Why Are Offensive Pro Bowlers Worse Recruits?
While many evaluation factors are at play, the main culprit appears to be older offensive linemen.
Bruce Feldman of The Athletic recently noted that the offensive members of the NFLPA All-Pro team tended to be worse recruits coming out of high school than their defensive counterparts. The coaches, recruiting staffers, and scouts he polled reported theories such as the difficulties in evaluating OL/QBs, geographic biases in recruiting rankings, and college position changes. I looked at the 2015-2022 Pro Bowl rosters to see if Feldman’s observation held in a larger sample of good players. The observation holds, however I believe the largest driver of this statistical anomaly is the time and experience it takes to develop as an offensive lineman at the NFL level. This is not mutually exclusive with the explanations presented by Feldman’s sources, but I believe was understated.
Note: For those unfamiliar with recruiting rankings, 247sports, the source of my recruiting data, does not designate 1 star ratings for players who receive a composite score. The lowest they give a player is 2 stars, with 0 stars indicating no recruiting service rated the player.
My research took a different form than Feldman’s in looking at Pro Bowl rosters instead of All-Pro rosters. This was because I believe the interesting question posed in his original article was not the specific “Why were offensive players in the NFLPA 2022 All-Pro teams lower rated recruits than the defensive players?” but rather the more general “Why were high-level offensive NFL players lower rated recruits than defensive NFL players?” The fan-voted NFL Pro Bowl does not perfectly select top performers, but over the course of several years it gives us a sizable sample of players who are considered to be good by teams and the public. The 2015 Pro Bowl was selected as the starting point as that was the first year where former class of 2011 recruits could reasonably be in the Pro Bowl. The importance of that class in specific will be discussed later.
Before we get ahead of ourselves, let’s check if the observation holds for our 2015-2022 Pro Bowlers sample.
Even after removing players who graduated high school before 2002 since they were not rated, it holds! Offensive players had an average star rating of 2.99 while defensive players had an average star rating of 3.28. Additionally, 43.0% of Pro Bowl defenders were blue chips (4 or 5 star recruits) while only 39.9% of offensive Pro Bowlers were. As a group, defensive players were better high school recruits than offensive players. Why is this?
It would be hubris to dismiss the testimonials of coaches, recruiting staffers, and scouts in potential explanations. Athleticism and size are much better predictors of success for defenders, and they’re the most viscerally apparent aspect of a player to coaches and scouts. If one were to approach the impossible task of gauging if a teenager would be an extremely accomplished professional in 5-8 years, they are probably the best places to start for football players. This means by the nature of the beast, defensive projections, particularly along the DL, are likely to be more accurate. That being said, whenever looking at historic recruiting data it is vitally important to take into account time effects of ratings around when 247sports was founded in 2010 and first covered a full recruiting cycle in 2011.
After 247 began covering recruiting, rankings became much more accurate. As evidence, consider the rate at which 5 star recruits according to 247sports composite ratings were drafted, and how rapidly it increased following 247sports’s first full cycle in 2011. Note that several 2018 5 stars are still in college or are in the current draft cycle but are considered undrafted by matter of convention for this exercise. Correlation does not equal causation, so we cannot say it was 247sports in particular giving more accurate ratings which led to this increase. But it does give us reason to mentally weigh the recruiting ranking of Trent Williams (Class of ‘06) different than Laremy Tunsil (Class of ‘13). Keeping this in mind, what proportion of Pro Bowlers at each position graduated high school after 2011?
Running backs, unsurprisingly given their low career lengths, are the youngest group. They’re closely followed by WRs, who have supplanted the older generation and likely have better training due to the proliferation of 7-on-7 camps. The DL group is a case of Simpson’s paradox that cannot be resolved in this case without great difficulty due to Pro Football Reference’s inconsistent position namings. Most of the older DL are interior players (i.e. Aaron Donald, Ndamukong Suh, Cameron Heyward, etc) while the younger players are primarily EDGEs (Bosa brothers, Myles Garrett, Brian Burns, etc). DBs are a healthy mix of young phenoms and savvy veterans who can rely on technique. QB gets a major boost from the AFC’s youth movement, and the flawed Pro Bowl voting system. LBs and TEs similarly have many savvy veterans who can perform at a high level.
The oldest group is the offensive line, where less than a third (32.4%) of Pro Bowlers from 2015-2022 graduated high school after 2011. Of the 213 individual offensive Pro Bowlers from 2015-2022, 71 of them (33.3%) were offensive linemen. Even for the 2022 Pro Bowl, 7 of 16 OL graduated high school before 2011.
If we dig down into the 23 Pro Bowl OL who graduated post-2011, only two were 0 or 2 stars coming out of high school. They were 2-star Joe Thuney, who weighed 250 pounds as a senior, and 0-star Dion Dawkins who attended a military academy for a year between his senior year and freshman season at Temple. They stand in stark contrast to the 12 blue chip recruits amongst those 23 players. Compared to the 44 pre-2011 (and post-2002) OL, 19 of them were 0 or 2 stars (39.6%) while only 7 (!) were blue chips.
This tells a simple story behind the recruiting difference between the two groups of Pro Bowlers. Offensive Pro Bowlers are being drug down by older offensive linemen who graduated high school before an increase in recruiting rating accuracy and were thus more likely to be misevaluated. As far as potential explanations for the discrepancy between high-level offensive and defensive NFL players as recruits, this is satisfactory evidence to consider. But this leads to a more fundamental question: Why are so many older OL making Pro Bowls? A full, detailed answer is outside of the scope of this article, but I’d like to present one theory here.
Pro Bowl OL generally have more experience when compared to other positions over this time span. The median experience of a Pro Bowl OL is 6 seasons, meaning 50% of them from 2015-2022 had more than 6 years of experience when they were voted into the event. In my opinion, this suggests two non-exclusive possibilities:
1. It takes several years before good OL play is recognized enough by the public to generate Pro Bowl nods.
2. Even talented young OL generally take \time to develop into high-level NFL OL.
I lean heavier toward into option two, as there is some empirical evidence to suggest OL’s extended learning curve, but option one cannot be ignored. In any case, this suggests that even if a talented OL enters the league there is a several year lag before they perform at a Pro Bowl level, and/or get enough recognition to earn votes. Combined with the late increase of recruiting ranking accuracy, this further exacerbates the difference between offensive and defensive players.
To illustrate explicitly, consider Landon Dickerson. He was a highly touted 4-star recruit in the 2016 class that committed to Florida State before transferring to Alabama and declared for the 2021 draft. He was selected in the top of the 2nd round by the Philadelphia Eagles, started as a rookie, and quickly made the Pro Bowl in his second season on the best team in the NFC in 2022. Even in nearly the best case scenario, there was a 6-year lag between Dickerson as a recruit and making his first Pro Bowl. Because of this unavoidable lag, I suspect that with younger OL making the Pro Bowl as time goes on the difference in recruiting rankings between high level offensive and defensive players will shrink.
Thank you for your time, and I hope you enjoyed! Next week’s installment will discuss parity in the NFL across eras. Please subscribe if you wish to have these directly in your inbox every Wednesday at 7 AM ET -CRM