Article Text

The impact performance of headguards for combat sports
  1. Andrew S McIntosh 1 , 2,
  2. Declan A Patton 1
  1. 1 ACRISP, Federation University Australia, Ballarat, Victoria, Australia
  2. 2 McIntosh Consultancy and Research, Sydney, Australia
  1. Correspondence to Dr Andrew S McIntosh, ACRISP, Federation University Australia, P.O. Box 663, Ballarat, VIC 3353, Australia; as.mcintosh{at}


Background/aim To assess the impact energy attenuation performance of a range of headguards for combat sports.

Methods Seven headguards worn during combat sport training or competition, including two Association Internationale de Boxe Amateur (AIBA)-approved boxing models, were tested using drop tests. An International Organization for Standardization (ISO) rigid headform was used with a 5.6 kg drop assembly mass. Tests were conducted against a flat rigid anvil both with and without a boxing glove section. The centre forehead and lateral headguard areas were tested. Peak headform acceleration was measured. Tests from a selection of drop heights and repeated tests on the same headguard were conducted.

Results Headguard performance varied by test condition. For the 0.4 m rigid anvil tests, the best model headguard was the thickest producing an average peak headform acceleration over 5 tests of 48 g compared with 456 g for the worst model. The mean peak acceleration for the 0.4, 0.5 and 0.6 frontal and lateral rigid anvil impact tests was between 32% and 40% lower for the Top Ten boxing model compared with the Adidas boxing model. The headguard performance deterioration observed with repeat impact against the flat anvil was reduced for impacts against the glove section. The overall reduction in acceleration for the combination of glove and headguard in comparison to the headguard condition was in the range of 72–93% for 0.6 and 0.8 m drop tests.

Conclusions The impact tests show the benefits of performance testing in identifying differences between headguard models.

  • Head
View Full Text

Statistics from


Boxing and martial arts, referred to collectively in this paper as ‘combat sports’, are sports in which the primary objective of each contestant is ‘to strike, kick, hit, grapple with, throw or punch one or more other contestants’.1 Combat sports are contested at many levels and in a variety of formats, including the Olympic Games. As a result of head and other injury risks in competition and training, headguards and other equipment are often worn. Head injury risks in combat sports include head wounds and concussion. There are various levels of control over equipment used in combat sports, which range from none to government regulations.2–4

The proportion of head injuries reported in boxing varies from 10% to approximately 50% of all injuries depending on study design.5–10 In Taekwondo competition, Pieter et al 11 reported that approximately 30% of all injuries in men and 15% in women were to the head and neck region.

Many combat sports have equipment regulations that govern headguards, although few combat sports have performance specifications or refer to standards.3 , 4 Typically, a combat sport will define dimensional criteria: the area of head coverage, the maximum thickness of the headguard and the maximum mass.3 Unless this approach is supported by performance-based assessment of headguards or formally conducted epidemiological studies, the protective benefits of headguards meeting those dimensional criteria are unknown.12–14 There are few performance standards for headguards intended for combat sports.15

Helmets are commonly assessed on their ability to reduce the impact force (ie, attenuate impact energy) by dropping an instrumented headform fitted with the helmet from a specified height onto a specified anvil.16 , 17 Drop tests are repeatable and relatively inexpensive. Drop tests have limitations in some areas, for example, the headform's angular kinematics are not measured and/or are constrained by the test method.16 However, drop tests are able to differentiate between the level of impact performance provided by helmets and form the basis for many successful systems of helmet supply.16 One characteristic unique to many sports helmets is the requirement for the headguard to function over repeat impacts during a contest, and over many contests and/or training sessions.

Few studies have assessed the protective potential of boxing headguards.9 , 18–22 All studies have reported that headguards reduce the magnitude of the head's acceleration in the range 15–33%, depending on the impact test method. The objectives of the tests reported in this study are to assess the performance of a range of headguards for combat sports with reference to: impact energy attenuation during a single impact; impact energy attenuation over multiple impacts; and, influence of a boxing glove on impact energy attenuation.


Headguard selection

A selection of headguards worn during training or combat sport competition was obtained. The parent study was focussed on boxing headguards; therefore, available Association Internationale de Boxe Amateur (AIBA)-approved headguards were included. The initial selection criterion for headguard models was compliance with a standard or sports-specific regulation. In addition, models that claimed some impact performance characteristics, for example, ‘designed’ or ‘tested’ to a technical specification, were considered. Design features that might potentially reveal desirable impact performance functionality are the density of the main material in the headguard, and the thickness of the main body of the headguard. These characteristics were also considered in the selection of the headguards. Finally, a cross-section of models from different combat sports was sought. Seven headguard models were selected for testing, including two AIBA-approved headguards (figure 1).

