No one who watched the game quite understands how Brandon Steburg got hurt. The defensive end for Papillion–La Vista High School, just south of Omaha, Nebraska, had made some big plays, the last of which was a sack of the quarterback, who landed on top of him. It looked routine. Steburg even participated in the next couple of plays. But his gait slowed. Then he put his hands to his head and walked off the field.

On the sidelines, Steburg couldn’t stop vomiting. An ambulance rushed him to the hospital. There, to relieve swelling and bleeding in his brain, doctors removed most of one side of his skull. They weren’t certain he would live or, if he did, whether he’d wind up enfeebled. As Steburg began his slow recovery, images of the gruesome scar around his bloated head appeared in the media. A TV interview showed the alternately cheerful and tearful young man as he awaited the operation to replace the missing part of his skull. It all made clear how close he had come to a tragic end.

The September 2016 incident, set against the backdrop of a long-building panic over brain injuries in the sport, led to a surge in orders at Riddell, the Des Plaines–based manufacturer that rules the football helmet market. Those purchases—1,200 in total from the Omaha area before the 2017 season—were not just for helmets, but for helmets equipped with Riddell’s relatively new InSite Impact Response System. Its sensors register the force and location of impacts. If a player suffers a particularly dangerous hit, a transmitter alerts a coach or trainer, who can pull the kid from the game to evaluate his condition. The system even allows teams to track the knocks players take over time. In theory, InSite might have provided a heads-up that Brandon Steburg had a problem.

“A coach and a trainer cannot have their eyes everywhere on every play,” says Stephen Eubanks, head of athletics at Omaha Public Schools, which purchased 910 InSite-equipped helmets for players at all seven of its high schools. “Sometimes a hit doesn’t look impactful on a kid, but it is. And students often don’t want to self-report [and risk having to leave the game]. But this gives us data on everyone.” The wired headgear doesn’t come cheap. The sensor system adds $150 to helmets that can already cost $400. So how did the cash-strapped Omaha school district come up with the dough? Eubanks got a $364,000 grant from the Sherwood Foundation, started by Susan Buffett, Warren Buffett’s daughter.

Innovation in sports often follows calamity. Nearly 80 years ago, Riddell invented the plastic football helmet to slow an epidemic of broken skulls. Now, amid the growing realization of how rampant brain injuries are in the sport, the company is reimagining the football helmet, pushing beyond protection into detection. And that’s meant reinventing itself as something more like a tech company—not an easy task for an 88-year-old business built on manufacturing. “This is a very big bet for us,” says Thad Ide, who runs Riddell’s research and development arm. “We think that in a few years, it will be hard to buy a helmet without sensors. We’re all in.”

The InSite system, which Riddell rolled out less than five years ago, is now used by more than 1,000 football programs around the country—in youth leagues, high schools (60-plus in Illinois), and colleges. Additional teams use sensors by other manufacturers. (Riddell’s biggest competitor, Schutt Sports, based in Litchfield, in central Illinois, recently introduced them in chin guards.) Still, that represents only a tiny portion of the universe of football players—at a time when the game itself is under attack because of safety concerns. While adherents like Eubanks call sensors a blessing, there are skeptics. Some big-time college programs have rejected them outright, and the NFL, despite the heat it has taken of late for player safety, has so far snubbed them. The question is why.

Slim, fit, and angular, Thad Ide is, like nearly everyone at Riddell, a former high school athlete—track and baseball, in his case. After earning a bachelor’s in mechnical engineering and a master’s in applied mechanics from Michigan State, he worked briefly at Riddell in the early 1990s, then cycled through gigs at Schutt, Nike, and safety-testing labs that evaluated and designed helmets for, among others, riot police, firefighters, and athletes. In 2000, he landed back at Riddell, where he oversees product innovation.

As he walks through Riddell’s offices in Des Plaines (the company moved from Rosemont in June), there’s energy in his step—and maybe more than usual when he’s giving a tour of the hospital-clean new testing lab. White helmets, stacked from floor to ceiling on steel shelves, await their torture. Passing a couple of giant freezers, Ide explains that extreme weather can compromise both plastics and flesh, so some helmets get frozen before they get pounded here. Others get heated. Still others get treated with hair products.

