It happened again. Just when it looked like we might get through a Sunday without an NFL quarterback getting injured for the first time in 2026, Tampa Bay's Baker Mayfield went down in pain against the Vikings. On a day when Mayfield was sacked six times and knocked down on four more dropbacks, a glancing blow proved to be the most painful hit, as the Bucs starter suffered a dislocated thumb while trying to get the ball out on fourth down. The injury is expected to cost Mayfield about three weeks, which only makes life harder for the winless Buccaneers, who will now turn to Jalon Daniels in the meantime.
Then, on "Monday Night Football," both teams dealt with quarterback injuries. The Bears started Case Keenum after Caleb Williams injured his hamstring and Tyson Bagent spent most of the week in concussion protocol. The Eagles briefly lost Jalen Hurts to a concussion check of his own, and while he quickly returned to the game, he was out of the lineup just long enough for Andy Dalton to throw a critical interception in the end zone to end an Eagles drive.
Depending on who you define as each team's primary option, Mayfield was somewhere between the sixth, seventh, eighth or even ninth starting quarterback to go down so far this season. The Bears likely would have started Bagent if not for the time in concussion protocol, but he didn't actually miss any action. Hurts was back quickly, as Dalton threw only one pass. You could argue that the Falcons would have started either Tua Tagovailoa or Michael Penix Jr. in Week 1; both were injured, leaving them to start Cooper Rush instead.
The NFL made it exactly zero drives without a starting quarterback going down this season, as Sam Darnold suffered a glute injury before the end of Seattle's opening possession in the season opener. He's already back after missing a couple of weeks, as is Kyler Murray, who left quickly in Week 1 because of a concussion.
Subscribe: 'The Bill Barnwell Show'
The news was good on Jayden Daniels (elbow) and Williams, who luckily avoided long-term injuries. Giants fans were less fortunate, as early suggestions that Jaxson Dart would be OK eventually turned more ominous. Dart underwent knee surgery to end his season last week. He's currently the only starting quarterback lost for the year.
So, to recap: Seven quarterbacks have been forced out of games by injuries through the first three weeks of the NFL season. Is that a lot? Is it a problem that's getting worse? Are QBs getting hit more often? Are we seeing more backup quarterbacks than ever before? Should teams be investing more in those QB2s as a result? These are all legitimate questions to ask. So, I actually did the research this week to try to figure out how often quarterbacks get hurt early in the season.
Going back through 2002 -- when the league expanded to 32 teams -- I looked at where every team and their preferred starting quarterbacks stood through the first three weeks of the season. I looked at every team that had two or more quarterbacks throw at least five passes through those first three weeks to see how often teams had been forced to turn to their backups for even a modest number of snaps. Let's see what history says about all these quarterback injuries.
Jump to:
Are we seeing more QBs than usual?
Are there more QB injuries than usual?
Why are so many QBs getting hurt?
Are QBs getting hit more than ever before?
Should teams invest more in QB2s?
Are we seeing more quarterbacks playing than usual?
Not really. Through three weeks, 41 quarterbacks have thrown at least five passes. That's up from a year ago, when 40 quarterbacks were on the board through three weeks, and tied for where we were in 2024. There was a four-year stretch between 2020 and 2023 when we saw fewer quarterbacks than normal, including just 35 passers appearing through three weeks in 2022.
Since 2002, though, the average has been just over 39 quarterbacks with at least five pass attempts over the first three weeks of the season. And that number hasn't even changed all that much since the league emphasized roughing the passer rules heading into the 2018 season. Teams have averaged 38.9 quarterbacks through three games over that span.
If it seems like we see more backup quarterbacks on a weekly basis than we might have 20 or more years ago, that's likely a product of simply seeing more football overall. Over the first three weeks of the season, the NFL put on 10 primetime "island" games to go along with widely available access to Sunday Ticket via streaming. In 2006, there were eight island games through three weeks, and Sunday Ticket was available only via satellite. In 1986, the league had five island games through three weeks, with no way to watch otherwise out-of-market games. We see more of every kind of quarterback than we did in the past.
Yes, 41 quarterbacks is slightly more than the historical average but hardly out of line with what we've seen during the 32-team era. And if you go back even further into the past, coaches were cycling through QBs on an even more regular basis. In 1999, for example, 47 different signal-callers threw at least five passes through the first three weeks of the season despite the fact that there were only 31 teams in the league at the time. The same number popped up again in 1995, when the NFL had 30 teams. Was that a product of an era when defenses had more latitude to hit quarterbacks or coaches trying to find a better answer under center?