Figure 1

Headguard and glove models and their properties (ASTM, American Society for Testing and Materials).

Impact test rig

Impact tests were conducted at the Roads and Maritime Services (RMS) Crashlab in Sydney using an International Organization for Standardization (ISO) ‘M’ rigid headform and the AS/NZS 2512.3.1:2007 test method.23 The headform circumference was 600 mm and drop assembly mass 5.6 kg. In each test, the headform and drop assembly were raised to a specified height and then dropped onto a flat anvil and the headform acceleration measured. The peak headform acceleration and the head injury criterion with a 15 ms duration limit (HIC15) are reported.

Glove sections

Tests were also conducted using the striking face of the glove. An AIBA-approved 10 oz glove was disassembled and the inner moulded foam insert was cut to sit on the anvil and to provide an impact interface. The heights of glove samples I and II were 57 mm, the maximum thickness of the glove, but the cross-sections differed slightly. All headguards were tested with sample I, except for the two Macho models which were tested 2 months after the other headguards with glove sample II. Repeat impacts to the same headguard and at the same location were also conducted. The five repeat tests were completed within 90 s. Headguards were tested under ambient conditions.

Impact severity

There are few studies that have identified the energy transferred to the head in a typical boxing punch. Viano et al 24 reported that the velocity change of the head as a result of a punch delivered by 11 Olympic boxers was much lower than the velocity of the fist. The hand speed was reported to be between 7 and 11 m/s, whereas the change in velocity of the anthropomorphic test device head was approximately 3 m/s. On the basis of the head's change in velocity and an assumption that head mass was 4.45 kg, the energy change of the head was approximately 20 J. In the context of the ISO ‘M’ headform, this is equivalent to a drop of 0.36 m. Initial tests were conducted at a drop height of 0.5 m to assess the potential of the headguards. The drop height was raised to 0.8 to assess the maximum impact that might be included in the test matrix, and then 0.2 m to assess a meaningful least severe impact. Impacts were then focused on the 0.4, 0.5 and 0.6 m drop heights.

Impact location

The headguards, when examined, typically had two areas of substantial padding in common: (1) frontal (the centre forehead) and (2) laterally. All impacts were directed radially towards the head's centre of mass. Unless otherwise reported, headguard sites were only impacted once.

Data analyses

Descriptive results are presented. The percentage change in peak headform acceleration is presented in the combined glove and headgear tests.


In total, 213 impact tests were conducted. The test results for peak headform acceleration are presented in tables 1 and 2. Results for HIC15 are presented in online supplementary material. For frontal and lateral drop tests onto the rigid flat anvil, the Macho Warrior was the best performing model, and the Adidas Taekwondo was the worst performing model (tables 1 and 2). A drop height of 0.4 m equates to an impact energy of 22.0 J, 0.6 m to 33.0 J and 0.8 m to 43.9 J. The performance of the Top Ten AIBA boxing headguard was superior to the Adidas AIBA boxing model (tables 1 and 2). All headguard models demonstrated that performance deteriorated with repeat flat rigid anvil impacts (tables 1 and 2). This was especially the case when the headguard model was close to its performance limit. These observations reflect the viscoelastic properties of the headguards.

Table 1

Results for all frontal drop tests onto the flat rigid anvil

Table 2

Results for all lateral drop tests onto the flat rigid anvil

The performance of the two glove samples (I and II) on their own in tests without headguards was generally superior to those with only headguards. Three tests per sample at three drop heights were conducted. In glove sample I, the mean peak headform acceleration in a 0.4 m drop was 28.5 g (SD=2.0 g), at 0.6 m the mean was 57.3 g (SD=2.2 g), and at 0.8 m the mean was 98.4 g (SD=3.2 g). In glove sample II, the mean peak headform acceleration in a 0.4 m drop was 55.0 g (SD=6.7 g), at 0.6 m the mean was 123.9 g (SD=6.2 g), and at 0.8 m the mean was 236.6 g (SD=6.3 g). As evidenced by the SDs, there was little change in the impact energy attenuation on repeat impacts.

Combinations of gloves and headguards were tested at 0.6 and 0.8 m drop heights in lateral impacts (table 3). For example: in the 0.6 m drop tests, the mean peak headform acceleration for the Top Ten (AIBA) boxing headguard reduced from 382 g in the rigid impact to 33 g in the glove impact, and, in the 0.8 m drop tests, the peak headform acceleration for the Top Ten (AIBA) boxing headguard reduced from 665 g in the rigid impact to 47 g in the glove impact. The overall reduction in acceleration for the combination of glove and headguard in comparison with headguard only was in the range of 72–91% for 0.6 m drop tests. The overall reduction in acceleration for the combination in comparison with glove only was in the range of 42–75% for 0.6 m drop tests. The overall reduction in acceleration for the combination of glove and headguard in comparison with headguard only was in the range of 81–93% for 0.8 m drop tests. The overall reduction in acceleration for the combination in comparison with glove only was in the range of 53–80% for 0.8 m drop tests. Similar trends were observed with HIC15.

