Comparing Impact and Concussion Risk in Leatherhead and Modern Football and Hockey Helmets : Neurosurgery

Secondary Logo

Journal Logo

Research—Laboratory: Sports and Rehabilitation

Comparing Impact and Concussion Risk in Leatherhead and Modern Football and Hockey Helmets

Huang, Jaxon J. BHSc*; Goya, Kellie N. BS*; Yamamoto, Brennan E. PhD; Yamamoto, Loren G. MD, MPH, MBA*

Author Information
Neurosurgery 92(6):p 1297-1302, June 2023. | DOI: 10.1227/neu.0000000000002355
  • Free
  • Infographic



confidence interval of the mean
G-force acceleration
National Football League
standard error
Summation of Tests for the Analysis of Risk.

Head injury has been a part of American football since the game's introduction in 1869. To reduce head injury risk, one of the first interventions introduced in 1896 was the development of the leatherhead helmet. Further modifications to the helmet soon followed, with the development of the face mask in 1935 and the patenting of the first plastic helmet by John T. Riddell in 1940. Despite the use of the plastic helmet, 38 football-related fatalities occurred nationwide in 1969.1 During the 1960s, thick foam padding was installed into the helmets, and full-face masks were added. In 1986, the first polycarbonate helmet was introduced, with polycarbonate surrounding a layer of aluminum and vinyl foam on top of plastic and a thin layer of leather.2 Today, the newest Riddell Speedflex Diamond helmet contains a three-dimensional printed impact absorbent lattice liner crafted using Carbon Digital Light Synthesis technology (Riddell).

Improvements in technology in the modern football helmet have been shown to have beneficial effects in reducing concussion risk in high school football players.3 However, when comparing the leatherhead football helmet to the modern football helmet in receiving near-concussive and subconcussive impacts, it was found that the theoretical injury risk based on linear acceleration, angular acceleration, angular velocity, neck force, and neck movement measures was similar or lower for the leatherhead helmet. In both the front and oblique front impact directions, the leatherhead helmet allowed comparable linear acceleration performance to the modern helmet. Furthermore, the differences in near-concussive and subconcussive impacts were not associated with a significant head injury risk reduction even in instances where the modern helmet showed superior angular responses. In these impacts, the stiffness of the modern helmet may be less than optimal in energy absorbency.1 Another study evaluated leatherhead football helmets and the modern football helmet using drop tests on a near-rigid surface. Significantly, higher head accelerations were produced in the leatherhead helmet, and it was concluded that all modern helmets are substantially superior to the leatherhead helmet.4 Therefore, the research into whether the leatherhead helmet has protective elements that can be used to optimize helmet design for all impact doses remains inconclusive.

The development of the hockey helmet followed a progression similar to the football helmet. In the 1930s, the first hockey helmet introduced was made from leather, typically only worn by players recovering from concussions. In the 1960s, molded plastic helmets were introduced, followed by further advances in foam technology, visors, and hard plastics.5 Hockey helmets and football helmets were evaluated using the Hockey Summation of Tests for the Analysis of Risk (STAR) methodology, which found that football helmets provided a lower risk of concussion in comparison with hockey helmets. This can be attributed to the hockey helmets' thinner padding system, resulting in less compression during impact.6 The STAR methodology was used to develop the Virginia Tech Helmet Ratings, where a 5-star rating correlates to the helmets that are best at reducing acceleration of the head during impact. Twenty-six of 28 football helmets tested received a 5-star rating, whereas only 7 of 59 hockey helmets tested received a 5-star rating.7

Each year, the National Football League (NFL) and the NFL Players Association appoint biomechanical engineers to evaluate current football helmet models. An educational poster is created from the results that ranks the helmets based on ability to reduce head impact severity. The helmets are ranked as “top-performing,” “not recommended,” and “newly prohibited” or “prohibited.” Helmets in the “not recommended” category are not permitted to be worn by players unless they had worn them during the previous season. Helmets in the “prohibited” category are strictly not permitted to be worn.8 At the end of the 2021 to 2022 season, 99% of NFL players were wearing helmets in the “top-performing” category. However, 6 helmets in the “top-performing” category in 2021 were downgraded to the “not recommended” category in 2022.9 This demonstrates the constant technological advancements and research being used to continuously improve the design of football helmets.

