During the last 10 years, there have been 319 first round picks (Patriots forfeited one of their picks because of Spygate). Using data from Wikipedia that includes the player's position and whether the player has made a Pro Bowl, I've done some data mining.
|Picks 1-8||Picks 9-16||Picks 17-24||Picks 25-32|
|Position||Pro Bowl||Total||Pro Bowl %||Pro Bowl||Total||Pro Bowl %||Pro Bowl||Total||Pro Bowl %||Pro Bowl||Total||Pro Bowl %|
|Center||- -||- -||- -||0||1||0%||3||3||100%||1||3||33%|
|Guard||- -||- -||- -||- -||- -||- -||1||3||33%||2||3||67%|
My first chart is a break down, by player position, of the number of Pro Bowl players for a position ("Pro Bowl"), the number of players picked for a position ("Total"), and the percentage of players who were picked that made a Pro Bowl ("Pro Bowl %"). I also split the data out into the Picks 1-8, Picks 9-16, Picks 17-24, and Picks 25-32 of the first round.
|Position||Pro Bowl||Total||Pro Bowl %||Avg. #||Expected Picks||Reach %|
My second chart is a breakdown, by player position, of the total number of Pro Bowl players for a position ("Pro Bowl"), the total number of players picked for a position ("Total"), and the percentage of players who were picked that made a Pro Bowl ("Pro Bowl %"). The last three columns of the second chart need a little more explanation.
The "Avg. #" column is the number of players you would normally expect to play at a given position. Since there are both 3-4 and 4-3 defenses, I split a player between DT and LB. Also, nobody picks FB so the total number is 21. The WR number could be a little higher because 3 WR sets are common, but I didn't know what would be a good number so I let it be.
The "Expected Picks" column is the number of picks you would expect if all the positions were valued equally and for every 21 picks, on average, you would draft a complete defense and complete offense (minus the fullback).
The "Reach %" (or Reach Percentage) column is the percentage difference between the "Expect Picks" column and the total number of picks that were actually made at a particular position. A positive Reach Percentage means that a player position was overrepresented in the first round and a negative Reach Percentage means that a player position was underrepresented in the first round.
In looking at the data and based upon my own observations of the draft, I believe a common failure among GMs centers on the observation "just because you take a guy in the first round doesn't make him worthy of a 1st round pick." Many times, GMs/teams say to themselves, "we need to get a 1st round pick at [X] position because [X] position is important."
As the data bears out, there are only so many draftable "Pro Bowl" caliber players in any given year. However, what appears to happen is that when the time comes for the team to draft (unless they have a really high pick) what is left are guys that have some flaws and teams rationalize picking these players in the belief that if they were the best player available at X position then they are a first round caliber pick. In listening to the NFL Network's coverage this weekend, on Saturday, Mike Mayock made the point (and I'm paraphrasing) that 'teams will overreach to get guys at the premier positions.' That comment is born out time and time again from the data.
QB & RB
It should be no surprise that QBs and RBs have the highest Reach Percentage. They are the two players that have the ball in their hands for the longest time. They are typically the faces of the franchise and are scrutinized like no other players. As a result, QBs and RBs are selected almost twice as much as you would expect if all positions were valued equally.
By way of example, a total of 58 QBs and RBs were taken in the first round over the last 10 years. There is only a single QB and single RB on the field at any time (i.e. 2 players), which is the same as the number of offensive guards on the field on any given play. However, only 6 offensive guards were taken in the first round over the last 10 years.
Based upon the data, it looks like GMs do an average job selecting quality RBs and QBs as their individual Pro Bowl Percentage (36% and 30%, respectively) are close to the average Pro Bowl Percentage (i.e., 32%) for all positions. There are, however, some differences between the two.
Again, as no surprise, QBs tend to be taken very high in the first round (8 QBs were taken #1 overall in the last 10 years). However, QBs taken in the first half of the first round are far more likely to make it to the Pro Bowl (8/20 or 40%) than QBs taken in the second half of the first round (1/10 or 10%). As such, unless you are taking Aaron Rodgers in the back half of the first round, odds are that you are not going to get a top QB with picks 17-32.
RB picks are equally distributed through the first round. However, unlike QBs, the odds of getting a really good RB does not change when the RB gets picked. Although I don't have the data in front of me, I think it is common wisdom that you can get a quality RB almost anywhere in the top rounds of the draft. A few special players are picked high (e.g., Adrian Peterson), but you don't need a 1st round pick to get a good RB.
