Sports Biomechanics the Mechanics of American Football Cleats On

This article was downloaded by: [Mr Richard Kent] On: 26 June 2015, At: 09:17 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Sports Biomechanics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/rspb20 The mechanics of American football cleats on natural grass and infill-type artificial playing surfaces with loads relevant to elite athletes Richard Kenta, Jason L. Formana, David Lessleya & Jeff Crandalla a University of Virginia Center for Applied Biomechanics, Charlottesville, VA, USA Published online: 26 Jun 2015.

Click for updates

To cite this article: Richard Kent, Jason L. Forman, David Lessley & Jeff Crandall (2015): The mechanics of American football cleats on natural grass and infill-type artificial playing surfaces with loads relevant to elite athletes, Sports Biomechanics, DOI: 10.1080/14763141.2015.1052749 To link to this article: http://dx.doi.org/10.1080/14763141.2015.1052749

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 Sports Biomechanics, 2015 http://dx.doi.org/10.1080/14763141.2015.1052749

The mechanics of American football cleats on natural grass and inﬁll-type artiﬁcial playing surfaces with loads relevant to elite athletes

RICHARD KENT, JASON L. FORMAN, DAVID LESSLEY, & JEFF CRANDALL

University of Virginia Center for Applied Biomechanics, Charlottesville, VA, USA

(Received 12 December 2014; accepted 5 May 2015)

Abstract This study quantified the mechanical interactions of 19 American football cleats with a natural grass and an infill-type artificial surface under loading conditions designed to represent play-relevant manoeuvres of elite athletes. Variation in peak forces and torques was observed across cleats when tested on natural grass (2.8–4.2 kN in translation, 120–174 Nm in rotation). A significant (p , 0.05) relationship was found between the peak force and torque on natural grass. Almost all of the cleats caused shear failure of the natural surface, which generated a divot following a test. This is a force- limiting cleat release mode. In contrast, all but one of the cleat types held fast in the artificial turf, resulting in force and torque limited by the prescribed input load from the test device (nom. 4.8 kN and 200 Nm). Only one cleat pattern, consisting of small deformable nubs, released on the artificial surface and generated force (3.9 kN) comparable to the range observed with natural grass. These findings (1) should inform the design of cleats intended for use on natural and artificial surfaces and (2) suggest a mechanical explanation for a higher lower-limb injury rate in elite athletes playing on artificial surfaces.

Keywords: Athletes, biomechanics, sports equipment, injury, performance Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 Introduction The interaction between a player’s foot and the playing surface is of critical importance in sports for both performance and injury risk. A player’s interaction with the surface dictates how fast he/she can accelerate, stop, and change direction. This same interaction also may be a factor in the risk of injury to the lower extremities. Among lower extremity injuries in American football and soccer, 21–61% are ‘non-contact’ type injuries—injuries that do not result from direct loading of the affected limb by another player or object (Agel, Arendt, & Bershadsky, 2005; Bradley, Klimkiewicz, Rytel, & Powell, 2002; McHugh, Tyler, Mirabella, Mullaney, & Nicholas, 2007; Ramirez, Schaffer, Shen, Kashani, & Kraus, 2006; Woods, Hawkins, Hulse, & Hodson, 2003). It has been postulated that these types of injuries may be caused by foot ‘entrapment’ that can cause injurious stresses within the lower extremities

Correspondence: Richard Kent, University of Virginia Center for Applied Biomechanics, 4040 Lewis & Clark Dr., Charlottesville, VA, USA, E-mail: [email protected]

q 2015 Taylor & Francis 2 R. Kent et al.