Our study compared the head impact acceleration force produced in a leatherhead football helmet vs a modern football helmet, as previous studies have found conflicting evidence. The stiffness of the modern football helmet may affect its ability to absorb energy on impact, and this study analyzed the effects of placing softer padding inside the helmet. Previous studies demonstrated that players deliberately collide in a helmet-to-helmet fashion although initiating contact with the helmet is illegal in football and harmful to the brain.10-12 It could be that the modern football helmet provides a false sense of security and protection while there is less perceived protection by the leatherhead helmet. Another explanation is that the lack of a face mask in the leatherhead helmet could raise concern for facial injuries, which may have contributed to fewer intentional helmet-to-helmet strikes in the days of the leatherhead helmet. Despite centuries of technological advancements to the football helmet, we have yet to create a concussion proof helmet.

The purpose of this study was to measure the head impact linear acceleration force when wearing a leatherhead football helmet vs a modern football helmet vs a modified modern football helmet with softer padding vs a modern hockey helmet in helmet-to-helmet strikes. We hypothesize that the acceleration of impact will be the lowest in the modified modern football helmet, followed by the modern football helmet, the modern hockey helmet, and finally, the leatherhead football helmet. In addition, the acceleration produced between 2 modern football helmets will be less than between 2 modern hockey helmets. By providing further insight, we hope to clarify the conflicting findings regarding the protective nature of the leatherhead football helmet and modern hockey helmet in comparison with the modern football helmet. With additional research, we can continue to improve helmet design and minimize concussion risk in contact sports.


Study Design

ADXL326-5V ready triple-axis accelerometers (Analog Devices) were secured onto the forehead (frontal), apex of the head (apex), and right ear (parietal) of a Century Body Opponent Bag manikin (Century). This head and torso is a boxing manikin that mounts onto a weighted base through a hollow plastic tube, which provides body and head flexion to mimic movement during impact. The MacGregor H612 leatherhead football helmet, previously worn in the 1940s to 1950s, was purchased from Ebay.13 The Riddell Head Impact Telemetry System helmet (Riddell) was purchased from a local Riddell product representative for the modern football helmet. This helmet was used previously in testing helmet-to-helmet strikes but was never worn. The modified football helmet was made by replacing the existing inner padding of a used Xenith X1 (Xenith) helmet with polyurethane foam material. A new Bauer RE-AKT 100 hockey helmet (Bauer Hockey) was purchased from Pure Hockey for the modern hockey helmet. This study was an experimental design with no human or animal subjects.

The leatherhead football helmet, modern football helmet, modified football helmet with soft padding, and modern hockey helmet were each placed on the manikin and struck in the left parietal region with the modern football helmet. The striking football helmet contained a 4.42 kg weight for a more realistic striking force and momentum that would be sustained in a typical helmet-to-helmet collision. A hockey helmet to hockey helmet collision was conducted by replacing the striking football helmet with a modern hockey helmet. The striking helmet was hung from a rope at a fixed length (117 cm) and raised 45° from horizontal and then released to strike the helmet placed on the manikin (Figure 1). Each helmet-to-helmet strike was repeated for a total of 20 trials.

An illustration of the testing apparatus with the Riddell Head Impact Telemetry System modern football helmet striking the helmet placed on a Century Body Opponent Bag manikin.

