Draft ranges

http://www.nationalfootballpost.com/draft-probabilities-by-position/

Tony Villiotti
Draft probabilities by position
Posted April 11, 2014 · Add Comment

In an earlier article the probability of achieving certain milestones by draft choice range was reviewed. To refresh your memory, the seven draft choice ranges are (1) the first 13 draft selections, (2) draft selections 14-24, (3) draft selections 25-46, (4) draft selections 47-73, (5) draft selections 74-114, (6) draft selections 115-187 and (7) draft selections 188 and higher.

It only makes sense, though, that the probabilities are different depending on playing position. These differences can be caused by some combination of a varying level of injury or scouting risk and the number of available positions for drafted players. For example, there is only one starting quarterback but five starting offensive linemen. It makes sense, then, that offensive linemen have a higher success rate than quarterbacks, especially given that a disproportionate number of quarterbacks are drafted.

In this article I will review the probabilities by playing position. For purposes of this article, in certain cases, playing positions were combined and others were split into two. The position groupings are:

• Quarterbacks
• Smaller Running Backs (210 pound or less)
• Larger Running Backs (over 210 pounds)
• Wide Receivers
• Tight Ends
• Offensive Linemen
-I do have individual data for centers, guards and tackles but the degree of position switching along the line makes it, in my opinion, more meaningful to report the information for the combined positions
• Small defensive ends and outside linebackers
-Small defensive ends were defined as less than 265 pounds
-This grouping acknowledges that defensive ends and outside linebackers are to some degree interchangeable, with a player’s professional position dependent on the type of defense his pro team plays
• Large defensive ends and defensive tackles
-Large defensive ends are defined as weighing 265 pounds or more
-This grouping acknowledges that 3-4 defensive ends are more like defensive tackles
• Inside linebackers
• Cornerbacks
• Safeties

Fullbacks and special teams players were excluded from the analysis due to the low number of draftees.

This article will address the playing position probabilities from two angles. First, I will address differences within the draft choice ranges for each of the first four ranges. This will take us into the beginning of the third round. Then, I will discuss observations by playing position. There may be some overlap between the two angles, but after much thinking about the subject I think it is the best way to proceed. The principle metric used in this review is the percentage of five-year starters, with the 1994 through 2008 draft classes used as the study period.

Selections 1-13
• Average % of five-year starters: 71.3%
• Among playing positions with at least 10 players drafted with these selections during the study period, the probability of drafting a five-year starter ranged from 60% to 95%
• Least risky: Offensive linemen (95%, or 20 of the 21 offensive linemen selected from 1994 through 2008 became five-year starters)
• Most risky: Large running backs (60.0% five-year starters), but quarterbacks (64.0%) and wide receivers (61.5%) were close
• Other observations:
-Over 51% of drafted offensive linemen made the Pro Bowl at least once compared to the 43.5% overall average
-80% of offensive linemen started as rookies compared to the average of 70%
-178 of 195 drafted players had careers of five years or more

Selections 14-24
• Average % of five-year starters: 63%
• Among playing positions with at least 10 players drafted with these selections during the study period, the probability of drafting a five-year starter ranged from 36% to 71%
• Quarterbacks, small running backs, tight ends, safeties and inside linebackers had too few selections to consider as part of the discussion
• Least risky: Offensive lineman with 71% of draftees becoming five-year starters
• Most risky: Large running backs with 36% of draftees becoming five-year starters
• Other observations:
-Biggest difference between 1-13 and 14-24 draft ranges is that 1-13 produces about 1.6 times as many Pro Bowl players
-Despite their overall higher risk profile, 32% of wide receivers drafted in this range were selected to the Pro Bowl at least once (versus 27% average)
-The outside linebacker/small defensive end group had only 16% of draftees selected to at least one Pro Bowl
-30% of drafted offensive linemen were selected to at least one Pro Bowl

Selections 25-46
• Average % of five-year starters: 48%
• Among playing positions with at least 10 players drafted with these selections during the study period, the probability of drafting a five-year starter ranged from 17% to 77%
• Least risky: Inside linebackers (76.5%), outside linebackers/defensive ends (64%), offensive lineman (61%)
• Most risky: Small running backs (16.7%), large running backs (23.5%) and defensive tackles/large defensive ends (35%)
• Other observations:
-Offensive linemen and inside linebackers were most likely to start as rookies
-Running backs and defensive tackles/large defensive end were least likely to start as rookies
-Only 58% of small running backs lasted five years in the versus the average of 73%
-Inside linebackers, safeties and tight ends were the most likely to play five years or more

