Anatomical simulator · Football head trauma · Evidence as of Oct 2026
Padded helmet caps are sold on a headline: fewer concussions. But a concussion is a rare, extreme hit, and CTE tracks the running total of every hit a player ever takes. A small softening of all hits can cut the rare ones sharply while barely moving the total. These figures show where that gap opens up, from the skull inward.
The brain sits in fluid inside the skull. When the head is spun by a hit, the skull turns first and the brain catches up a few milliseconds later. That lag is what strains brain tissue, and padding outside the skull cannot remove it.
| Per impact | Helmet | + Cap | Change |
|---|---|---|---|
| Peak head spin acceleration, rad/s² | |||
| Change in head spin speed, rad/s | |||
| Brain–skull lag (strain proxy) |
Every hit a player takes lands somewhere on this curve. Nearly all are routine. Concussions come from the thin right-hand tail, where the risk curve finally climbs. Shift the whole curve left by a little and the tail shrinks by a lot. The total, which is what CTE tracks, shrinks by only that little.
Read it this way: a drop of about half in concussions is what you would expect from a modest across-the-board softening. It does not mean half the brain trauma went away. Impact distribution and risk curve are fitted to the Virginia Tech–Brown–Dartmouth helmet-sensor dataset of 301,034 college head impacts (Rowson et al. 2012). Tail shading is risk × frequency, scaled up so it is visible.
The same player, the same seasons. On the left, a standard helmet. On the right, a cap in practice. The microscope view is the depth of a single brain fold, where CTE's defining lesion forms: abnormal tau protein, stained brown, gathered around a small blood vessel. Deposits here scale with lifetime load, not with concussion count.
Brain donors in the 2023 CTE study played about 12 years on average.
Most college players log 200–400 sensor-recorded impacts a season; linemen and linebackers have topped 1,400.
Caps are worn in practice, not games. Sensor studies count fewer hits per practice than per game, but far more practices; the default puts about two-thirds of hits in practice. Set 100% for full-time wear.
The largest on-field study found no change in hit count, but did find a shift toward hits to the back of the helmet. Raise this to see how little behavior change erases the gain.
What this shows: with the NFL cap's best lab figure, worn in practice, the model removes roughly one season's worth of load from a 12-season career while cutting concussions by about a third. With the on-field measurements, it removes nothing. Cutting contact itself moves the total far more. The CTE odds figure applies the association in Daneshvar et al. (2023), about 22% higher odds per million rad/s² of cumulative rotational acceleration, to the change in load. That study is observational, so treat the number as a scale, not a prediction for any person.
Helmets have changed a lot since the hard shells and dense foam of the 1990s and 2000s, and unlike the cap, the better ones have on-field evidence behind them. They still work the way the cap does: they soften each hit, they do not remove hits. Run through the same tail-and-sum model, a helmet that cuts concussions by half trims the average hit by only about an eighth.
| Change | On-field evidence | Concussions | Lifetime load | Covers |
|---|---|---|---|---|
| Guardian Cap, NFL model | No measured change in on-field head impacts; no link for concussions from hits to the shell | Practice only, at your Fig. 3 settings | ||
| Top-rated helmet vs a 2000s design | 54% lower concussion risk across 1,833 college players, Riddell Revolution vs VSR4. In the NFL, helmets now rated Not Recommended had about 30% more concussions than the top-ranked ones. | −54% observed | Every hit, practice and games | |
| Halve full-contact practice | Contact limits, with athletic trainers on staff, are among the high-school measures tied to 64% fewer concussions | Fewer hits overall; not modeled | Removes hits outright |
Shells built to give on impact, like a car bumper. VICIS's ZERO1 introduced a soft outer layer that deforms; Riddell's Axiom adds flex panels and openings to its shell.
Evidence: graded inside overall lab rankings. No on-field study isolates the shell itself.
Printed elastomer struts replace foam blocks. Riddell's Diamond liner has more than 140,000 struts tuned for linear and rotational energy. VICIS designed its lattice to start stiff and then buckle, absorbing energy as it collapses.
Evidence: lattice-lined models top recent NFL lab rankings. Performance claims for the lattice itself come from the makers.