Statistical Analysis

G-force acceleration (GFA) was determined in each helmet-to-helmet strike. Each accelerometer on the apex, frontal, and parietal regions recorded linear acceleration (in g units, 1 g = 9.8 m/s2) experienced by the manikin's head in x, y, and z vectors. The maximum net acceleration on impact was calculated as a net vector from each of the 3 accelerometers' vectors. The G-forces were determined using descriptive statistics (mean, SD, and standard error) for each helmet-to-helmet strike. Ninety-five percent CIs of the mean (95% CIM) were calculated to compare the G-forces sustained by each helmet type across all 5 helmet-to-helmet strike combinations.


Peak GFA values for each of the 3 accelerometers of 100 total helmet-to-helmet impacts were obtained in this study. The results are summarized in Table 1 and Figure 2.

TABLE 1. - Summary of Results
Hockey to hockey Modern football to hockey Modern football to leather football Modern football to modern football Modern football to modified football
 Number (N) 20 20 20 19 21
 Mean GFA 16.12 16.99 28.53 19.24 17.47
 SD 1.67 1.78 5.62 3.08 2.89
 SE 0.37 0.40 1.26 0.71 0.63
 95% CIM 15.39-16.85 16.21-17.78 26.06-30.99 17.85-20.62 16.24-18.71
 Number (N) 20 20 20 19 21
 Mean GFA 24.11 27.34 44.19 38.39 37.76
 SD 2.91 1.25 3.47 7.99 3.70
 SE 0.65 0.28 0.78 1.83 0.81
 95% CIM 22.84-25.39 26.79-27.88 42.67-45.71 34.80-41.99 36.18-39.34
 Number (N) 19 20 20 20 22
 Mean GFA 21.98 23.65 33.64 25.45 29.52
 SD 2.57 1.55 3.86 1.34 6.54
 SE 0.59 0.35 0.86 0.30 1.39
 95% CIM 20.83-23.14 22.97-24.33 31.95-35.34 24.87-26.04 26.79-32.26
CIM, confidence interval of the mean; GFA, G-force acceleration; SE, standard error.
Values are expressed as mean GFA, SD, SE, and 95% CIM at the apex, frontal, and parietal regions. Helmet type is labeled as striking helmet to manikin helmet.

Summary of results. G-force acceleration in gravity g units (1 g = 9.8 m/s2) for each helmet-to-helmet (swinging helmet to manikin helmet) impact at the apex, frontal, and parietal regions. Error bars represent the 95% confidence interval of the mean. CIM, confidence interval of the mean.

When the leatherhead football helmet was placed on the manikin head and struck with a modern football helmet, significantly greater G-forces were produced in all 3 accelerometers (28.53, 44.19, and 33.64 g in the apex, frontal, and parietal regions respectively) in comparison with when a modern hockey helmet (16.99, 27.34, and 23.65 g) and modern football helmet (19.24, 38.39, and 25.45 g) were struck. The manikin's head sustained the highest impact G-forces when fitted with the leatherhead helmet. There was a statistically significant 16% increase in impact G-forces in the modified football helmet, compared with the modern football helmet in the parietal accelerometer. There was no significant difference in the frontal and apex accelerometers. Hockey to hockey helmet strikes and football to hockey helmet strikes provided a 16% to 59% and 8% to 40% reduction, respectively, in impact G-forces on the manikin's head compared with football to football helmet strikes. The reduction was statistically significant in all 3 accelerometers.