Selections 47-73
• Average % of five-year starters: 34%
• Among playing positions with at least 10 players drafted with these selections during the study period, the probability of drafting a five-year starter ranged from 8% to 49%
• Least risky: Offensive linemen (49%) and safeties (46%)
• Most risky: Quarterbacks (8%), small and large running backs (combined percentage of 11%) and wide receivers (25%) 
• Other observations:
-Offensive linemen were most likely to have five-year or longer careers
-Large running backs were least likely to last five years
-Offensive linemen and inside linebackers were most likely to start as rookies

The rest of the article discusses individual playing positions. The tables in each section include information about the probability of (1) a player having a five-year or longer career and (2) a player starting for at least five years. The table shows the average across all positions, the number of players drafted at that position from 1994 through 2008 (which is the same for both five-year careers and five-year starters) and the probability for that draft choice range.

Observations:
• 13% of players drafted with selections 1-13 were QBs with almost as many QBs taken with those selections as with selections 14-73
-If a team wants a top rated QB it better take him early
• The small number of data points makes it difficult to draw many additional conclusions
• The number of All Pro QBs is very low because Peyton Manning “hogged” most of those honors, holding down the total number of QBs to earn All Pro honors

Observations:

• Not many small running backs are taken with early selections
• There is a low likelihood of grabbing a starter at any point in the draft but players in this group often return kicks
• The probability of drafting a five-year starter is lower than the average for each draft choice range

Observations:
• Drafting running backs is a risky propositions and the percentage of five-year starters is below the average in every data choice range
• Only two of the 92 large running backs selected after the 114th pick ended up as a five-year starter
• The probability of having a five-year career is considerably lower for large running backs than the average of all positions

Observations:
• There is virtually no difference in the results of drafting wide receivers between draft selection ranges 1-13 and 14-24
• Despite having one of the lowest percentages of five-year starters with selections 1-13, wide receivers have the third highest number of rookie starters
• There is a large drop-off in the percentage of five-year starters after selection #24 but wide receivers taken with draft selection ranges 25-46 and 47-73 are about the same
• Selections between 115 and 187 have probably the lowest probability of success across the board

 
Observations:
• Only 10 tight ends were selected in the first 24 picks over the 15-year study period
• Out of those 10, nine became five-year starters
• There is a higher percentage of drafting five-years starters at tight end than the average of all positions through the 114th selection

Observations:
• The % of offensive linemen who end up as five-year starters exceeds the average in all seven of the draft choice ranges
• The probability of an offensive lineman earning Pro Bowl honors at least once is significantly higher than the average through selections 46, but then drops below the average
• Offensive linemen, when all performance measures are considered, are probably the list risky position to draft regardless of draft position
• One-third of all five-years starters from picks 188 and higher are offensive linemen

Observations:
• The probability of drafting a five-year starter is pretty much the same through the first 46 selections of the draft
o The chances of drafting a rookie starters are about the same as well
• The probability of drafting a Pro Bowl player does drop off considerably after the first 13 selections
o The probability for selections 14-46 is below the average of all positions
• Outside Linebackers/Small Defensive Ends have the highest percentage of five-year careers with selections 115-187

Observations:
• 40 of 54 draftees with selections 1-13 were rookie starters, the most at any position
o This is based on a 1994-2013 study period
• There is a big drop-off in the probability of drafting five-year starters from choices 14-24 to 25-46
o 14-24 was exactly at average while 25-46 was considerably below average
o Defensive tackles/large defensive ends had the lowest percentage of five-year careers from selections 14-24
• There is another drop-off from 47-73 to 74-114, with 74-114 being considerably below average
• There is a better than average chance of drafting a five-year starter late in the draft (after selection 114)

Observations:
• Drafting inside linebackers has been a very safe selection pretty much anywhere in the draft
• All 10 selections in the first 24 picks became five-year starters and practically all have earned Pro Bowl honors
• None of the 32 inside linebackers selected after the 187th pick have become five-year starters

Observations:
• There has been very little difference between cornerbacks selected 1-13 and 14-24
o 14-24 has a higher % of five-year starters while 1-13 has a higher % of Pro Bowl selections
• Cornerback selections after the 73rd pick have not done well
• As a general rule, take cornerbacks earlier rather than later

Observations:
• Teams tend to stay away from safeties early (although that has been changing in recent years)
• Safeties have been a low-risk pick
• The probability of drafting a five-year starter has been higher than the average of all draft choice ranges in all groups except the 74-114 draft selections
• Safeties have the highest percentage of players with five-year careers from selections 188 and higher

Follow Tony on Twitter @draftmetrics

Probably more information than you ever wanted to know. Taking coffee break. 😉

http://walterfootball.com/nfldraftology408_1.php

Matt McGuire’s NFL Draftology 408: <br> The 10-Year NFL Draft Study: Round 1

Send Matt an e-mail here: mmcguire5@gmail.com
All other e-mail, including advertising and link proposals, send to: wpc112@gmail.com

Updated July 12, 2010

The first round can make or break a team’s draft. This is where the top prospects fall most of the time. Casual fans watch the first round of the NFL Draft and then tune it out. However, have the expectations for how successful teams should be drafting in the first round become too high now?