Ide picks up one of the gray dummy heads that don the helmets for the tests and bounces it in his hand as if weighing a cantaloupe. “People are usually surprised how heavy a head is,” he says. (These weigh 15 pounds.) “Car crash dummy headforms are not this sophisticated. The one we use has a soft ‘flesh’ outer covering. Under the hard skull is a cavity filled with a material to mimic the gray matter of your brain.” Dummy heads in other industries do not simulate how the brain shifts in the skull upon impact. The head in Ide’s hands also has empty sinus cavities to mimic how the force of an impact travels from the torso and neck to the brain and to the neurons and nerve fibers that move information to the brain. “Everything is designed to resonate like a human head,” he says.

Next to Ide is the drop-test machine, a device that looks unsettlingly like a gallows. The operator is a large bald man in, gulp, all black. Skewered sideways on a post, a head donning a Riddell helmet is slowly raised about 60 inches over a steel stump. Though Ide has seen this test innumerable times, he still smiles wryly, because he knows that what happens next will make a visitor jump. A bell rings, and in a split second, the helmet crashes down. Boom! It hurts just to hear it. A severe hit in football has roughly the same physics, Ide says. Think of that next time you see a couple of All-Pro linemen slam into each other.

The machine attests to what an engineering miracle a modern football helmet is. Studies show that helmets with the highest safety rating can absorb about 50 percent of the force of a hit, which, at the concussion-causing high end of the spectrum, can deliver a g-force of 100 or more. Riddell, under Ide, was the first company to offer helmets with such effective impact reduction. In 2011, the year Virginia Tech released results of the first comprehensive independent test of helmets, only one, from Riddell, earned five stars, the highest rating. Now many do.

Stefan Duma, a biomedical engineer who founded the Helmet Lab at Virginia Tech, says that no helmet can entirely eliminate concussions and other head injuries. But, he adds, “200 live primate tests clearly show the higher the acceleration is on a brain, the higher the risk it gets injured. And all the research shows that the lower the acceleration, the lower the risk.” That’s where the newer, better helmets come in. Duma points to research on about 20 college football teams whose players were tracked with accelerometers in their helmets; higher-rated helmets reduced the risk of concussion up to 50 percent. That fact, he adds, is rarely, if ever, mentioned by manufacturers. They are too wary of litigation to publicly quantify any level of improvement, even if research supports it.

The drop test evaluates only how well helmets withstand linear acceleration, the kind of impact delivered by straight-on hits, like the blunt force you might experience running into a brick wall. That’s what football helmets dissipate best. But in reality, linear acceleration is nearly always joined by rotational acceleration, which results from the jerking of the head and the movement of the brain inside it. That, in turn, causes damage to neurons and nerve fibers.

Next on Ide’s tour is the linear impactor, also known as the pneumatic ram. It’s a robotic punching arm that shoots into a helmet the way a SWAT team knocks down a door. The helmeted dummy head sits atop a bendable metal neck that flexes according to the direction and force of the blow. Despite its name, the machine simulates rotational acceleration as well. The ram, which strikes lightning fast, shows how violent a football hit can be on the head and neck. Watch closely and you’ll see the metal neck ricochet and the head whipsaw like a boxer’s speed bag.

As good as the modern helmet is, its protection has limits. “There is only so much you can do with all the materials known on the planet and the small space between the helmet shell and the head,” says Duma. “You might see a helmet in the future that’s 5 percent better, but unless somebody discovers a new miraculous material, you probably won’t see much more than that.” Enter sensors and the industry’s push into impact-detection systems.

To understand where Riddell is going, you first have to understand how it got where it is. The company was born from a revolutionary invention. In the early 1920s, John T. Riddell, a math teacher and football coach at Evanston Township High School, looked for a new way to build traction in his players’ footwear. Back then, players relied on cobblers to nail leather nubs into the soles of boots they also wore on the street. The nubs worked poorly and wore out quickly, and the boots weren’t built for speed. Riddell invented the screw-in metal cleat and, with the help of Bears coach George Halas, athletic shoes that shifted players’ balance forward. In 1929, he set up his namesake company to make both.

