On the day before the NFL draft, I wrote about a company called Achievement Metrics that analyzes the speech of college football players to predict how they will perform in the NFL. The basic premise is that future backup quarterbacks talk like backups, and future starters talk like starters. Many of you were intrigued but wanted Achievement Metrics to put all of its cards on the table. The big question: Who do the company's analysts believe will be the stars and busts of the 2011 draft? Did the Carolina Panthers make the right call by picking Cam Newton?
Achievement Metrics' Steven Hofmann has answered your call. The chart below plots quarterbacks based on their positive power score and in-group affiliation score. Positive power is the belief in one's ability to influence events and outcomes: "I refuse to consider that" or "I promise that it won't happen again." In-group affiliation is judged based on the use of expressions that suggest team spirit: "We did a great job today" or "Our linebackers were amazing." This updated version includes the top 2011 quarterbacks as well as Tim Tebow and Sam Bradford, the two QBs chosen in the first round in 2010.
As you can see, Cam Newton is in what I've termed the "quadrant of doom." He talks like other first-round draft picks—Matt Leinart, Alex Smith, Brady Quinn—who have bombed in the NFL. Does this mean Newton is destined for a disastrous career? Absolutely not. This year's No. 1 pick could defy the probabilities and become a success in the NFL, just the way that Carson Palmer has done.
More jarring, perhaps, is the presence of Tim Tebow in the quadrant of doom. Though his on-field readiness for the NFL has been debated endlessly, nobody disputes his skill as a leader and orator. The pledge he made after a 2008 loss to Ole Miss—"You will never see someone push the rest of the team as hard as I will push everybody the rest of the season"—is even inscribed on a plaque at his alma mater. Although this sounds uplifting to our ears, the algorithm heard something different. It detected a guy who had a little less love for the team and a little less belief in himself than his peers did.
The big surprise in the first round was the Minnesota Vikings' pick of Christian Ponder, a quarterback whom many had projected to go in the third round. On the NFL network, the assembled pundits couldn't stop repeating "there's something about this guy." He's in the "Julius Caesar" quadrant, along with his fellow first-round draft picks Jake Locker and Blaine Gabbert. The other highly regarded quarterbacks in the draft, Andy Dalton (taken by the Bengals as the 35th pick) and Ryan Mallett (taken by the Patriots in the 3rd round) also talk (at least) like winners.
Now that we've come this far, let's go further down the rabbit hole. Steven Hofmann and Achievement Metrics have another way of looking at quarterbacks. They measure the amount of conceptual complexity in their speech. A good way to understand this trait is to think of a wine critic. She will taste some pinot noir and describe it as "medium-bodied with subtle hints of cinnamon and raspberry." By contrast, you might say, "Yummy!"
Players who demonstrate a lot of low conceptual complexity in their speech are not simpletons. Rather, they use a greater proportion of speech that is indicative of black-or-white thinking, such as "Exactly" or "We don't have any other choice but to" or "Obviously." Players with high conceptual complexity use terms and expressions that suggest an ability to see nuance in one's environment, such as "That's disputable" or "The alternative is to" or "Generally." Here's a chart of 72 NFL quarterbacks and the 2011 prospects plotted by conceptual complexity.
There are two clusters of Pro Bowl quarterbacks. In the upper-left, you have Michael Vick, Jay Cutler, and Tom Brady. These quarterbacks talk about football in a straightforward, no-nonsense style—call them "instinctual" quarterbacks. In the lower-right, you have players who speak with more shades of gray: Aaron Rodgers, Drew Brees, Tony Romo. Call them "intellectual" quarterbacks. (Note that there are some duds in both categories.)
What stood out for the analysts at Achievement Metrics was what they termed the Zone of Separation. (You can think of other names for it. I like to call it the Trench of Mediocrity.) Remember, all of these quarterbacks are being analyzed based on press conferences held while they were in college. (And, yes, correlation does not equal causation, but let's go out on a limb here.) For some reason, players whose speech is balanced between low conceptual complexity (or instinct) and high conceptual complexity (or intellect) tend to become backups and also-rans. As for the 2011 draft picks, Cam Newton looks better on this chart, whereas Jake Locker appears destined to be carrying around a clipboard in a few years.
Clearly, Achievement Metrics has not pulled the sword from the stone. There is no single variable that predicts the future, no data-powered crystal ball. But even these probabilistic, suggestive numbers can shed some light on football arguments. I wrote in my previous piece that the Wonderlic, an "intelligence" test, does not correlate with NFL performance. What the conceptual complexity graph shows is that pro quarterbacks don't all share a single mindset. There are multiple ways to understand the game, and different approaches can work. You just don't want to fall in the Trench of Mediocrity.
So there are the 2011 numbers, splattered on the chart—another variable to consider as you pick your fantasy teams, shout at your giant flatscreens, and do battle on the message boards. Let's all meet here again next year to see how the predictions pan out.