended up making the pro bowl than to know that Y% of all pro bowl running backs were taken in the 2nd round.
Nonetheless, I only have what I have. And it might be interesting to look at what percentage of the league's production in various categories has come from players drafted in each round.
First, look at the running backs. Here are the total scrimmage yards in 2005 by running backs drafted in each round:
Rnd TOTYD
==========
1 24070
2 13401
3 9554
4 8366
5 1339
6 3078
7 1190
X 8016
Just to make sure we're clear, what that says is that in 2005, running backs who were former first-round picks accounted for 24,070 rushing plus receiving yards. Running backs who were former second-round picks accounted for 13,401 total yards, and so on.
If it's true that there are generally roughly an equal number of running backs taken in each round, and if these 2005 numbers are representative of other years, then this means that an average first-round back can expect to gain about 80% more yards per year than an average second-round back. I'm not sure how close to true those ifs are, and I don't want to speculate, but maybe some kind soul will hop over to drafthistory.com and investigate the first one for us.
Here it is for quarterbacks. All passing plus rushing yards for QBs, by draft round:
Rnd PYD
==========
1 48322
2 11697
3 6822
4 7069
5 4500
6 12313
7 6095
8 4096
9 1938
11 28
X 13162
All receiving yards by wide receivers, by draft round:
Rnd RECYD
==========
1 24167
2 16866
3 8534
4 6807
5 3421
6 972
7 4061
8 466
12 917
X 7875
All tackles recorded by defensive linemen:
Rnd TKL
==========
1 1713
2 869
3 692
4 599
5 510
6 379
7 383
X 858
All tackles recorded by linebackers:
Rnd TKL
==========
1 1440
2 1453
3 1260
4 1140
5 587
6 349
7 425
X 1579
All tackles recorded by defensive backs:
Rnd TKL
==========
1 2025
2 2665
3 1218
4 1159
5 658
6 541
7 392
X 1798
This entry was posted on Tuesday, May 2nd, 2006 at 4:35 am and is filed under NFL Draft. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
I took a look at the drafts from 2000 through 2005 (6 years). Here is how many players were drafted at each position, by round:
Thanks, Richie!
My first reaction is surprise at how few linebackers go in the first round. 46 defensive lineman, 32 defensive backs, and only 14 linebackers? Interesting.
That formatting sucked...
Now here is the average yards per draft pick for each round:
QB
Rnd Yards
1....3020
2....3899
3.....758
4.....785
5.....409
6.....724
7.....339
RB
Rnd Yards
1.....1337
2.....1117
3......531
4......270
5.......84
6......205
7.......38
WR
Rnd Yards
1....806
2....602
3....388
4....324
5....110
6.....31
7....107
DB
Rnd Tackles
1...63
2...54
3...28
4...25
5...15
6...12
7...10
DL
Rnd Tackles
1...37
2...26
3...25
4...19
5...18
6....8
7....7
LB
Rnd Tackles
1...103
2....54
3....36
4....41
5....21
6....22
7....16
DB's really jumped out at me. Every round consistently drafts a lot of DB's.
Look at the QB stats in the chart I just posted. 2nd round QB's actually were slightly more efficient in 2005 than 1st round QB's. The data is probably a little skewed since I only went back to the 2000 draft, and I bet most QB's who played in 2005 were drafted before 2000.
But, it kind of backs up the theory that there are a lot of 1st round busts at QB.
I looked a little deeper at QB. In the 20 drafts from 1986 through 2005, here were the number of QB's in each round:
1-46
2-16
3-27
4-37
5-16
6-36
7-43
Going back the extra 14 years, actually increases the rate of drafting QB's in the first round. From 2000-2005, 16% of all QB's drafted were taken in the first round. From 1986-2005, 21% of all QB's drafted (in the first 7 rounds) were taken in the first round.
Each of the first 4 rounds actually has a higher rate when I go back 20 years.
It seems like a lot of DBs are always drafted because there an even bigger crapshoot than normal. Does that sound plausable to you Richie?
It seems like DBs are very hit or miss in the NFL, and most teams figure if they take a solid crop of them every year, one or two should pan out (see: Nathan Vasher, 4th round).
Sure, it sounds plausible. Also, DB is a more broad term than any of the other positions. Each team "starts" 4 DB's. Of the other positions, only LB "starts" as many as 4 players. Then there are nickel and dime defenses, so that each team probably has more DB's on their roster than any of the other positions. But I'm just guessing.
Richie scooped me! I figured the distribution of draftees by position across rounds might not be uniform -- which could account for the discrepencies in production. This is most dramatic with 1st v. 2nd round QBs.
I've been known to carp about the hysteria surrounding top players prior to the draft -- stocks rising and falling with little apparent justification. Still, it appears that investigating players before the draft (thus playing a large role in determining where a player is taken in the draft) helps in predicting ultimate productivity as a pro. Gosh!
Thanks, Doug.
A lot of those numbers are skewed by a few players. The 7th round for WRs is up from the 6th because of TJ Houshmandzadeh and Donald Driver, and the 6th round for QBs is way up because of Brady and Bulger. Good stats though, kinda cool to look at.
I wanted to comment on two things.
One, on the running backs, it certainly appears as though if you are in need of help at running back you'd be ahead to take one in the first two rounds. In 2005, 11 of the top 15 rushers in the NFL were taken within the first 50 picks of the draft.
However, no team with a running back who was taken in the first half of the first round has won the Super Bowl in a really, really long time.
When you juxtapose those two criteria, it seems to me that you're best off taking a running back somewhere between 17-50. Look at the running back who were taken there this year: Maroney, Williams, White.
Second is about quarterbacks and draft position. Over the last decade or so, and excluding 2004 and 2005 (because it's just too early to tell with those guys) 23 QBs have been taken in the first round. The ratio of decent QBs to busts is about 1:1 and there have been twice as many busts as there have been franchise-type QBs.
Every QB who was taken in the first round started at least one game by his second season with one exception (Chad Pennington). The QBs who have gone on to have good careers have generally established themselves by their second season.
Moral of the story: If you need a QB, go ahead and take one in the first round but understand it's a crap shoot. But if it's his second year and you've still got question marks about him, then you might as well go ahead and take another QB in the first round again. Guys who are still shaky their second year rarely become quality NFL QBs.
If you're a good team, drafting in the back half of the round, take a RB with your first pick. If you're a bad team, drafting towards the front of the round, you can wait until the second round for that.