Are we seeing more quarterback injuries than usual?
Yes. This is the most injured season through three weeks we've seen from NFL quarterbacks in recent memory. Defining an injury depends on your methodology, but in terms of quarterbacks who would have been in the lineup if healthy but weren't able to make a start, or backup quarterbacks who were forced into the lineup to throw at least five passes because the quarterback ahead of them on the depth chart wasn't healthy enough to play, the league has dealt with eight injured passers through three games. (While he hasn't yet thrown five passes, I'm counting Jalon Daniels as a backup who has been forced into the lineup, because we know Mayfield will be out for a few weeks because of his thumb injury.)
Again, the evidence suggests that while the league is on the higher side relative to historical averages, it's not an outlier or indicative of a broader trend. Eight injuries to quarterbacks through three games is up from five in 2025 and four in 2024. The 2022 season was an outlier, with just two injuries through three weeks, but that's the lowest figure we've seen since 2002.
The last time eight quarterbacks were unavailable by our criteria through the first three weeks of the season would have been all the way back in 1999. There were seven quarterbacks sidelined through three weeks of the 2019 season, a grouping that included Andrew Luck, who dealt with a calf injury all summer and then shockingly retired because of the ailment just before the start of the campaign.
Over the course of the past 25 seasons, an average of almost exactly 5.0 quarterbacks have been injured by the end of Week 3. You can split that up into 4.6 quarterbacks who have been hurt during the regular season and an average of 0.4 quarterbacks who were injured and unavailable heading into Week 1.
Depending on whether you consider Penix or Tagovailoa the preferred starter in Atlanta, we would have had either zero or one injured starter heading into the season, which isn't always the case. In 2016, for example, Teddy Bridgewater and Tony Romo suffered serious injuries before the season began. (These numbers don't include situations involving players who were holding out to begin past seasons, such as Deshaun Watson and Carson Palmer, or players who were suspended to open the season, such as Watson and Ben Roethlisberger.)
Since 2018, the NFL has ended Week 3 with an average of about 4.7 quarterback injuries. And of course, there have been seasons when multiple starting quarterbacks have been lost for the year before the start of Week 4, as was the case when Cam Newton and Roethlisberger went down with season-ending injuries in back-to-back weeks during the 2019 season. So far, Dart is the only quarterback expected to miss significant time because of his injury.
OK, but why are there more quarterback injuries than any other time in recent memory?
Great question! My answer is probably not going to be very satisfying: I believe it's just ... randomness. And the reason has to do with the nature of those injuries and how and when they've occurred. There has been a really wide range of injuries to quarterbacks this season. Let's quickly run through them:
Penix tore an ACL last season and was still recovering when the 2026 season began. He's now healthy enough to play. The ACL tear came when Penix was blitzed off the edge and brought down by a tackle.
Tagovailoa sustained an oblique injury during a passing drill in practice. We don't have footage of the injury, but this doesn't sound like a contact injury.
Williams pulled a hamstring while being chased down on a naked bootleg by Blake Cashman of the Vikings. It looked in the moment as if Cashman might have brought down the Bears star, but the All-22 angle makes it clear that Williams was not touched on the play.
Murray sustained a concussion at the end of a scramble on a hit by Packers safety Javon Bullard, who was flagged for unnecesssary roughness but not fined after the game.
Dart was at the end of an extended dropback and hit in the knee as he released a pass, suffering a season-ending knee injury in the process.
Darnold was sacked as he attempted to leave the pocket at the end of a dropback and sustained a glute injury. This was a more traditional injury.
Mayfield was hit as he tried to scramble away and release the ball from yet another Brian Flores-schemed pressure on fourth down. While Mayfield was knocked down on the play and took plenty of hits throughout the game, the injury came from his hand striking Cashman's arm as he followed through on his pass attempt. This was a contact injury, but it would be difficult to classify it in the same bin as the injuries sustained by Penix and Darnold.
Daniels had his foot stepped on by guard Chris Paul as he took a snap, fell down, tried to break his fall with his arm and dislocated his elbow. Is this a contact injury? I guess, technically?