Table 3

Comparative reduction in peak headform acceleration is reported with respect to (‘wrt’) both headguard only and glove only tests


The impact performance of a range of headguards was assessed using repeatable and routine drop test methods. The tests were not intended to replicate directly combat sport impacts. The tests revealed a broad range of impact performances by headguard models, and the benefits of the combination of a glove and headguard in maximising the impact energy attenuation. The results are not directly comparable with other studies of combat sport headguards because each study has applied different methods, for example, headforms, energy and impact interface.22 , 25

Boxing headguards

With regards to boxing, the results showed that the Top Ten headguard's impact performance was superior to the Adidas boxing model. AIBA rules do not specify any performance requirements, and neither model claimed compliance with any other relevant helmet standard, for example, ASTM F2397—09 (American Society for Testing and Materials).16 The mean peak acceleration for the 0.4, 0.5 and 0.6 m frontal and lateral rigid anvil impact tests was between 32% and 40% lower for the Top Ten model compared with the Adidas boxing model. The impact tests show the benefits of performance testing in identifying differences between headguard models. The performance differences are also sufficiently large to suggest that it is necessary to control for headguard model in epidemiological studies of boxing headguards. It is challenging to relate the results of these impact tests directly to the likelihood of concussion in boxing because angular head acceleration was not assessed, the headform was rigid, and head impacts in boxing are delivered through a punch. These issues are addressed in a companion paper by McIntosh AS and Patton DA.26

Comparisons of headguards

The results demonstrate the opportunities to improve headguard performance through material selection and design.27 Setting performance specifications for headguards in boxing and other combat sports, and resolving usability issues around mass and thickness would encourage the supply of better headguards.12 , 15 , 16 , 22 , 25 The best performing headguards were either the heaviest—the Rival RHG 10 at 0.53 kg (average thickness 25 mm, density 140 kg/m3)—or the thickest—the Macho Warrior at 37 mm (mass 0.3 kg, density 130 kg/m3). The worst performing headguard was the Adidas Taekwondo model, which was the lightest and thinnest headguard. The two Macho brand headguard models had similar foam densities (125 kg/m3), but the Warrior's average thickness was 37 mm compared with the Dyna's average thickness of 25 mm. The additional thickness explained the Warrior's superior performance. Comparatively, the Macho Warrior was between seven and eight times more effective in reducing headform acceleration compared with the Adidias Taekwondo model in rigid impacts, but with only a difference in mass of 0.09 kg. The opportunities available to designers are to (1) maintain the thickness of the headguard and increase its density, (2) increase the thickness and maintain density or (3) do both.


The combination of headguard and glove was superior to either on its own and reduced the deterioration in headguard performance with repeat impact loading, which has been observed in padded headgear and headguards.17 , 21 The most likely reason for this change is a reduction in the headguard deformation with the glove. Therefore, the hysteresis of the polymeric foam in each headguard changed as a result of there being less deformation.


The authors contend that the range of results observed in this study reflects the absence of performance standards that specify impact tests and performance criteria for headguards in combat sports. Without mandated requirements, suppliers can produce a range of products with limited effectiveness and potentially batch-by-batch variations. If drop tests were considered to be appropriate for assessing headguards for combat sports, further work is required to identify the test requirements and performance criteria. The results highlight the opportunity available through careful selection of material thickness and density to improve impact performance.

What are the new findings?

  • Impact testing highlights the performance differences, from poor to good, between headguards intended for combat sports, including boxing, and their limitations.

  • Improvements in the impact performance and likely head injury risk mitigation are achieved through the thickness and density of the foam materials from which the headguards are constructed.

  • In the absence of performance-based standards, including impact testing and performance criteria, the full injury mitigating benefits of headguards are unlikely to be provided to combat sport competitors.

  • Epidemiological studies of injury and safety interventions in combat sports may be confounded by the large variation in performance of headguard models worn by participants. It is recommended that headguard model is controlled as a confounding variable.


The authors would like to thank: the IOC for funding the project, in particular Dr Torbjorn Soligard, who has been the point of contact, Cherine Fahmy and Professor Lars Engebretsen; Roads and Maritime Services Crashlab staff, in particular Devaraj Muniswamy, and David Felice from Zoe's martial arts and boxing supplies in Sydney. Their assistance is gratefully acknowledged.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.