Wondering about the Wonderlic: Does it Predict Quarterback Performance?

By Austin Tymins and Andrew Fraga

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:

Position Average Score
Offensive tackles 26
Centers 25
Quarterbacks 24
Guards 23
Tight Ends 22
Safeties 19
Linebackers 19
Cornerbacks 18
Wide Receivers 17
Fullbacks 17
Halfbacks 16

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.

To determine whether the Wonderlic has any effect on QB performance, we examined the test results and NFL performance of 50 quarterbacks dating back to 2007. We included 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
QBR .0049
Sack Percentage -.1071
ANet Yards Per Attempt .0535
Passer Rating .1217
Interception Rate Per Attempt -.1944

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.

Variable QBR Sack ANetYa~A Passer~g IntPA
p>|t| .979 .56 .771 .507 .286

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.

About the author

harvardsports

View all posts

9 Comments

  • Disagree with your study design. Far too many variables when trying to compare a single metric to overall success. There are far too many confounding variables.

    My guess is that if you peel off the bottom 10% of Wonderlic scores you would find a high degree of failure at the QB position, particularly when compared with other QB drafted in those same positions.

    I would also attempt to normalize for draft position where you would see “smart” quarterbacks drafted late with a somewhat greater chance of carving out an NFL career, at least as a long-term backup.

  • As a correlational metric, it seems to me that solely basing the overall efficiency and productivity of a quarterback (or any position for that matter) is indeed problematic. However, as simply a piece of information to better understand a team’s prospects, I think this test has some merit; why would it continue to be used so often within the NFL? Of course, I know that “because it has been done before always” is not a valid argument. I see it as simply one small piece of info, just as players are given personality profiles, to holistically assess any given player. To assume that this test proves the validity and productivity of a pro quarterback is quite narrow-minded, at best.

  • Thanks for doing this analysis.

    Perhaps with many players, and particularly QBs, there may be other elements of attention involved. There are players I’ve seen and known who play by intuition…there is a descriptive feeling of what is going on and they can intuit almost instantaneously what to do.

    One of the goals is coaching is to refine a player so that his reaction time to the play is minimized…literacy is too slow in this regard. Great for instruction and relating info and data in the learning mode and as a feedback mechanism, but a different sensibility.

    Another attention element is proficiency…players who naturally have the requisite physical and mental skills and have the capacity for developing a high level of attention to the physical process of the game…they’re just players. They may not play any other game well, but football (or the chosen sport); it is their discipline.

    The amplifier is leadership ability, which has a certain dependency on team composition. At best a player in a key position can push a team to its best performance. The ability may be intuitive to some degree. Not the sort of process that would lend itself to literate testing unless it was a designed part of the test.

    I don’t doubt there are more elements that might be pertinent to a good design, like electrons around a central proton.

  • Interesting look at the Wonderlic test. I’ve often wondered if time given to players should be examined and if it influences their test scores. Have you completed any research regarding that possible issue?

  • What about first year players? Do the better ones have higher scores? I always thought the wonderlic wasn’t so much a measure of career performance, but rather how quick a player picks up the game. Also do teams with players of higher wonderlic scores expect more from their intelligent players and make their roles more complicated while players with lower scores have their tasks simplified? My guess is that coaches and managers know what their players are capable of and assign their players corresponding roles and expectations.

    Interesting article though. Keep up the good work guys.

  • To my knowledge, there is one visual-spatial intelligence test that predicts objective, on-field performance in American Football. The visual-spatial test was discussed in an Evolution of Sports talk at the MIT Sports Analytics Conference by Dr Damon Bryant, who is now the Lead R&D Scientist at Infor. The test was shown to be better in measuring intelligence than the Wonderlic. Here is a link to that presentation:
    http://www.sloansportsconference.com/?p=5479

    After receiving funding from Dan Gilbert’s Bizdom accelerator, 1st & Goal Recruiting completed a study with high school and college athletes who took the visual-spatial intelligence test described at the MIT conference. Results showed that for some offensive and defensive positions, visual spatial intelligence predicts the probability of scoring touchdowns while controlling for physical characteristics such as height, weight, speed, strength, and agility. Here is a copy of the presentation:
    http://t.co/9i6iT0zHrx

Leave a Reply

Your email address will not be published. Required fields are marked *