That's eight players. Four (Penix, Murray, Dart and Darnold) sustained the sort of contact-driven injuries we would typically think of when a quarterback gets hurt. Daniels, Mayfield and Williams were under some semblance of duress or pressure, but their injuries look more like bad luck than anything else. And Tagovailoa injured himself during a drill. Not every quarterback injury in the past is a product of a direct hit, but this does seem like an uncommon number of freak injuries or ones that weren't a direct product of a tackle.
Then again, go back to 2019 and you'll see what might be an even weirder or unlikely group of injuries. Nick Foles (clavicle) and Trevor Siemian (ankle) had serious injuries on direct hits. Drew Brees injured a thumb when he hit it on Aaron Donald's hand during a follow-through, much as Mayfield did this past week. Two of the injuries were more long term, as Cam Newton (Lisfranc) and Ben Roethlisberger (elbow) were dealing with issues that were aggravated during Week 2 without being attributable to a specific play. Luck strained his calf during the offseason, and it never healed to his satisfaction. And Darnold was sidelined by mononucleosis.
Are quarterbacks getting hit more often in 2026?
Yes. According to ESPN data, quarterbacks were contacted on about 16% of plays in 2006, which is the first year for which we have data. That has steadily risen over the ensuing 20 years, owing at least in part to quarterbacks running more often and attempting to extend plays while creating outside of structure. Quarterbacks have been hit on more than 20% of plays every year since 2020, and so far in 2026, that number is up at a record high of 22.2%. While going 16% to 22.2% over 20 years might not sound like a massive difference, that's about 1,220 more hits per season for quarterbacks to endure.
So, if hits and injuries are at record highs, it makes sense that one is causing the other, right? No, the correlation between hit rate and quarterback injuries going back through 2006 is minus-0.29, suggesting that there isn't much of a relationship at all. Contact rate and injuries are up this season, but contact rate was also at relative record highs between 2021 and 2024, when quarterback injuries were happening less often than historical averages. And 1999 -- the prior high-water mark for quarterback injuries -- was in an era when QBs were likely getting hit less often.
All hits aren't created equal, either. Quarterbacks have been afforded stronger protections over that timeframe, including the "Brady Rule" that barred defenders from targeting quarterbacks below the knees in 2009 and further protection against defenders using their full body weight to sack quarterbacks in 2018. Runs between the tackles are often more dangerous and result in hits from bigger players than scrambles to the sideline.
If we assume that hit rates were lower in the 1990s, the fact that we saw so many backup quarterbacks making appearances through the first three weeks further hints at no relationship between volume or frequency of hits and early-season QB injuries. Even with 30 (or 31 teams) for most of the decade as opposed to the 32 teams the league has played with since 2002, we saw an average of just over 41 quarterbacks see the field between 1990 and 1999.
Let me be clear: It's difficult to imagine a scenario in which being hit more often is better for quarterbacks, and I'm not suggesting that's the case. But there's no relationship between the sheer volume of hits the broader QB pool faces and how often they've been injured early in seasons so far.
One alternate argument could be that quarterbacks and offensive linemen don't play in the preseason anywhere near as often these days, which could lead to rusty protection schemes and quarterbacks who aren't quite as prepared for the season as they might have been in years past. That could certainly be reasonable, but again, that has been a trend for years in the NFL. The Rams have mostly treated the preseason as vacation time for their starters under Sean McVay, and it hasn't led to an increase in injuries before this season.
Quarterbacks who aren't suiting up in preseason games aren't getting hurt in those games, which is great news. But those quarterbacks can also get injured during training camp practices, as we saw with Tua Tagovailoa this year and players such as Teddy Bridgewater and Ryan Tannehill in the past.
There's no doubt that quarterback play is evolving. They take more hits, scramble more and extend plays longer than quarterbacks did in years past. A decade ago, about 67% of quarterback sacks took place within four seconds of the snap. In 2026, that has fallen to 54%, with signal-callers getting better at escaping pressure and resetting to find receivers. The league's emphasis on the boot and naked game also creates natural space for passers to extend plays as they run toward the sideline, although defenses have gotten more aggressive at pursuing quarterbacks and punishing offensive architects for leaving defenders unblocked on those concepts.
Simultaneously, though, early-season injury rates for quarterbacks really had not risen before this season. If anything, 2021 to 2025 was the healthiest five-year early-season stretch for quarterbacks of the past two decades. And while it's tougher to find reliable information on injuries stretching into the mid-1990s, I don't see any evidence that starting quarterbacks are less likely to stay on the field now than they were 30 or 40 years ago.