Key Results

In this study, we determined the head acceleration produced when performing helmet-to-helmet strikes with football and hockey helmets. The results demonstrated that the manikin's head sustained the highest impact G-forces when wearing the leatherhead football helmet. The pendulum model used in our study provided subconcussive impact doses, similar to the model used by Bartsch et al.1 However, Bartsch et al1 found leatherhead helmets to be comparable or more protective to the modern helmet and suggested that modern helmet stiffness provided less than optimal energy absorbency in lower severity impacts. A subsequent study conducted by Rowson et al4 implemented a drop test method for the leatherhead helmet producing acceleration values within concussion threshold levels. Although the drop test methodology may not generate the linear and rotational head motions observed on field, they concluded that all modern helmets were superior to the leatherhead helmet in reducing head acceleration.4,14 Rowson et al contributed their contrasting findings to Bartsch et al to the low impact severity used in Bartsch et al.’s study that was not able to fully compress the more compliant padding of the striking helmet.1,4 Our study demonstrates that even at low helmet-to-helmet impact doses, the modern helmet is superior in reducing impact G-forces when compared with the leatherhead helmet. An advantage of our study design is the attachment of the accelerometers directly to the manikin's head, allowing for a direct measurement of the impact G-forces sustained by the head. Other studies have been conducted by measuring the impact sustained by the helmet or the sensors within the helmet.6,15,16

The modified football helmet that contained softer padding on the inside did not demonstrate lower impact G-forces compared with the modern football helmet, indicating that softer foam may not be superior in reducing impact potential. Mills et al investigated peak head acceleration and padding deformation when adjusting the densification of the foam layer of a football helmet. Their results showed that the difference in acceleration with material changes was not significant; rather, impact speed had the greatest effect. In addition, the location of padding relative to site of impact played a role in padding compression. Impacts that were in line with the foam pad resulted in uniaxial compression, whereas impacts in line with the corner of a pad resulted in translational movement away from the site of impact.17 It is possible that the low impact dose and location of impact contributed to the increased G-forces in the parietal region of the modified football helmet observed in our study.

The hockey helmet demonstrated lower impact G-forces, suggesting that there is an element of the hockey helmet that is providing superior protection over the football helmet. These results are not consistent with the STAR methodology, which rated football helmets as providing a lower concussion risk when compared with hockey helmets. However, the STAR methodology determined concussion risk by assessing head acceleration at near-concussive, subconcussive, and concussive thresholds.7 Although the thinner construction of the hockey helmet is not optimal at reducing head acceleration when concussive thresholds are considered, this design may provide an element of protection over football helmets at lower severity impacts.


Testing helmets at lower impact doses is important for youth football players, who do not endure the higher momentum doses seen in adult professional football. It was found that the 95th percentile head acceleration from impact in youth players aged 10 to 12 years is 48.8 g, whereas the average head acceleration of a concussive impact in adult players is 105 g.16,18,19 In addition, low-energy impacts account for 80% of head impacts in youth football players and greater than 90% of head impacts in hockey.7,18 Because youth helmets are essentially scaled down versions of adult helmets, they are optimized to absorb high impact levels that are rarely produced by youth players, rather than lower-dose impacts that are more commonly occuring.1 It is possible that the optimal football helmet in youth would be one that is similar in construction to the hockey helmet, which is more effective at reducing head acceleration at lower-energy impacts. Although the lowest concussive impact reported in literature is 42 g, there is evidence that the impact concussion threshold is individual specific, with some players being more susceptible to concussion regardless of acceleration magnitude.16,20 Repetitive impact exposure is thought to affect concussion susceptibility and tolerance, highlighting the importance of reducing head acceleration at all impact doses.20 Additional studies are required to develop a helmet that provides optimal protection across all levels of impact for every age range.


A limitation of the study was that only low impact G-forces were studied, as the pendulum model could not produce impacts at concussion threshold levels. The leatherhead helmet was aged with stiff leather and may not have been an accurate representation. When leatherhead helmets were in use, player head-to-head strikes were leatherhead helmet to leatherhead helmet, which our study was not capable of simulating. In addition, few angles of impact were investigated because the data were obtained solely from strikes to the left parietal region. The manikin used was representative of a human subject in size and form; however, the material composition does not exactly match that of a human body.

Future studies include testing helmets from various other sports. Helmets that are more expensive and marketed to be more protective can be compared with those that are less expensive. It is interesting that the heavy hand impacts sustained in celebration may result in additional head trauma, and these forces can be measured.