DE & DT
The next highest Reach Percentage positions are along the defensive line. Once a team has their "marquee players," they start working on the defense and the place most teams like to start is along the defensive line. This project originated from me doing research on defensive lineman drafted in the first round and their sack numbers. Since I've done more work on this particular position, I'm going to save my analysis on defensive ends and defensive tacks for another post.
Listen to draft coverage and the commentators frequently discuss a team's need to "get a franchise left tackle to protect the QB." The QB is the heart of the franchise and finding a great left tackle to protect the QBs blindside is so important that they made a movie about it. Besides the defensive line, the offensive tackle position has the lowest Pro Bowl Percentage of any position. Although the Reach Percentage is only 25%, that number is grossly understated because most of the offensive tackles being taken in the first round are expected to play the blindside (most typically left) tackle position.
As in common in many positions, a high Reach Percentage accompanies a low Pro Bowl Percentage. Taken in the first half of the first round, 38% (8 of 21) made it to the Pro Bowl. However, only 1 out of 17 offensive tackles taken in the second half of the first round made it to the Pro Bowl. This is a position where the really good ones go quickly and if you don't get a good one quick, odds are that you are not going to get a high quality one.
CB & WR
Obviously, these positions mirror one another, and they have similar (but not identical) Reach Percentages and Pro Bowl Percentages. From the data, it looks like once you get past the first 8 picks, your chances of getting a good WR drop significantly from 6/10 or 60% to 4/27 or 15% in the last 24 picks. Good corners seem to go deeper. You'll have a very good chance of getting a good corner in the first half of the first round (8/15 or 53%), but once the back half of the first round hits, the percentage drops significantly to 6/25 or 24%.
Linebackers are the first of the positions that I have discussed in which the Reach Percentage is negative (i.e., this position is underrepresented in the first round). However, like all other positions in which the Reach Percentage is negative, the Pro Bowl Percentage is higher than the total average (i.e., 32%).
One thing I did look at was whether the linebacker played outside or inside. Of the 12 Pro Bowlers, 7 played in the middle (i.e., ILB or MLB), 1 was an OLB in a 4-3 defense, and 4 were OLBs in a 3-4 defense. What this tells me is that inside backers picked in the first round are much more than likely to be a Pro Bowler than an OLB picked in the first round. Also, like most positions, once you get past the first half of the first round, you chances of getting a top notch players drop significantly.
G, TE, S, C
Such different positions yet I have clumped them all together. The reason why is pretty straight forward. These positions are almost always significantly underrepresented with picks in the first round yet when these players are picked in the first round, they have a much better chance of being a Pro Bowl-caliber player. Interestingly enough, not a single guard and only a single center have been taken in the first half of the first round in the last 10 years. FYI – that lone interior lineman was Mike Pouncey, who was picked last year (he didn't get selected to the Pro Bowl but he was named as an alternate as a rookie).
Many people advocate that the Cowboys should trade down to get more picks. Based upon the data, that is probably not a good idea. You have a 2 in 5 chance of getting a Pro-Bowl player in the first half of the first round. However, that drops to 1 in 4 when you select in the second half of the first round. If the Cowboys trade down, they'll be trading down into the part of the first round where finding a stud player can be really hard. I probably wouldn't have a problem trading down if the Cowboys were in the 20s, but at 14, the Cowboys still have a good chance at getting a very good player at 14.
The only mitigating factor would be if you planned on taking a C/S/G in the second half of the first round. Although they are positions of need for the Cowboys, I don't think there will be a center, guard, or safety worth taking in the second half of the first round. Peter Konz might fit the bill, but he has a incomplete grade because of what he didn't do at the combine.
I'm not a scout. I don't have the resources to evaluate each player with more sophisticated measures. Grading success/failure on whether a player makes a Pro Bowl is imperfect at best. Many players make the Pro Bowl that don't necessarily deserve to be Pro Bowlers. For example, both of the Roy Williams and Mike Jenkins made it to the Pro Bowl. Moreover, some deserving players don't get to go because they play behind some great players in their conference. Additionally, some of the players just recently drafted haven't reached their potential. That being said, given a large data set (i.e., 10 years of data), I believe things tend to even out. Also, the positions I have placed these players under are those that Wiki indicates.