How well future pros sprint, jump and elevator weights at the NFL Scouting Combine may tell us just a little flake about their career success.

(Within Science) -- Despite his superstar condition equally quarterback of the New England Patriots, Tom Brady's athleticism never turned any heads. At the 2000 NFL Scouting Combine, an almanac gathering where players dreaming of playing in the NFL evidence off their strength, speed and explosiveness in a series of drills, Brady's performance was famously underwhelming. His xl-1000 dash fourth dimension and vertical leap height are mediocre at best, with numbers mayhap more typical for linemen 100 pounds heavier.

In the draft, the New England Patriots picked him in the sixth round, as the 199th actor and seventh quarterback chosen.

Yet, nearly xix years afterwards, on Feb. 3, Brady won his record sixth Super Basin, prompting many to bless him the greatest histrion of all time.

As a new batch of prospects convenes this week in Indianapolis for the 2019 NFL Scouting Combine, teams volition one time again be scrutinizing every dash, bench printing and leap in search of the next Tom Brady -- or, at to the lowest degree, someone who will help them win. But Brady's combine results didn't portend success, and he'due south far from the only player whose career surpassed his combine performance. At the same time, many other players with standout combine results never succeed in the NFL.

Today'due south NFL is equally sophisticated as ever, with thick, complex playbooks, yr-round scouting, and always-improving sports medicine, and has, in recent years, adopted an increasingly statistical approach, paralleling the analytics revolution that has already changed the MLB and the NBA. All of which raises the question: Can a few elementary drills really predict success in a sport equally complex as football?

More stories about football research from Inside Science
Math to Football game Coaches: 'Be More Aggressive on quaternary Downward'
Will the NFL's Rush for Position-Specific Helmets Pay Off?
Why Making Decisions at Game Speed Can Pb to Penalties for NFL Players

Finding the measurements that affair

The first combine, held in 1982 in Tampa, Florida, served as a more efficient way for teams to get together and share concrete and medical information on prospects. Since then, the combine has grown to include various psychological and concrete tests, becoming an consequence in and of itself -- televised live equally a four-day caricature for ardent football fans.

Each year, more 300 prospects undergo six timed or measured drills. The 40-chiliad dash and the vertical jump are 2 of the almost well-known. In the broad jump, athletes have to get-go from a stance and leap as far alee as they tin can. The three-cone drill requires prospects to run around three cones bundled in an 50-shape. The shuttle run involves sprinting laterally back and forth. And finally, prospects must bench press 225 pounds equally many times as they tin.

These drills don't quite replicate what happens on the field, and aren't relevant for all positions. For a quarterback like Brady, whose master job is to stand up in the pocket and throw the ball accurately, it doesn't matter then much that he's slow. But the drills exercise approximate full general athleticism, and it'southward reasonable to recollect better athletes would likely make amend football players.

Undoubtedly, NFL teams and their number crunchers have analyzed the predictive power of the combine. Only their results tend to be proprietary, and considering data hasn't always been hands available, just a few bookish studies accept been washed. Ane of the showtime was published in 2008, when Frank Kumitz and Arthur Adams at the Academy of Louisville looked for statistical correlations between success in the NFL -- equally measured by games played, statistics, draft positions and salaries -- and the combine information of quarterbacks, running backs, and broad receivers. Other than sprint times for running backs, nevertheless, the researchers establish no correlation at all.

Merely their approach might have been limited. Researchers led by sports statistician Masaru Teramoto at the University of Utah used a different arroyo that enabled them to better determine exactly which combine outcome has more predictive power. Their 2016 study found that the time over the showtime 10 yards of the 40-yard-dash was the most predictive of a running back's rushing yards per attempt. For a broad receiver, the cistron that most predicted success was only his height. His leaping ability was likewise of import, as the higher he could jump, the more receiving yards he would gain -- a reasonable link, since receivers often have to jump over defenders to catch the brawl.

Most recently, in peradventure the well-nigh comprehensive study to appointment, sports scientists Lisa Vincent, Bryan Blissmer and Disa Hatfield of the University of Rhode Island analyzed the predictive ability of the combine not just for broad receivers and running backs, merely also for quarterbacks, defensive ends, defensive tackles and linebackers.

Their analysis, published in January in the Journal of Strength and Conditioning Research, used an approach like to Teramoto's study, simply didn't look at the bench press or three-cone drill, which is similar to the shuttle drill in evaluating agility. The bench press, Hatfield said, isn't relevant for football.

Mayhap surprisingly, the analysis revealed that the shuttle drill didn't correlate strongly with anything. But the other events did, modestly predicting success in some metric for each position, similar yards gained or tackles fabricated. What especially stood out, Hatfield said, was the usefulness of the oft-overlooked wide jump in predicting success for running backs, defensive ends and defensive tackles. Again, this makes intuitive sense, Vincent said, as these players almost ever start each play crouched and ofttimes with i hand on the footing.

