It’s April, that time of year where the draft becomes the central focus of everybody that loves football. And typically I kick off that part of the calendar by posting grades of some of past draft classes. Last year, I looked back at the 2009 draft. The prior year, I looked at the 2007 and 2008 drafts. This year, I will turn my attention to 2010.
The conventional wisdom states that individual drafts cannot be accurately graded until three years out. I’ve put a twist on that and have decided to wait five years.
Three years should rather be considered the minimum, while I believe waiting five years gives a far more accurate and complete picture of an NFL players’ career. By then, the majority of players should have had an opportunity to reach free agency and garner a second contract.
The perception of a player can change dramatically when waiting another two years to assess his impact on professional football. A very good example of this is former Atlanta Falcons defensive tackle Corey Peters.
If judging Peters from his first three years in the NFL, then the simple truth is that he wasn’t very good which can be evidenced by the very low positional grades on premium website Pro Football Focus. His overall grades were always in the negative, and both as a rookie in 2010 and in 2012, he was among the lowest-graded defensive tackles in the NFL. However in 2013, Peters’ fourth year, he really turned things around. Peters graded out at almost perfectly average, ranking 33rd among 69 defensive tackles. His overall grade went into the positive side of things this past year, ranking 34th among 81 graded defensive tackles, identifying him as an above average player.
Fellow Falcons defensive tackle Jonathan Babineaux is another perfect example of this. Babineaux spent the bulk of his first three seasons in the NFL as a rotational backup with the Falcons. He got his first opportunity to be a full-time starter in his fourth season (2008) and graded near the middle of the pack according to Pro Football Focus. Then in 2009, his fifth season, Babineaux was that site’s top-rated defensive tackle in the NFL. And Babineaux’s career since then has seen him rated at least in the middle of the pack if not significant higher among NFL defensive tackles.
Now that we can look back at a decade’s worth of play for Babineaux, it’s clear that had we waiting until his fifth season, we would have gotten a more accurate snapshot of the player he was and is. I don’t think Babineaux is the exception either, but rather much closer to the rule if most NFL players were put into that context.
You can even look at the opposite end of a spectrum for a player that was really good early in his career but fizzled as time wore on. One such was former San Diego Chargers linebacker Shawne Merriman. Merriman was a dominant player his first three years in the NFL, combining for 39.5 sacks in his first 42 games. But he was injured in his fourth season, being limited to just one game. Over the next four years, Merriman managed to appear in 32 games and recorded only six sacks.
From that same draft class was a quarterback by the name of Aaron Rodgers. Rodgers only threw a total of 59 passes over his first three years in the NFL. By the end of his fifth season, he had emerged as one of the league’s premier quarterbacks and eventually led the Green Bay Packers to a Super Bowl victory in his sixth season. If only taken in the context of three years, Merriman would have even edged out a player like DeMarcus Ware as the top player selected in that 2005 draft class. After five years, Ware more accurately shoots to the top, with Rodgers not far behind.
I could go on and on illustrating why the five-year mark is a more accurate judge, but I’m sure by now you get the picture.
With that in mind, two years ago I went back and started grading drafts from 2007 and on with a five-year cut-off. Meaning I looked back at the 2007 draft to see where it’s players were after the 2011 season and so on. Now that the 2014 season is done, I can now go back and look at the 2010 draft class.
Using a simple (and subjective) grading system, I assigned every player drafted an A, B, C, D or F grade based off my assessment of the sort of careers they have had up until then.
Players given an A grade were the so-called “elite” or blue-chip players. “B” players were the guys that were among the best at their positions, but not quite on that elite level. “C” players were ones that were deemed quality starters. “D” players were below average starters, role players and backups. “F” players were ones that were out of the league by the time their fifth year ended.
History Indicates Most Draft Classes Are Roughly The Same
The main discovery I came across was that the overall difference between individual drafts wasn’t significant. There was certainly variance between drafts, but not to the degree that is often stated. That was a big reason why I’ve concluded that oft-made statements describing the strength or weakness of an individual draft class are often overrated.