I think they are based on my research in the NFL Draft Study. In the NFL Draft broadcast, sure a couple prospects are talked about as being overrated, but consistently in my study at least 10 players busted in the first round each time and we should expect this in the future. This means at least every third player will bust in the first round, and in some past drafts half the players busted in the first round.

Fans can’t comprehend that there is at least a 33-percent chance of their 2010 first-round pick busting, but it’s true (33% – giving the benefit of the doubt that the scouting in 2010 is better than in recent years).

Here are my numbers for the NFL Drafts from 1998-2007:

image:
Busts in the NFL Draft

There are a number of ways you can use this. Go back into the 2009 and 2010 NFL Drafts, and you can scope out 10 or more players whom you think have a very legitimate possibility of busting. It’s annoying how some analysts (i.e. Jon Gruden) talk about every prospect like they will pan out, but the reality of the situation is that a lot of these players will bust.

Why is this? Is scouting this bad? I don’t think the scouting is bad, but I feel bad drafts are certainly made for teams.

Even though teams invest hundreds of thousands of dollars into their scouting department and spends thousands of hours evaluating prospects from January the April, the reality of the situation is that this is simply an extremely difficult process. It’s really hard to figure who will thrive in the NFL and who won’t.

Is it a crapshoot? It’s a little bit of a crapshoot because teams can only do so much and then players have to take some level of responsibility for their performance on and off the field. However, I think teams do make mistakes and while I feel like some of these busts can certainly be avoided, the bottom line is teams don’t want to deviate from the consensus on talent evaluation throughout the league.

There is good news. It looks as though evaluation is getting better and better as the percentages are going up. Is it this, or were the 2002 and 2003 NFL Drafts just so terrible that they skewed the stats? I think a little bit of both. Teams are doing more thorough background checks (often now to the point of over-analysis) and learning from some of the league’s mistakes. I am confident the percentages will improve in the future.

Here is how you can apply these statistics. For one, stop expecting every first-round pick – or even the vast majority – to be successful. Secondly, stop expecting your team to nail every first-round pick and give them the benefit of the doubt sometimes. Go back through your team’s drafts in the last five years. If your team only has one or no successful first-rounders (assuming you have five first-round picks – adjust accordingly if you have more or less) then you have a bad drafting team in the first round. Two is below average; three is solid; four or five is excellent.

You can access this 10-year NFL Draft study first round only (the six other rounds will be released eventually) by downloading it here. Every first-round pick from 1998 to 2007 is accounted for. If you have any comments or complaints about whether a player was a hit or a bust (and quite frankly this isn’t going to be perfect; I’m bound to be off on a couple) then please e-mail me.

Now I’m going to touch on some of the controversial players in my study as to whether I categorized them as a hit or bust, and explain why. Keep in mind this is an NFL Draft study – so even if the player presently has a successful career, he might have been a bust for the team that drafted him in his first 3-5 years in the league, then went on to success elsewhere.

Anthony Simmons, LB, Seattle, 1998: I thought the Seahawks made a good pick here with Simmons and he was productive enough to be a “hit” before succumbing to some bad luck with injuries. I have no idea if he had durability issues at Clemson, and if he did and this could have been more avoidable then he’d be categorized as a “bust.” Simmons started every game for the Seahawks from 1999 to 2001 racking up seven sacks, four interceptions, and 279 tackles. I think injuries really derailed his career and I’m not blaming the Seahawks here.

Shaun Williams, S, N.Y. Giants, 1998: Williams had a very promising career in New York before he suffered a season-ending knee injuries in 2003 and 2004. It is hard for me to label him a bust because he started every game for the Giants from 2000 to 2002 before having some bad luck.

Daunte Culpepper, QB, Minnesota, 1999: We all know who Culpepper is now and the argument can certainly be made he was a product of Randy Moss, Cris Carter and Dennis Green’s coaching with the Vikings. However, Culpepper did make the Pro Bowl three times, and he has more than 24,000 yards passing in his career. Who knows what would have happened if he hadn’t blown out his knee. He was very productive for the Vikings for five years. If you don’t like that he’s not a bust – tough. I think he was good enough to be a “hit.”