The numbers with predictive power weren't just raw combine scores, either. The researchers incorporated combine scores with body weight to calculate various types of power, a dissimilar metric that complemented raw scores. Horizontal and vertical power, for example, combines body weight with raw scores from the 40-k-nuance and vertical leap, respectively. Both these numbers pointed to defensive ends who made more sacks and tackles, and quarterbacks who rushed for more yards.

Overall, the analysis suggests that the combine can predict most 20-25 percent of a player'southward future NFL success, Hatfield said.

While these studies used a few years of combine data, David Hedlund, a researcher in sports data and analytics at St. John'due south University, in Queens, recently compiled 15 years of data. He compared players who earned All-Pro honors, those named to the Pro Bowl, and those who received neither honor. All-Pro players represent the best at their position in a given season, as voted on by the media. Pro Bowl players, are chosen by coaches, players and fans to play in an annual all-star game. All-Pro players had higher average combine scores than Pro Bowlers, who in turn had higher average scores than the balance did. Although the written report lacked the statistical rigor of others, information technology at least hints at an credible association between combine scores and NFL success.

But one of the primary limitations of these kinds of studies is how success is measured. Tackles and yards might mean victory in fantasy football, but they don't always reflect a histrion's skill or value.

"At that place's no mode we tin always account for the complexity of the game," Vincent said. How many yards a running back gains depends on the blocking prowess of his offensive line. A quarterback relies on elusive receivers who can become open. Wide receivers depend on accurate quarterbacks. And then there's the 24-hour interval's game plan, specific game situations, and the team's overall style of play, which all touch on the box score.

In recent years, however, teams and analysts are looking beyond the box score, using increasingly sophisticated numbers that more accurately reflect a player's value. For example, ESPN uses a proprietary number called total quarterback rating, or QBR, a number based on a formula incorporating a quarterback'southward passes, turnovers and other statistics.

From the combine to success on the field

Even with better measures of on-field success, the predictive power of the combine is express. Not just practice drills neglect to capture the complication of football, they ignore intangible traits like leadership, drive and "mental makeup" -- the quality the Patriots cited when drafting Brady.

The best predictor of future success in the NFL is even so past success in higher. "Combine measures are not as relevant equally on-field functioning measures," said Matt Manocherian, director of enquiry and evolution and football at Sports Info Solutions. "That'southward based on whatsoever statistical test you tin can construct."

His visitor builds statistical models that evaluate players, based primarily on on-field functioning -- those avant-garde stats such as an offensive lineman's failed blocks. But combine data is also added to improve the model.

Combine scores are useful as baselines, said Manocherian, who has previously worked as a sentry for the New Orleans Saints and Cleveland Browns. You want to identify prospects with at to the lowest degree some minimum corporeality of athletic ability, so y'all can, say, filter out the slowest wide receivers.

Merely considering of the rise of analytics, teams have gotten more than statistically sophisticated and better at assessing players. "Because nosotros tin can evaluate players better," he said, "information technology allows us to not rely on the combine and so much."

For some teams and analysts, however, that might be a mistake. "My bet is that teams probably disbelieve the combine besides much," said Brian Shush, a senior analytics specialist at ESPN. Many previous models couldn't detect a strong correlation between combine scores and NFL success considering of a statistical bias called Berkson's Paradox, he said.

Because the combine participants don't correspond the general population -- they're already the foam of the crop of football players -- it turns out combine scores and future success can appear to be much less correlated than they actually are, he said. "Quantitative analysts have been discounting the value of the combine for years and years, when the reality is that it can predict career outcome if you analyze it correctly," Burke said.

While such a selection effect could take come into play for his study, there'south no good way to account for it, Teramoto said. For the University of Rhode Isle analysis, Hatfield said they took care to make sure that this kind of bias was probable not an issue.

Information technology's not just about the measurements

Even afterward bookkeeping for Berkson'south paradox, the combine data shouldn't be relied on too much. "It definitely should not be everything," Shush said.

Just as an event and an annual leaguewide convention, its value extends beyond the drills. "The about important purpose of the combine," Manocherian said, "is for the doctor to look at the guys."

The combine too allows teams and coaches to interview prospects and to get to know them as people. "The combine," Hatfield said, "is absolutely necessary because it gives those teams an opportunity to interact with the players."

In one case private, the combine is now a highly anticipated event, drawing the most media of any NFL outcome other than the Super Bowl. It reportedly rakes in more than $8 million in annual acquirement for Indianapolis. "I believe today, the combine is taking on more of a marketing office than anything else," Hedlund said.

Simply for the purposes of evaluating players, the challenge is to better empathise combine data and their implications. "I'yard less concerned with under or overuse of combine metrics than misuse of combine metrics," Manocherian said. "We need to start understanding what they mean a lot better in terms of how predictive they are when we integrate them with on-field performance data."

And as the league becomes more statistically savvy and continues to analyze the combine and on-field performance, that understanding will grow. "We know a lot more than what we used to know," he said. "But we know nothing compared to what nosotros will know."

Author Bio & Story Archive

Marcus Woo

Marcus Woo

Marcus Woo is a freelance science writer based in the San Francisco Bay Surface area who has written for Wired, BBC Earth, BBC Future, National Geographic, New Scientist, Slate, Observe, and other outlets.