Wondering About the Wonderlic: Does It Predict Quarterback Performance?

By: Austin Tymins and Andrew Fraga
Published originally at Harvard Sports Analysis

During the 2014 NFL Draft, all 32 NFL teams will be on the clock to invest in the future of their franchises. Decision makers will feel immense pressure to secure a top-notch first round pick, find the next Tom Brady in the sixth round, and, most importantly, avoid selecting a bust. College stats, highlight reels, and NFL Combine results will all be evaluated. The draft, however, isn’t just about physical prowess; in addition to the 6 workouts at the NFL Combine, such as the 40-yard dash and bench press, draft prospects must also complete the Wonderlic Test, an examination designed to gauge mental aptitude.

Prospects must complete this 50-question quiz within a 12-minute time limit. Scores on the Wonderlic Test range from 1 to 50 with the average draftee scoring about a 20. Charlie Wonderlic Jr., President of Wonderlic Inc., claims that a score of 10 and above suggests literacy, while the average score of 20 is comparable to a middling 100 IQ score. In theory, this brief exam should offer teams a comprehensive look at their prospects’ mental capacity, a crucial component to succeeding in any sport played at the highest level.

In its rather long history, The Wonderlic has produced a few top scores. Tennessee Titans quarterback and Harvard graduate, Ryan Fitzpatrick, notched a notable 48 on the exam, the highest reported among active players. Punter and Harvard graduate Pat McInally, however, remains the only player to have ever scored a perfect 50 on the test.

On the other end of the spectrum lie some horribly low scores. Quarterback Vince Young and San Francisco running back Frank Gore, for example, both received a measly score of 6. In addition, Dallas Cowboys cornerback Morris Claiborne scored a 4 on the exam, understandably so, however, given his reported learning disability.

Richard Sherman, you ask? He beat out a majority of draftees with an above-average score of 24.

ESPN provides a sample Wonderlic Test (here for those interested) with questions like: “Paper sells for 21 cents per pad. What will four pads cost?” And “Which number in the following group of numbers represents the smallest amount?”

Paul Zimmerman of Sports Illustrated reported in his book, The Thinking Man’s Guide to Pro Football, the average score by position on the Wonderlic as follows:
Average Score
Offensive tackles
Tight Ends
Wide Receivers

Football is a much more complicated game than it appears to the casual observer. Coaches expect players to analyze hours of film to pick up on their opponents’ tendencies. During the game, players must adjust to constantly shifting dynamics. And no position is as cerebral as quarterback.

We looked at different measures of quarterback efficiency including QBR, Sack Percentage, Adjusted Net Yards Per Attempt, Passer Rating, and Interception Rate Per Attempt. From the various tests we ran, we found a negligible correlation between all the variables and Wonderlic scores of quarterbacks.

Performance Statistic
Correlation with Wonderlic score
Sack Percentage
ANet Yards Per Attempt
Passer Rating
Interception Rate Per Attempt

Not a single variable tested had a correlation above .2 (or below -.2), suggesting a minimal or very weak correlation between quarterbacks’ Wonderlic scores and the other variables at best.


Furthermore, the results of the regressions we ran tell a similar story. After individually regressing QBR, Sack Percentage, Adjusted Net Yards Per Attempt, Passer Rating, and Interception Rate Per Attempt on the corresponding Wonderlic scores, we did not find a single relationship that proved to be statistically significant at the 5% level, and most are not even close. That is, a quarterback’s score on the Wonderlic Test does not serve as a significant predictor for any of the metrics we analyzed.

It’s unclear whether intellectual proficiency isn’t as important for quarterback as we might think, or that the Wonderlic isn’t very good at measuring it; regardless, it’s very clear that the Wonderlic isn’t, and shouldn’t be considered, a good predictor of quarterback performance.  At the end of the day, scouts are better off watching tape, pro days, and the combine rather than reading test scores.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

5 Responses to “Wondering About the Wonderlic: Does It Predict Quarterback Performance?”

  1. Anonymous says:

    About 77% of the teams that made the playoffs the past 4 years were lead by QBs with top-10 EPA/P. That is, even counting teams tht were clearly defensive-driven juggernauts like the 2010 Jets, Playoff QBs tend to have the best EPA/P.

    You could remove QBs who did not contribute much to wins, but whose teams had elite defenses, and you would have a list of arguably the best QBs in football.

    Now, what is the average wonderlic of these QBs? How much higher is their wonderlic over the NFL mean? How much higher are these QBs' wonderlic scores over the NFL QB mean?

    Without looking, I'm going to guess something like 28.

    If I were an NFL GM, I would have the hard time pulling the trigger on a QB with a sub 28 Wonderlic score unless they also had great passing numbers, like adjusted QBR, Pass EPA, and Total EPA combined with a sub 4.67 40 yard dash.

    I would not however, draft a QB with a high wonderlic expecting that to correlate by itself to a high QBR.

  2. Unknown says:

    I would have been very surprised to find a relationship, given the structure of the model you proposed. There are too many covariates involved. One would need to control for other factors--physical skill level/talent being chief among the other factors. Absent the data to parse out contributions to success to intelligence, physical skills, coaching? quality of teammate? it's hardly surprising. Furthermore, I think Wonderlic's usefulness, if it has any, would be more as a "qualifier". In other words, one must have at least some minimum threshold to have a chance. Kind of like arm strength. A QB may not need to throw a ball through a brick wall, but if they have a pop-gun arm, odds aren't very good of ever being a top performer, no matter how intelligent. Maybe not impossible, but if the draft is about probabilities, if the Wonderlic can assist in separating candidates who are otherwise equal, or eliminating candidates who are probably too dumb for their position, I think that's where it potentially could be predictive. But maybe not. But I think that's a more likely scenario to find potential use in it than what you proposed.

  3. Reilly Pinkelnwert says:

    Sometimes I really do worry about whether people are becoming overly enamored with regression analysis. Does anybody really think that the performance of a QB will gradually improve as his Wonderlic score goes up? I doubt it. I suspect the real questions is whether there is a cutoff point, below which a player's stupidity starts to significantly effect their ability to perform at a reasonably high level. That doesn't seem so implausible to me, though it is a different sort of question from the one being asked in this post.

  4. Unknown says:

    Cedera Lutut, Vidal Naik Meja Operasi

    Berita terbaru dan terkini dari agen bola City Holiday – Arturo Vidal mungkin tak akan membela Juventus lagi di sisa musim ini.
    agen bola City Holiday Terpercaya – Gelandang asal Chile itu naik ke meja operasi untuk menyembuhkan cedera lututnya.
    Operasi ini akan dilakukan di Barcelona pada Rabu 7/5/2014 dan dipimpin oleh Professor Ramon Cugat.
    "Pemain berposisi gelandang ini akan menerima operasi arthroscopic pada meniskus lateral lutut kanannya besok di Barcelona," demikian keterangan Juventus di situs resminya.
    Football Italia melaporkan bahwa operasi ini tak akan mengancam partisipasi Vidal di Piala Dunia 2014. Gelandang berusia 26 tahun itu diprediksi akan pulih tepat waktu dan bisa tampil di Brasil.
    Sumber http://cityholidaybet.com/

  5. Anonymous says:

    Range restriction, error, additional variables that are not controlled for. These would likely explain that lack of a correlation.

Leave a Reply

Note: Only a member of this blog may post a comment.