Thomas Jones, RB, Arizona, 2000: Jones spent his first three years in Arizona starting only 15 games there and only averaged 3.5 yards per carry. He had marginal success in Tampa, then has enjoyed a fairly successful career in Chicago and especially with the Jets. He has rushed for more than 2,700 yards with New York in the past two seasons. Despite all this, Jones was a bust for Arizona. This is an NFL Draft study, and he severely underperformed in Arizona and quite frankly I don’t think there are any excuses when you are drafted at No. 7 overall. Jets fans might not like this, but if Jones were enjoying success now in Arizona and he didn’t do anything for you if you drafted him in the top 10, then you would feel differently.

Leonard Davis, T, Arizona, 2001: Davis was drafted at No. 2 overall and expected to be an elite left tackle in the NFL. Davis couldn’t even handle playing offensive tackle in the NFL and has enjoyed success with the Cowboys at right guard. However, in this study we don’t account for positional changes. You don’t spend the No. 2 overall pick and the money coming with it hoping he works out at guard if he busts at tackle. If you are drafted as a tackle and you bust as a tackle, then you’re a bust in my book. You severely failed to live up to the expectations of the position you were drafted as.

Andre Carter, DE, San Francisco, 2001: Carter had an enormous 12.5-sack sophomore season in San Francisco and has 63.5 career sacks. He hasn’t lived up to the hype of being the No. 7 pick in the NFL Draft, but he hasn’t busted either. He’s a solid pass rusher though a bit inconsistent. I don’t think he’s a bust, but I can’t blame 49er fans for expecting a little more production in his time there.

Marc Colombo, T, Chicago, 2002: Colombo did nothing in Chicago because he had three years where injuries really held him back. He has enjoyed a lot of success with the Cowboys, though he missed a lot of games in 2009 due to injuries. Colombo has been in the NFL for eight years, but has only had three productive seasons. He has to be a bust.

Calvin Pace, DE, Arizona, 2003: Calvin Pace did absolutely nothing in his first four years in Arizona, then had a great season in his contract year and has been very good for the Jets. Arizona might have gotten screwed here. Pace had the talent all along, but he obviously lacked the work ethic and competitiveness to be great for the team that drafted him. He’s not a bust with the Jets now, but for his first four years he did nothing – and the drafting team that invested the No. 18 pick in the NFL Draft got burned.

Robert Gallery, T, Oakland, 2004: See Davis, Leonard.

Alex Smith, QB, San Francisco, 2005: The jury is still out on Alex Smith, but for the moment I’m giving him the benefit of the doubt.

Cedric Benson, RB, Chicago, 2005: See Pace, Calvin.

Antrel Rolle, CB, Arizona, 2005: Rolle was a bust at corner in Arizona. He is now a solid safety, but he greatly failed to live up to expectations as a cornerback. Bust.

Mark Clayton, WR, Baltimore, 2005: While Clayton hasn’t been extraordinarily productive in Baltimore, he has been able to lock up the No. 2 receiver position for most of his career and hold off other receivers. Clayton is certainly in the gray area, and I think you can make the case he has been both a “bust” and a “hit,” but because he has been able to hold onto the No. 2 receiver job I gave him the benefit of the doubt.

The second round will be coming later in the week, so stay tuned and do your duty as a Draftnik and look back at past drafts!

Read more at http://walterfootball.com/nfldraftology408_1.php#rIBKPvqTkaC8oLS2.99

Matt McGuire’s NFL Draftology

Send Matt an e-mail here: mmcguire5@gmail.com
All other e-mail, including advertising and link proposals, send to: wpc112@gmail.com
NFL Draftology Lectures

NFL Draftology 408: The 10-Year NFL Draft Study (updated 7/16)

NFL Draftology 234: The NFL Positional Value Pyramid (updated 7/4)

NFL Draftology 101: Introduction to NFL Draftology

NFL Draftology 112: Need Versus Value

NFL Draftology 205A: Positional Demand

NFL Draftology 205B: Positional Impact

NFL Draftology 205C: Positional Impact Continued (Updated 6/9/10)

NFL Draftology 320: Dissecting the Trade Value Chart

NFL Draftology 321: Revising the Trade Value Chart

NFL Draftology 378: The Safest Pick Myth

Read more at http://walterfootball.com/nfldraftology.php#GrCfDDsAjKBL5yRj.99

Go here for a bunch of links if you wish to investigate further. The links are the same, whatever. 😉 http://walterfootball.com/nfldraftology.php