There was variance between the distribution of A, B, and C players, but generally speaking every draft class from 2007 to 2009 produced about the same amount in total, ranging between 56 and 61 players that combined to be quality starters or better.
Also, the number of players that were still in the league by the end of their fifth years wasn’t that different, ranging from 134 to 146 between the 2007 and 2009 drafts.
My conclusion is that the NFL draft, as far as producing NFL talent, falls along normal distribution, or the bell curve. At least when charted, it appears to follow the normal downward slope we associate with that curve.
Nearly half the players drafted every year are going to be out of the league by their fifth seasons. And roughly half of the remaining number will become backups and/or role players. That means roughly a quarter of all players drafted will turn into quality starters. And in terms of the players good enough to receive A or B grades, there’s about a one in 20 shot of snagging one of those guys.
Another big conclusion I’ve come to because of that is that draft prospects are probably for the most part overrated. The fact that roughly only five percent of the players drafted any given year are really going to turn into upper-echelon NFL players really illustrates the “crapshoot” nature of the draft.
Typically, the majority of those players are selected in the first half of the first round, but at least from 2007 to 2009, their distribution appeared to be a lot more random than probably many would have assumed.
2010 Draft Produced More Depth Than Previous Years
Now that I’ve crunched the numbers on 2010, I haven’t seen much reason to make major changes to any of the above conclusions. Here are the 2007 through 2010 drafts side-by-side using my simple “A through F” grading system:
Draft Grades (2007-10)
The biggest takeaway is that the 2010 class seemed to produce a lot more players that hung around as backups than previous draft classes.
And to be honest, I don’t quite know why that is. My first guess would be the number of underclassman that entered the draft. In 2010, 53 underclassmen officially declared for the drafting, which tied the record number from 2008. And it’s notable that 2008 also had a significantly higher number of “D” players that stuck around than the class ahead and behind it.
That might be a worthwhile conclusion because the 2011 class, at least when looking at it from four years out, also appears to have a very high number of D players. Now that’s typically normal after four years. Past experience shows me that the number of “A, B and C” players only makes slight shifts as the fourth year turns into the fifth year. That’s another worthwhile takeaway is that most quality starters or elite players are already established by their fourth years. As that applies to the Falcons, hopes that 2012 draft picks in Peter Konz, Lamar Holmes and everyone’s favorite, Jonathan Massaquoi, suddenly turn things on their fourth years are probably naive.
Typically the biggest change from year four to five is that many of the borderline guys that are already in the “D” category get washed out of the league by the end of training camp and don’t come back before the year is done.
A year ago, the 2010 group preliminarily had 127 “D” players after four years and wound up with 107 with after five, a change of 20 players. Not quite coincidentally, the number of “F” players went from 71 to 93, a shift of 22 players.
After four years, the 2011 class has 123 players graded as D. Even if 20 or so players wash out of the league before the end of the 2015 season, that would still represent a number of “D” players on par with the 2010 class.
Underclassmen and Three-Day Format Might Help Keep Players in NFL Longer
It’s also notable that the 2011 broke 2008 and 2010’s record of underclassmen with 56 early entrees declaring for the draft. What will be interesting to see is if there is a similar trend as the number of underclassmen only trended up after 2011. In 2012, 65 underclassmen declared for that drive. That record was broken a year later with 74 in 2013. And then the 2014 class featured 98 underclassmen, dwarfing the previous records.
Why should that affect things? Well considering that underclassmen are typically a year or two younger than the seniors selected, it might compel NFL teams to keep them around another year or so because of hopes that they show greater growth before their initial contract is up.