The aged leatherhead football helmet was the least protective, and the hockey helmets were the most protective, with the football helmet being intermediate in G-forces sustained by the manikin's head. This study provides additional insight into the inconclusive evidence about the protective qualities of the leatherhead football helmet. These results will be useful in the design of football and hockey helmets to reduce concussion risk in the future.


This research was supported by a grant from the Hawai'i Pediatric Association Research and Education Foundation (HPAREF).


The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.


1. Bartsch A, Benzel E, Miele V, Prakash V. Impact test comparisons of 20th and 21st century American football helmets. J Neurosurg. 2012;116(1):222-233.
2. Stamp J. Leatherhead to Radio-Head: The Evolution of the Football Helmet. Smithsonian Magazine; 2012.
3. Collins M, Lovell MR, Iverson GL, et al. Examining concussion rates and return to play in high school football players wearing newer helmet technology: a three-year prospective cohort study. Neurosurgery. 2006;58(2):275-286.
4. Rowson S, Daniel RW, Duma SM. Biomechanical performance of leather and modern football helmets. J Neurosurg. 2013;119(3):805-809.
5. Bak R. A Short, Heads-Up History of the Hockey Helmet. Vintage Detroit Collection. 2011. Accessed September 13, 2022.
6. Rowson B, Rowson S, Duma SM. Hockey STAR: a methodology for assessing the biomechanical performance of hockey helmets. Ann Biomed Eng. 2015;43(10):2429-2443.
7. Virginia Tech Helmet Ratings. Accessed September 16, 2022.
8. National Football League. NFL and NFLPA Release 2022 Helmet Testing Performance Results. NFL Football Operations. Accessed September 13, 2022.
9. National Football League Associated Press. VICIS has Top Four Helmets in Impact Testing by NFL, NFLPA. National Football League. Accessed September 13, 2022.
10. Solomon J. College Football Safety Gets Shaped by Research Including Possible Head-Hit Counts. Alabama Media Group. Accessed March 10, 2022.
11. National Football League Operations. Use of Helmet as a Weapon - Rule 12, Section 2, Article 17. Accessed September 13, 2022.
12. National Football League Player Health & Safety. NFL Health and Safety Related Rules Changes since 2002. National Football League. Accessed September 13, 2022.
13. National Museum of American History. MacGregor Football Helmet Model H612. Smithsonian. Accessed September 14, 2022.
14. Bartsch A, Benzel E, Miele V, Prakash V. Editorial: leather football helmets. J Neurosurg. 2013;119(3):803-804.
15. Nakatsuka AS, Yamamoto LG. External foam layers to football helmets reduce head impact severity. Hawaii J Med Public Health. 2014;73(8):256-261.
16. Rowson S, Duma SM. Development of the STAR evaluation system for football helmets: integrating player head impact exposure and risk of concussion. Ann Biomed Eng. 2011;39(8):2130-2140.
17. Mills ST, Young TS, Chatham LS, Poddar S, Carpenter RD, Yakacki CM. Effect of foam densification and impact velocity on the performance of a football helmet using computational modeling. Comput Methods Biomech Biomed Eng. 2021;24(1):21-32.
18. Campolettano ET, Gellner RA, Sproule DW, et al. Quantifying youth football helmet performance: assessing linear and rotational head acceleration. Ann Biomed Eng. 2020;48(6):1640-1650.
19. Pellman EJ, Viano DC, Tucker AM, Casson IR, Waeckerle JF. Concussion in professional football: reconstruction of game impacts and injuries. Neurosurgery. 2003;53(4):799-814.
20. Rowson S, Duma SM, Stemper BD, et al. Correlation of concussion symptom profile with head impact biomechanics: a case for individual-specific injury tolerance. J Neurotrauma. 2018;35(4):681-690.

Concussion; Football; Helmet; Hockey; Leatherhead; Sport helmets

© Congress of Neurological Surgeons 2023. All rights reserved.