(Lambson, Barnhill, & Higgins, 1996; Orchard, Seward, McGivern, & Hood, 2001; Torg, Quedenfeld, & Landau, 1974). Particular attention has been paid to the role of pivoting (rotational) foot entrapment in ACL injury (Lambson et al., 1996; Torg et al., 1974), although entrapment during other loading motions (e.g. lateral movement causing inversion of the ankle (Bloemers & Bakker, 2006) may also cause injury. Meyers and Barnhill (2004) evaluated American football players from 8 high schools over five competitive seasons playing on natural grass or on one particular brand of infill artificial surface. They identified significant playing surface effects by injury type, location, mechanism, and time lost and recommended further investigation to elucidate these observations. Hershman et al. (2012) analysed the incidence of lower extremity injuries on natural grass and on one brand of infill artificial surface in 5,360 team games in the National Football League. They found that the rates of knee sprains and of ankle sprains on the infill surface were 22% higher than on grass. When the injuries were further subdivided, the rate of sprains to the anterior cruciate ligament was found to be 67% higher on that infill surface than on natural grass, and the rate of eversion ankle sprains was found to be 31% higher. In contrast, the rates of sprains to the medial collateral ligament and of inversion ankle sprains were not significantly higher on that infill surface than on natural grass. Understanding the relationship between cleat design and an athlete’s interaction with a playing surface remains one of the most challenging topics in sports biomechanics. Several factors in cleat design—including the size, number, and distribution of cleats—are highly variable. Developing models to predict the interaction mechanics is challenging, as both the cleats and the playing surfaces are typically composed of composite materials that experience large deformations and exhibit complex structural and material behaviours. As a result, the study of cleat designs has relied on mechanical testing, either with rotational or translational types of tests (Andreasson, Lindenberger, Renstrom, & Peterson, 1986; Bonstingl, Morehouse, & Niebel, 1975; Bowers & Martin, 1975; Dura, Hoyos, Martinez, & Lozano, 1999; Heidt et al., 1996; Livesay, Reda, & Nauman, 2006; Torg, Stilwell, & Rogers, 1996; Torg et al., 1974; Villwock, Meyer, Powell, Fouty, & Haut, 2009). The interpretation of the findings of these earlier studies with respect to elite American football players is unclear because the methods they employed did not generate the load magnitudes, rates, or durations typically experienced by those athletes in game-play scenarios (Kent et al., 2012). These past studies are also limited because they did not explicitly evaluate the release mechanics between the cleat and the surface.

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 Limited other studies have investigated the effect of various cleat patterns on ground reaction forces during manoeuvres performed by volunteers (Queen et al., 2008). While they did try to simulate some game-relevant manoeuvres such as running and cutting, those studies did not result in conditions that tested the limiting performance of the cleats. Instead, all of the cleat patterns studied maintained grip with the playing surface, and as a result all generated similar ground reaction forces. This information is of limited utility since the functional differences in cleat patterns will only manifest at their performance limits, where one cleat pattern will break traction with the ground and another will not. Kent et al. (2012) described a device and method for studying the interactions between cleats and playing surfaces that overcomes some of the limitations described above. The device was termed the BioCORE Elite Athlete Shoe-Surface Tester (BEAST). It can test shoe-surface interactions in situ at loads and speeds representing elite athletes at the limits of cleat performance through three tests that reﬂect generic classes of tasks: (1) translation test; (2) rotation test; and (3) translation/drop test. As a ﬁrst step towards understanding the effect of cleat design, the goal of this study was to perform a descriptive analysis of the range of forces and torques generated by various contemporary American football cleat patterns on The mechanics of American football cleats 3

natural grass and in-fill type artificial surfaces using the test methodologies of Kent et al. (2012) and to evaluate the release mechanics of different cleats on these surfaces. Several specific hypotheses were evaluated. First, that there is a positive association between the horizontal forces generated under pre-loaded translation tests and the torques generated under pre-loaded rotation tests for various cleat patterns. Second, that there is a positive association between the horizontal forces generated under pre-loaded translation tests and the horizontal forces generated under translation/drop tests for various cleat patterns. Finally, that the release mechanisms are different between natural and artificial surfaces, that they limit the loads generated in the test, and that they depend on cleat pattern. A further goal of the study was to quantify the ranges of forces and torques generated by various cleats in these loading modes to serve as a baseline, reference dataset for comparison with the performance of future cleat designs.

Methods Test device The mechanical operation of the BEAST is described in detail by Kent et al. (2012), as are the three types of tests that it performs (namely a translation test in which a preloaded cleat is pulled along the surface, a rotation test in which a preloaded cleat is twisted relative to the surface, and a translation/drop test in which a cleat strikes the ground with both vertical and horizontal velocity components). A 2.8-kN vertical preload was used in the translation and rotation tests reported here. The maximum potential pulling force in the translation tests is regulated by the pressure introduced into the actuator and was approximately 4.8 kN. The maximum potential applied torque in the rotation tests, likewise regulated by the pressure introduced into the actuator, was approximately 200 Nm. In both tests, the loads measured at the foot-form can be substantially below these values if some load-limiting mechanical interaction occurs at the cleat, surface, or cleat-surface interface (e.g. the surface fails in shear at a lower load or the cleat releases and moves relative to the surface). In conducting the translation/drop tests, a mass of 42 kg, a horizontal speed of 1.5 m/s, and a drop height of 67 mm were used. The BEAST employs a foot-form to which a shoe or portion of a shoe is mounted. In this study, the foot-form was oriented such that the applied horizontal force in the translation