In 1939, Riddell and son John Jr. designed the first plastic helmet as an alternative to the ineffectual leather headgear worn by players of the era. It took 10 years of modification—and the end of World War II to free up supplies of plastic—before Riddell produced a helmet that met the NFL’s approval. The product alleviated some of the dangers of the game, but hardly all. In 1968, a particularly fatal year, 36 players, across all levels of football, died from injuries, mostly to the head. Riddell and other manufacturers responded with better-padded helmets. That cut the death count dramatically.

Saving skulls, however, had unintended results. When the head is well secured in padding, the force of an impact moves to the soft matter inside the skull, leaving the brain still vulnerable. For helmet makers, negotiating that tradeoff between cracked bones and racked brains has been bedeviling—in part because the physics of the helmet can do only so much. The upside is that it has led to more innovation.

In mid-2003, Riddell’s current owner, the private equity investment firm Fenway Partners, bought the business for $100 million. The Riddell family had shed its interest in the late ’70s, and before Fenway came along, the company had changed hands five times. Even under Fenway, Riddell was combined with and separated from other sports equipment companies, including Bell and Giro. Now, as the last helmet maker in Fenway’s stable, Riddell is banking on sensors—on its InSite system in particular—as its future.

Riddell rolled out its first version of sensors in 2004 for research purposes. A New Hampshire company called Simbex was exploring how to apply the accelerometers that trigger automobile airbags to other protective gear, primarily for the military, and wound up partnering with Riddell. The sensors were added to the helmets of football teams willing to help collect impact data for researchers.

Once it had a critical mass of this data, Riddell designed a new generation of helmets, including the SpeedFlex in 2014. One big change was the material. The rigid shell was gone. The SpeedFlex is made from a polycarbonate plastic that yields on impact and spreads the force of a hit more effectively around the helmet and through the pads. Data even influenced the design of the facemask, which is now made of a lighter and stronger stainless steel.

The most noticeable feature of the SpeedFlex, however, is a groove, shaped like a jeans pocket, cut into the top front section. This contour, the thinking goes, allows extra give in the place that takes the most hits. In the next few years, Riddell plans to create helmets designed for specific positions, based on the kind of hits those players tend to take.

Riddell will continue to push the evolution of its helmets, but that is now inseparable from its move into sensors and software. The company’s initial effort to mass-market sensors came in 2013 with the introduction of InSite, which was a fraction of the cost of the old sensors and, except for annual battery changes, nearly maintenance-free. The first generation of InSite sensors was designed to collect and transmit data on just the most dangerous hits—the top 5 percent in terms of an aggregate of factors, including impact and duration. The current system can measure any hit. The company just released a web-based platform that football teams will be using next season. With it, they can analyze an assortment of data online (see “How the Sensor System Works,” above).

The shift to technology opens up the helmet, and the game itself, to more possibilities. In years to come, it may be possible to identify biomarkers—chemicals found in sweat and saliva, for example—that can actually diagnose trauma as it happens. Several labs around the country are working on that, and Riddell is funding some of the research. How exactly the company might engineer assays into helmets is still an open question. “I see the sensors now as part of a bigger process,” says Virginia Tech’s Duma. “We are going down a direction where everyone will be instrumented in every sport. Like everyone has cell phones.”

But to make that happen, at least in football, the NFL needs to get on board. Though its players make up only around 1,600 of the nearly four million Americans who participate in football, the league has an outsize influence. (Riddell supplies about 65 percent of NFL players with helmets; it has 90 percent of the college market and more than half of the youth and high school segment.) Until a couple of years ago, it looked as if the NFL, which had underwritten academic studies on the use of impact sensors, was on the verge of adopting them in helmets. Then, in 2015, the league suspended its sensor program, claiming that it had concerns about the accuracy of the data.

Duma, for one, rejects that explanation. “Alerts [from sensors] just provide team trainers with additional pieces of information.”

So why else might the NFL be dragging its feet? A recent visit to a Division I college football program offers a clue. At this Midwest school, money seems no object. The players are outfitted with Riddell’s latest (and most expensive) helmet, SpeedFlex Precision, which features customized padding for every player based on 3-D laser scans of his head. Each helmet costs around $1,750. None here are equipped with sensors.

Why not? The program’s head athletic trainer initially asserts that sensors are simply unnecessary: “I can tell every player by his walk. I know how they do in class. I see them day in and day out. If there’s any change in them, we know.” What’s more, he points out, during a game there are five trainers and three doctors on the sideline, along with one or two spotters who watch every play in the broadcast booth and then again in replay on TV, all looking for nothing but injuries.