Another factor that might explain the large number of backup-caliber players sticking int he NFL is the new Collective Bargaining Agreement that was signed in 2011. But I can’t quite wrap my head around why that would have benefited the 2010 class that much. You can make the argument that smaller contracts and all picks being signed to four-year deals might make teams more willing to keep guys a year or so longer than they normally would have. But again, there’s no reason why that should have much effect on the 2010 class. The three-year deals were slowly being phased out in the NFL by 2010, with only a few teams (Pittsburgh Steelers, Baltimore Ravens and Detroit Lions) still using them by then. But there wasn’t a dramatic enough reduction from 2009 to 2010 to explain the increased number of backups in the latter year.
The only other major change that occurred in 2010 was the draft’s move to the three-day format. It’s certainly plausible that having the ability to regroup after the first round on the first night of the draft, and then again after the third round on the second night helped teams better manipulate the draft.
But regardless of the reason why, perhaps the new trend will be that instead of slightly more than half, rather closer to two-thirds of players drafted can stick in the league after five years.
How will that affect things? My best guess would be that might make a bit more difficult for undrafted free agents to stick if more drafted players that teams are investing in are sticking. But more than likely, it probably hurts the older veterans that normally could have stuck around for another year or two before having to hang up their cleats.
That’ll be interesting to discover because one of the main goals of the NFL Players Association with the lockout that led to the new CBA in 2011 was to help older players. A lot of the logic of installing a rookie pay scale was to free up more money for teams that would in turn go to veteran players. That may ultimately not be the case, and NFL teams are instead just keeping low-cost, middling rookies a year or so longer rather than that ninth or tenth-year veteran that might command a slightly higher salary.
Approximate Value Provides More Objective Past Draft Analysis
One fair criticism of my grading system is that it might be a bit too subjective. I have come across the problem of being on the fence with some players. Are they an A or B, perhaps a B or a C, or a C or a D? Generally, there’s only a handful of guys that sit firmly on the fence in each draft class, but it’s fair to say that such a subjective grading system could be tweaked and refined.
My best answer to find a more objective evaluating tool was using Pro Football Reference.com’s Approximate Value. Approximate Value (AV) wouldn’t be described as a perfect system, but it certainly gives a more accurate and complete view of the distribution of talent within and between drafts.
It also raises some discrepancies with my grading system. In 2008, I placed A grades on Matt Ryan, Jake Long, Ryan Clady, Joe Flacco, Ray Rice and Carl Nicks. Those players represent six of the top eight players from that year’s draft class according to accumulated AV through their first five seasons. But based off AV, I should have also included running backs Chris Johnson and Matt Forte. I gave Johnson a B grade, but Forte received a C from me.
Had I used AV, other players like cornerbacks Brandon Flowers and Aqib Talib that I gave B grades to, probably would have been better placed in the “C” category.
So I think it’s probably fair to go back and use Approximate Value to determine if some of the major conclusions I noted previously remained true. It would be nice to see if there was an objective measure that backed up my subjective opinion that draft classes are fairly similar.
So here is the graph of the accumulated five-year AV of every draft pick in 2007 through 2010:
Here’s another chart that removes all the data points, so that you can get a better look at the trend lines, which are a better indicator of how each draft differs:
As you can see, the 2007, 2008 and 2009 trend lines are right on top of each other to the point that they’re practically identical. However, 2010 (in green) stands out from the rest. And it doesn’t appear to be that way because there were more “D” or backup-caliber players drafted than other years.
What exactly is a “D” player based on AV? That’s harder to figure out, but it did seem like the vast majority of players I graded out as “Cs” had accumulated five-year AVs north of 20. So let’s use 20 or below as an indicator of a D player just for illustrative purposes.
There were less players that had accumulated five-year AVs of 20 or less in the 2010 class with 189. Each of the other three draft classes each were within two of 200 players. However, I doubt the difference of a dozen or so “D” players explains the notable trend in the graph above.
2010 Draft Offered More “Hits” Early in Draft
Instead, it appears that the key about the 2010 draft is that most of its good players were taken early in the draft. There wasn’t as much random distribution, particularly with the earlier picks.