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 tests was parallel to the long axis of the foot and directed toward the posterior aspect. Thus, this test represents in an idealised sense the force applied to the foot when an athlete pushes off to start running forward. In the translation/drop tests, the foot-form was oriented such that the applied horizontal force vector was approximately perpendicular to the long axis of the foot and directed toward the medial aspect. Thus, this test represents in an idealised sense the lateral force applied to the right foot when an athlete cuts left and the applied force that would induce inversion and internal rotation in the ankle joint. The BEAST is instrumented to measure the three forces and three moments experienced by the foot-form during interaction with the ground, the horizontal and vertical displacements and the rotation of the foot-form, and the tri-axial acceleration of the foot-form. The inputs to the BEAST (vertical pre-load and pressure in the translation tests; drop height, dropped mass, and pressure in the translation/drop tests) were speciﬁed to represent an overload situation relative to the performance of elite American football players. Figure 1 illustrates how the BEAST is programmed to do this. That ﬁgure shows the average peak ground reaction forces generated by nine elite American football players performing a range of tasks (Riley et al., 2013) compared with the peak horizontal and vertical forces 4 R. Kent et al.

Figure 1. Peak horizontal and vertical ground reaction force generated by elite American football players in game- relevant tasks (open circles, Riley et al., 2013) compared with loads generated by the BEAST during translation (small x) and translation/drop (small filled circles) tests using several cleat patterns on infill artificial turf.

generated in translation and translation/drop tests using a range of cleat patterns on inﬁll artiﬁcial turf. As shown, the loads generated by the BEAST exceed those generated by the athletes during normal performance-related tasks and, as such, represent a situation where release of the cleat pattern would be desired.

Surfaces Tests were performed on two playing surfaces located at a single facility. These surfaces serve as practice fields for a professional American football team, and are maintained by a professional, full-time maintenance staff. One surface was natural grass (Kentucky Blue Grass) on an outdoor field. One surface was artificial, in-fill type turf installed on an outdoor field. Each surface was less than one year old at the beginning of this test program. For each repeated test the BEASTwas moved a distance of at least 0.7 m to a new location on the field. Care was taken to avoid seams and painted areas on the fields. The ambient air temperature, ambient air humidity, and ground temperature were recorded prior to each test. The

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 ambient air temperature ranged from 13–368C; the ambient air humidity ranged from 22 to 75%; the ground temperature ranged from 14–378C. All surfaces were tested in their as- found (i.e. dry) states.

Cleated foot-form To isolate the mechanical interactions between the test surfaces and the cleat patterns, this study used footforms consisting of production shoe bottoms rigidly attached to the test device (e.g. Kent et al., 2012; Livesay et al., 2006). The shoe bottoms were taken from size 12 shoes (right foot) designed for American football players. To simulate a player pushing, pivoting, or landing with a raised hindfoot (American Society for Testing and Materials, 2006; Kent et al., 2012; Queen et al., 2008), only the forefoot section of the cleats were used. To prepare a test sample, the forefoot portion of the shoe bottom was first cut from a new shoe. Automobile body filler then was applied to the upper surface of the shoe bottom to create a stiff, reinforced, flat surface that could be bolted to the flat plate foot-form of the test device. The mechanics of American football cleats 5

