Two Weeks Into NFL 2026 and the Quarterback Shock: What Lies Beneath the Surface of the Number 38?

On September 15, 2026, as Week 2 of the NFL season closed, a number began spreading across sports feeds: 38 quarterbacks had thrown at least 5 passes across the first two weeks. That is the highest mark since 2026, and just one short of the record 39 set in both 2026 and 2026. Sports Illustrated called it an “astonishing list” of injured stars. ESPN’s Bill Barnwell cited the figure as a warning sign about the sport’s most important position.

That is where I stop. Data is the surface layer, and I always dig three more layers.

Context: a different league, a different frame of reference

Before getting to the number, it must be placed in its proper geological layer. The NFL is the American professional gridiron league, with 32 teams, each built around a quarterback — the signal-caller and thrower, the position seen as having the highest leverage in the sport. When a quarterback is injured, an entire offense must reorient. Unlike soccer, where a central midfielder can be replaced by someone broadly comparable in skill, a backup quarterback is almost always markedly worse at reading the game, throwing accurately, and deciding quickly under pressure.

In the first two weeks of the 2026 season, events arrived in a rush. Sam Darnold suffered a muscle injury. Kyler Murray entered the league-mandated concussion protocol. Caleb Williams had a hamstring issue. Jayden Daniels dislocated his elbow — for a quarterback, an injury that can span multiple weeks or even a season. Brock Bowers, a rising young tight end, required meniscus surgery. The list ran long, and American media called it a “particularly complicated” Week 2.

Six teams — the Minnesota Vikings, Atlanta Falcons, Chicago Bears, Seattle Seahawks, Washington Commanders and New York Giants — were already using a backup quarterback within two weeks. In the language of American analysis, this is the “next-man-up” phenomenon at the most sensitive position. Each of these teams must reshape its offensive identity mid-season, while its playbook is still in the gelling phase.

But this is where I must be careful. I do not excavate stars; I excavate context.

Core analysis: three layers beneath the number 38

The first layer is the question of the counting threshold. The figure of 38 is established by a criterion of “at least 5 passes.” That is a very low bar. It does not measure who the starting quarterback is. It measures everyone who threw the ball at least five times — including wildcat passes, gadget throws from non-specialists, and mop-up appearances at the end of games. In other words, it is a participation indicator, not a performance indicator. A number measuring “appearances,” inflated by a low threshold, cannot be read as “38 teams lost their starting quarterback.”

The second layer is the cross-era comparison. The 39 from 2026 and 2026 is placed next to the 38 to create a sense of “unprecedented.” But the article itself concedes that earlier seasons had fewer teams. As the league expanded, the number of rostered quarterbacks grew, so comparing raw counts across eras is a basic methodological error. The conclusion of an “unprecedented crisis” loses part of its footing right there. A fair comparison would normalize by team count or by play count.

The third layer is what is genuinely worrying. Not the number 38, but the fact that six teams had to field a backup quarterback in the opening month. In gridiron football, when a team loses its starting quarterback, offensive efficiency — measured by process metrics such as EPA per play or DVOA — typically drops sharply, and offensive coordinators tend to shift toward safer play-calling: running more, simplifying the script, reducing deep-throw risk. That shifts division dynamics from the earliest stage of the season, while teams are still finding rhythm.

Those three layers reveal a picture different from the sensational headline: there is a real wave of quarterback injuries, but the degree of “crisis” is inflated by how the number is packaged. A player is not a number, but the number is where I begin the excavation.

Contrarian angle: when the surface layer is contaminated

The most counterintuitive thing here is not the number but the reliability of the source itself. This is where I want to linger, because it bears directly on how I work.

On close inspection, I found entity and team errors. A.J. Brown, a Philadelphia Eagles wide receiver, was placed on the New England Patriots’ injured reserve list. Kyler Murray, the Arizona Cardinals’ quarterback, was attributed to Minnesota. Tight end Eli Stowers was slotted into the Eagles’ roster. These are not trivial details. They signal that the content may have been aggregated automatically or machine-translated without cross-verification.

In my trade, a data map can point the wrong way if you do not read the terrain. In 2026, as a senior specialist at the Viettel youth football academy, I once underrated a 16-year-old midfielder because his BMI and speed fell below the national U17 benchmark. I concluded he lacked the physical foundation. I overlooked one detail: he had just returned from a ligament injury and was in a growth-spurt catch-up phase. Three months later, he debuted for the first team in the V-League with 4 assists in only 5 matches. That mistake forced me to add a new column to my dataset: “biomedical context.” An injury does not erase a talent’s name; it only pushes that talent down into the sedimentary layer.

By the same logic, when an NFL report misassigns the team of a famous player, I cannot use it as a foundation for conclusions. I must treat it as a lead, not an established fact. Whether a number is right or wrong depends on whether the names and teams within it are right. When the surface layer is already contaminated, I do not rush to build on it. A lesson from 2026 at Song Lam Nghe An deepened this: when training grounds closed during the pandemic, I had to interview the family of an 18-year-old striker online and analyze archived GPS data, rather than look only at a 0.8-goals-per-90-minutes rate on paper. It was historical data, not the aggregate number, that helped me recommend a professional contract — and he scored 6 goals when the 2026 V-League kicked off.

Blind spot: injury as a market variable

There is another blind spot media tends to miss: the positional concentration of injuries. Six of 32 teams were already on backup quarterbacks after two weeks. If this trend persists, it is not merely an injury story — it is a market story. The quarterback position is the highest-paid within the salary cap. Every starting quarterback lost forces a team to sign a replacement or elevate a practice-squad player, eating into already-tight cap space. This is a “cap-efficiency erosion” effect that standings cannot measure.

But I must admit my limits. The report gives no cap figures, no recovery timelines for any injury, and no performance data. I cannot model financial impact out of thin air. Humility before new variables obliges me to say clearly: this section is conditional inference, not conclusion. A Jayden Daniels injury could be the single biggest variable of the season — or just a few weeks of rest. The report does not tell me, and I will not pretend to guess.

Takeaway

Two weeks is far too short to speak of a trend. Any conclusion about an “injury wave” at this point rests on a small sample, and the story will resolve within one to four weeks, when recovery timelines for Jayden Daniels and the other stars are published.

The lesson I draw is not about the NFL. It is about how we read numbers. In Vietnamese youth football, I learned that a statistical table is only worth something when you know by what threshold it was counted, by whom, and against what backdrop. I do not excavate stars; I excavate context. And sometimes the surface layer is not where you begin — it is where you must check before you believe.

If 38 teams had truly lost their starting quarterback, that would be a crisis. But 38 people who threw the ball five times? That is a different number. The question remains open: if the counting threshold were lowered to 5 passes for every season, would 2026 — with fewer teams — really have been less volatile than 2026? I do not yet have the data to answer. And I will not pretend that I do.