Each of the first 12 picks accumulated AVs of 24 or higher. The 2008 draft was the only other one where even each of the top five picks hit that mark. That’s another thing that illustrates the crapshoot nature of the draft, that even picking in the top five of a draft doesn’t constitute being able to land on a “can’t miss” prospect.
Then after pick No. 13 in 2010 (Brandon Graham, whose AV was 10) the next 11 picks all each had AVs of 22 or higher, with Sean Weatherspoon’s 22 being the lowest among them. Then Tim Tebow (AV of 12) at pick No. 25 was the next guy that had a relatively low AV.
The simplest conclusion seems to be that what makes the 2010 draft stand out more than anything is the relatively few number of “busts” or “misses” that occurred in the early portion of that draft.
I haven’t looked at the 2011 draft class yet in terms of Approximate Value, but it does appear likely that it will fall along similar lines since it is another draft class where the majority of the best players in it were selected early, with one notable exception.
Nonetheless, it’s still worth looking at how players’ AV was distributed from each year. The following graph shows the number of players in each draft based on AV in intervals of five:
There are few intervals that have a difference of more than a dozen or players between the drafts with the highest and lowest numbers. Thus, even with a “better” draft such as 2010, there isn’t a high degree of variance between it and its predecessors.
Once again, the overall conclusion is that draft classes are relatively the same from year to year. What separates a “strong” class from a “weak” one may wind up only be a dozen or so players, roughly five percent. That’s basically what your standard margin of error would be in a data pool of 25o or so prospects.
It appears it’s more notable to decipher how talent is distributed throughout the draft than whether it produced more or less than another class. In the case of a draft like 2010, it seemed that the majority of NFL teams correctly identified the best players since a larger portion of the top picks turned into good players, at least as far as Approximate Value is concerned.
In a draft like 2009 that lacked as many “A-level” players as the other classes, it’s strength seemed to be in the latter part of the first round as well as the latter half of the second through the third rounds where players like LeSean McCoy, Phil Loadholt, William Moore, Sebastian Vollmer, Louis Vasquez and Mike Wallace were all selected.
Identifying talent distribution might prove easier said than done since so much of it can only be determined after the fact. How does a team figure out that picking 53rd, 54th or 55th in the 2009 draft might land them McCoy, Loadholt or Moore, while picking 10 spots ahead might land them Everette Brown, Pat White or Clint Sintim?
So I’m not sure if an NFL team’s draft process can be overly enhanced by this data unless they have been more successful than most at correctly identifying talent. Such a team would be better able to navigate the ebbs and flows that come within every draft. But whatever advantage is gained, the data suggests it won’t be significant.
NFL Teams Should Maximize Opportunities to Hit in Draft
But at the very least this data can potentially help crush long-held conventional wisdom about the relative strength of draft classes as well as illustrate how relatively few players that get drafted wind up being good NFL players.
And that might do more to help teams in their draft process. Perhaps drafting is less about finding the franchise-changing players like Matt Ryan, Patrick Willis, Ndamukong Suh or Earl Thomas given how relatively rare such players are. Instead, team’s drafting philosophies should be more about the cumulative process of acquiring as many good players over the long haul.
It’s not to say that drafting a couple of C-level players year after year will lead teams to be more successful. But perhaps NFL teams should be less willing to go “all-in” on one particular player that they’re trying to trade up to get. Teams might want to trade back more often than not, acquire as many picks as possible and increase your overall distribution of picks so that there’s a higher probability you might find one of those “sweet spots” where some talent has slipped through the cracks.
Ultimately the percentage of times that NFL teams hit (roughly 25 percent) in the draft is comparable to the amount of times that your average Major Leaguer gets a hit in the sport of baseball. And subsequently, teams’ draft philosophies should gear less towards the “home run” and more towards getting as many at bats as possible.
To continue the analogy, there’s nothing necessarily wrong with swinging for the fences once at the plate but maximizing the number of at bats might be the most important thing.