By Ross Benes
One way to see if high school recruiting ratings accurately measure talent is to examine the NFL draft.
If the highest-rated players really are the most talented, then they should also be more likely to get drafted and have success once they're in the league. To test that theory, I mined data to see if the recruiting services' talent evaluations translate to the NFL.
Recruiting ratings are inexact, and among the thousands of players graduating high school each year, the ratings still overlook some players who go on to become NFL stars. Two-star players becoming Pro Bowlers do make for great stories, but stories are only anecdotal. SB Nation has done some great work in applying probabilistic thinking to recruiting ratings, and they showed that blue chippers are about 10 times more likely than non-blue chippers to be drafted in the first round. The also found that high-rated recruits are more likely to get drafted in the first place and that they tend to get picked earlier in the draft than low-rated recruits.
But the SB Nation posts only examined one NFL draft apiece, which means the effects they found may be due to a small sample size. And they also didn't measure if those high-rated recruits actually became better pro players. To extend SBN's analysis I looked at the last five drafts, which produces a sample of nearly 1,300 players. I also examined if recruiting ratings translate into more success at the NFL level using Pro Football Reference's Approximate Value (AV), since that statistic quantifies the value of all players and allows for comparisons across positions.
Last year, I wrote in the Wall Street Journal that each bump in star rating is associated with a greater likelihood of getting drafted. As shown in the table below, relative to their proportion of rated high-school players, five-stars put out 4.4 times as many draft picks as two-stars.
In another WSJ article, I claimed that each increase in star rating corresponded with a better draft position and more success (higher AV) in the NFL. Those are two big claims, so let's break them down further.
According to Rivals ratings, on average five-stars are drafted almost two entire rounds earlier than two-stars and unrated players. The table below shows that every increase in star rating is related to a better draft position, which ranges from the end of the fifth round for unrated players to the end of the third round for five-stars.
|Star Rating||Avg Pick||Round, Pick|
|Unrated||161.3||5th round, 25th pick|
|2-star||151.9||5th round, 16th pick|
|3-star||124.2||4th round, 25th pick|
|4-star||116.2||4th round, 17th pick|
|5-star||89.3||3rd round, 25th pick|
Raw AV numbers show that five-star recruits have nearly twice the yearly output as two-star recruits. However, part of the reason for this effect is that many low-rated draft picks never appear in an NFL game, which brings their AV per year to zero, weighing down their cohort's average.
But even after controlling for draft picks who never appeared in an NFL game, each increase in star rating is still associated with an increase in AV per year to the point where five-stars produce a whopping 88 percent more AV per year than two-stars. As seen in the table below, a higher recruiting ranking also generally runs alongside a greater likelihood of appearing in an NFL game. The exception is that slightly more four-star draft picks failed to appear in a regular-season NFL game than their three-star counterparts.
|Star Rating||Raw AV Per Year||Control AV Per Year||% Never Played|
When it comes to output and the likelihood that a draft pick will appear in a game, the jump between two-stars and three-stars, and four-stars and five-stars is large. But the difference between three and four stars is pretty minimal. Taken altogether, the discrepancy between the lowest rated players and the highest rated players is huge.
To give you an idea of the difference between 2 AV per year (which two-stars average) and 4 AV per year (which five-stars average), Vance Walker had 4 AV last season for the reigning Super Bowl champion Denver Broncos team. To obtain 4 AV, Walker started four games and had 33 total tackles. Walker's teammate, Shane Ray, had 2 AV last season after starting zero games and recording 20 total tackles. The difference between Walker and Ray might not seem terribly large, but remember this is the average difference between the production of two-stars and five-stars. When you project this difference across hundreds of draft picks, it adds up to huge amount of production.
In summation, five years of NFL draft data show that a higher recruiting star rating is associated with:
- A better chance of getting drafted in the first place
- Getting picked earlier in the draft
- A greater chance that a drafted player will appear in an NFL game
- Greater production once a player reaches the NFL.
If the high number of blue chippers who become All-Americans, the correlation between recruiting ratings and college team rankings, and the fact that in a given college football matchup the team with the better recruits usually wins, haven't already convinced you that recruiting ratings matter, then the NFL draft should.
* * *
Ross Benes is a Sports on Earth contributor who has written for Deadspin, The Wall Street Journal, Esquire, and Slate. You can reach him at firstname.lastname@example.org and follow him on Twitter @RossBenes.