Sounds reasonable. But then, official interview apparently over, the trainer explains, in a confidential tone and not for attribution, that there is another reason the team doesn’t use sensors: “You cannot predict how all that data might get used against you in litigation.”

And there’s the rub. It’s not a stretch to figure that the NFL shares similar concerns, especially having agreed last summer to dole out a reported $1 billion to settle a lawsuit brought by retired players claiming brain injuries. Says Duma diplomatically: “There’s no question that there are nontechnical issues on sensors.”

Fear of litigation isn’t the only such issue. The players themselves might prove to be an obstacle in getting sensors into NFL helmets. Impact data collected by teams could threaten their playing time, their negotiating power—their livelihoods. Look what happened when teams started using sensors just to monitors players’ sleep. The NFL Players Association filed a grievance in 2015, arguing that the practice was too intrusive.

Players’ distrust of the league runs deep. “The NFL doesn’t give a shit about any one of us,” says Otis Wilson, linebacker for the legendary 1985 Chicago Bears. “Anytime you’re injured, the teams will use that against you.” Meanwhile, he doubts that teams will ever share the impact data with players, even in retirement, when they may be searching for explanations for cognitive impairments. “You’re going to have to fight for that information in the end.”

Coming on the heels of an unseasonably warm stretch, the crisp, cool early-November day feels, at last, made for football. At Stevenson High School in Lincolnshire, a perennial football power, the Patriots are preparing for their first playoff game of 2017, the team’s 29th straight year of postseason appearances. The players are all hustle. They thump, they clap, they yell. These guys are wired—in more ways than one.

Four years ago, the school began to shop for ways to make the game safer for its players. A North Shore tech startup was the first to offer solutions. “We started with a padding group,” says Tyler Kollmann, one of Stevenson’s four athletic trainers, “and also looked at a headband with sensors that was used for soccer but perhaps could have been fit inside a helmet. We also tried caps that go over helmets [to soften impact], but what would happen if the caps got soaked in the rain? Our players may not have the shoulder and neck strength for the extra weight.”

In the end, the school turned to Riddell’s InSite system. Stevenson has been buying a few sensor-equipped helmets every year and is now up to more than 30. Which players get them? “It’s a conversation we have in the training room,” says Kollmann. “It’s based on who needs them because of the vulnerability of their position, or maybe they have a health history.”

Just then, a large player, probably a lineman, runs over to the trainer’s cart. He removes his helmet and holds out a cut hand. As Kollmann bandages it, the player offers a look inside the helmet at the InSite membrane and radio transmitter. He says he never notices either during play. Then he runs back to join the team.

Stevenson was one of InSite’s earliest users, which made it a valuable test case for Riddell. “It’s been evolving,” Kollmann says of the system’s features. “At first it didn’t give enough data. The tech had not caught up with what I wanted. Then, over the last year or two, it began to offer more information on different hits, and by player, and let us track them. When we send the helmets back for reconditioning this year, we’re going to get the next generation, and it will have all kinds of data and analysis. We’re going to be able to see if a linebacker is getting more hits than an offensive lineman, and to see which of our players need more coaching.”

How effective is the system in warning of dangerous hits? “It provided less than 10 alerts over the season,” Kollmann says.

Ironically, none of those hits resulted in diagnosed concussions. Yet in 2017, the Patriots reported 10 to 15 concussions, all flagged not by InSite but in more traditional ways—by players coming off the field or coaches or trainers noticing something wrong. Still, Kollmann hasn’t lost trust in Riddell. “Not every concussion is from a top 5 percent impact,” he says. “And some are from multiple lower-intensity hits.”

In Omaha, though, InSite alerts did lead to concussion diagnoses in 2017, Eubanks says. And the system is changing the way coaches there approach the game. “We can already pull up players by position,” Eubanks says. “If we see the number of hits for linebackers is very different from that of defensive backs, we can work with that.” That might mean reinforcing safer tackling techniques or running different drills in practice to reduce impact. “The helmets have been a tremendous tool.”

As for Brandon Steburg, just a month after his injury, he showed up at school to attend a pancake breakfast fundraiser for his medical care. He wore a light helmet to protect his head. He told friends he might try track in the future but he was done with football.