Linemen take most hits on the front of the helmet; quarterbacks often hit the back of the head on the turf. VICIS's ZERO2 Trench (2021) adds front support for linemen, and the NFL now runs separate quarterback and linemen tests.
Evidence: nearly 30% of NFL linemen and quarterbacks wore one in 2024. Rankings are lab-based.
Riddell's fitting app scans 285 points on a player's head to shape the pads, replacing hand-inflated air bladders. A helmet that stays put protects as tested.
Evidence: sensible, but the NFL's rankings state they do not test fit, retention, temperature or long-term durability.
What none of them change: the number of hits. The injuring motion is mostly rotation, which shears tissue inside the brain; the brain does not need to slam into the skull to be strained. The NFL's own ranking poster warns that no helmet system can completely protect against serious brain injury, and that its results should not be applied to college, high school or youth football, where older and reconditioned helmets are common. The helmet row's lifetime load is inferred by running the observed 54% concussion drop backward through the Fig. 2 model.
The NFL cut practice concussions in half with the cap.
The league's own peer-reviewed analysis found a 54–62% drop for required positions. But for concussions from hits to the helmet shell, the part the cap covers, it found no significant link to cap use, and the authors said the drop could not be credited to the padding alone. Practice rules, technique coaching and better-rated helmets changed over the same years.
Lab tests prove the cap absorbs the hit.
In Virginia Tech's helmet lab, the NFL model cut rotational acceleration 14% and linear 8%; the lighter youth and high-school model cut them 5% and 3%. Players are not lab headforms. Instrumented-mouthguard studies of college practices have not found a difference, the largest covering more than 7,500 impacts with the same count and force of hits for wearers and non-wearers.
Virginia Tech Helmet Lab · Lynall & Schmidt, ACSM 2025 · Univ. of Nevada, Reno 2023
Caps are protecting high-school players.
Across 41 Wisconsin high schools and 2,610 players in the 2023 season, practice concussions were 33 among cap wearers and 31 among non-wearers, no statistical difference. The athletic trainers' national association said in 2022 that the evidence on helmet add-ons showed no benefit.
Fewer concussions means less CTE.
In 631 brains of former players, the number of concussions was not associated with CTE. The estimated number and force of all head impacts was: each 1,000 additional hits raised the odds of CTE by about 21%, and each added season by about 15%. In a mouse model, impacts produced early CTE-type tau changes with no concussion signs at all.
Daneshvar et al., Nat Commun 2023 · NIH summary · Tagge et al., Brain 2018
The padding keeps the brain from getting rattled.
Padding lowers the peak by stretching the impact out. The brain still lags the skull, and its strain depends on how much the head's spin speed changes, not only on the peak. Computer models of the head put the highest strain at the depths of the brain's folds, the same place the consensus definition of CTE puts its defining tau lesion.
Ghajari et al., Brain 2017 · NINDS/NIBIB criteria, Bieniek et al. 2021 · Gabler, Univ. of Virginia
Extra padding can only help.
Cap wearers had 36% lower odds of a facemask impact and 152% higher odds of an impact to the back of the helmet. The researchers call this preliminary, and they also found no rise in impact force. The pattern suggests the cap may change how players use their heads, which is the opening for risk compensation shown in Fig. 3.
New helmets stop the brain from hitting the skull.
The best current helmets do lower concussion risk, with on-field data to show it. But the brain is strained mostly by rotation, which a helmet can soften and cannot stop, and no helmet changes how many hits a player takes. The NFL's ranking poster says plainly that no helmet system can completely protect against serious brain injury.
Rowson et al., J Neurosurg 2014 · NFL helmet results 2025 · NFL/NFLPA 2026 poster
Every modeled number on this page comes from one of three small models built on published values. They are teaching models: transparent, simple, and meant to show direction and scale.
1/(1+e^-(−12.531+0.00196·α)), 50% at 6,383 rad/s², reproduces the observed rate of 57 concussions in 301,034 impacts. The cap multiplies every hit by (1 − reduction). With other fits we tried (log-normal matched to published percentiles, gamma), a 14% reduction cut modeled concussions 48–63%; the tail effect holds.