Should teams be investing more in their backup quarterbacks?
Did you watch Keenum against the Eagles on Monday night? The 38-year-old quarterback, a legendary Air Raid passer when throwing to Donnie Avery during his time in college at Houston, picked the Eagles apart with short completions and the play-action game in a 27-7 victory. We even got to see the "sucker sneak," with the Bears pulling both guards and freezing the Eagles long enough for Keenum to sneak right past two unblocked defensive tackles for a 1-yard touchdown plunge. It was beautiful.
Of course, you can go two ways from that performance. Does Keenum's success coming off the bench as Chicago's third-stringer prove that you need to have a QB2 you can trust? Or does it suggest that the way you win when you don't have your starting quarterback is through a great offensive architect and a smart game plan? Keenum, after all, didn't look anywhere near this good when he started twice for the Texans three years ago, posting a 61.1 passer rating across two starts.
I'd argue there's some truth to both sides of the coin. And like everything with the NFL, there's no one-size-fits-all answer. Teams with Ben Johnson or Sean McVay as their offensive architects probably have a lot more leeway in addressing the backup quarterback position than teams with less insightful playcallers. Organizations paying $60 million per year to their starter are already pricing in some confidence that their signal-caller will stay healthy and probably can't afford to commit high-end money to a backup, as well.
That backup quarterback market, though, hasn't really risen much at all over the past decade. In 2016, the Eagles signed Chase Daniel to a three-year, $21 million deal, originally to back up oft-injured starter Sam Bradford. Instead, Howie Roseman then pivoted, drafted Carson Wentz and eventually dealt Bradford to the Vikings for a first-round pick. Daniel ended up making $7.9 million for one pass attempt with the Eagles.
That $7 million average salary on Daniel's contract represented 4.5% of the league's salary cap at the time, which was $155.3 million. In 2026, the cap is $301.2 million, which is almost double what it was a decade ago. Salaries at most other positions have risen in kind over that same time frame. Daniel's deal in 2026 money would be closer to $13.6 million per season.
Defining backup quarterbacks can be in the eye of the beholder. There are teams that sign a quarterback to a deal knowing they're going to draft a signal-caller in the first round, such as the Raiders with Kirk Cousins this year. There are others that have a need at quarterback and take a chance on a player who is unable to win that job long term, such as Mike Glennon with the Bears and Mitchell Trubisky with the Steelers. I'm not counting those contracts as backups.
If we look at quarterbacks who were specifically signed for a backup role without a meaningful path to a starting gig, the backup QB market has mostly stagnated over the past decade. The only backups to make more than $7 million per season on a deal over that time are Jacoby Brissett and Marcus Mariota, who have each done so twice. Mariota's two-year, $17.6 million deal with the Raiders in 2020 was essentially the same as Daniel's deal after adjusting for cap inflation, but that's the only other example of a backup passer making 4% or more of the salary cap as an average salary over the past decade.
Even as the cap has nearly doubled, teams have collectively proven unwilling to pay backups more than the $8.8 million average salary Mariota earned with the Raiders during the 2020 season. In fact, Mariota earned $8 million on a one-year deal with the Commanders in 2025, played reasonably well over eight starts while filling in for an injured Jayden Daniels and ... took a $1 million pay cut to re-sign with the team in 2026. He has again stepped in and led them to a victory over the defending champion Seahawks.
Every organization will say it's important to have a trusted backup, but what teams do with their decisions is more meaningful than what they say. Teams might believe that things are hopeless if they don't have their starting quarterback, but we're not too far removed from Nick Foles replacing an injured Wentz at the end of the 2017 season and leading the Eagles to a Super Bowl.
There's no right or wrong answer here, but if quarterback injuries do spike and this isn't just a random one-year blip, I suspect we'll see organizations get more aggressive in valuing backups in the years to come. How many teams wish they had spent more than $6 million to sign Joe Flacco right now? The Giants just sent a fifth-round pick that projects to be worth $7.4 million in surplus value over the next four seasons to acquire J.J. McCarthy, who was mostly dismal in his 10-game stint as an NFL starter last season.
As organizations perpetually work to find potential places of value on their roster, backup quarterback might be an untapped or underappreciated mine, both in terms of potential trade value and what they might offer a team if its starter does go down. And it's a bit of a surprise that the backup quarterback market hasn't risen at the same level as the cap over the past decade.