Instrumentation The forces and moments generated by the interaction of the cleated foot-form and the playing surface were measured by a load cell (model 1914, R.A. Denton, Rochester Hills, MI, USA) mounted between the foot-form and the shaft of the BEAST (Kent et al., 2012). The vertical force, the horizontal force, and the torque about the vertical axis (in the rotation tests) are reported here. A linear displacement transducer (TLM series, Novotechnik, Southborough, MA, USA) measured the horizontal motion of the foot-form relative to the playing surface. A string potentiometer (161 series, Firstmark Controls, Creedmoor, NC, USA) measured the vertical motion in the translation/drop tests, and a rotary potentiometer (model SP22GS, ETI Systems, Carlsbad, CA, USA) measured the rotation of the foot-form in the rotation tests. In the combined translation/drop test condition the peak vertical force reflects the vertical stiffness of the playing surface. The peak horizontal force reflects both the horizontal stiffness of the playing surface and the interaction mechanics between the cleats and the surface under the vertical load. All signals were recorded at a rate of 10 kHz with a National Instruments (Austin, TX, USA) Compact DAQ data acquisition system and digitally filtered using the Channel Frequency Class (CFC) filtering procedure described in the Society of Automotive Engineers Recommended Practice for Instrumentation for Impact Tests (Society of Automotive Engineers, 1995). All force and torque signals were filtered using CFC 600. All displacement and rotation signals were filtered using CFC 180. Motions, forces, and torques were examined during a time window of 200 ms for the translation and rotation tests, and 100 ms for the combined translation/drop tests.

Cleat patterns and analysis methods Nineteen different cleat models from various manufacturers were studied. Each cleat pattern was subjected to the battery of tests described by Kent et al. (2012) on both the artiﬁcial and natural grass playing surfaces, with repeated tests performed in selected instances. For each surface (natural and artiﬁcial), a linear regression analysis was performed to test for a relationship between the horizontal forces measured during the pre-loaded translation tests and the torques about the vertical axis measured in the pre-loaded rotation tests. A linear regression was also performed to test for a relationship between the horizontal forces measured during the pre-loaded translation tests and the lateral horizontal forces

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 measured during the translation/drop tests for each surface. For cases where repeated tests were performed, the average values of the results for each cleat were used in this analysis. Significance was defined as p , 0.05 for both the slope of the regressions and for comparisons between groups. Of the nineteen cleat patterns, five were selected for further illustrative comparison. These were selected to represent the extreme data points resulting from the tests with natural grass. The designs of these five cleat patterns were qualitatively compared by cleat depth, style, and number.

Results Cross-plots of peak horizontal force in the translation tests versus peak torque in the rotation tests and versus horizontal force in the translation/drop tests are shown in Figures 2 and 3. The regression indicated a positive and signiﬁcant relationship between the horizontal force in the translation tests and the torque in the rotation tests for the tests performed on the natural grass surface. The regression also indicated a positive and signiﬁcant 6 R. Kent et al.

Figure 2. Cross-plots of the peak posterior (horizontal) reaction forces in the preloaded translation tests versus the peak moments (torques) about the vertical axis in the preloaded rotation tests. The datapoints are the mean values for each cleat. The error bars indicate ^ one standard deviation for cleats where multiple tests were performed. Note that for the artiﬁcial turf the forces and torques were limited by the capabilities of the test device for most cleats. The circled points are cleats selected for illustrative comparison (with the cleat identiﬁer numbers shown).

relationship between the peak horizontal force in the translation tests and the peak torque in the rotation tests for the tests performed on the artificial grass surface, however this was predominantly due to a clustering of most of the datapoints, with a single datapoint outside the cluster unduly influencing the regression. As such, those regression results are excluded from further analysis and discussion. No significant association was observed between the peak horizontal force in the translation tests and the peak horizontal force in the translation/ drop tests for either the natural grass or the artificial surfaces. In the translation and rotation tests on natural grass, all of the tests resulted in ‘breakaway’ where the cleats were eventually able to move relative to earth. With some of the less- aggressive cleat patterns, this ‘breakaway’ occurred as a result of the foot-form sliding over the surface of the grass (‘sliding’). Other cleat patterns resulted in a greater engagement with the grass, causing the engaged portion of the grass to tear away from the surface (‘divoting’). Downloaded by [Mr Richard Kent] at 09:17 26 June 2015

Figure 3. Cross-plots of the peak posterior (horizontal) reaction forces in the preloaded translation tests versus the peak lateral (horizontal) reaction forces in the translation/drop tests. The datapoints are the mean values for each cleat. The error bars indicate ^ one standard deviation for cleats where multiple tests were performed. Note that for the artiﬁcial turf the translation test forces were limited by the capabilities of the test device for most cleats. The circled points are cleats selected for illustrative comparison (with the cleat identiﬁer numbers shown). The mechanics of American football cleats 7

Figure 4. Cleat patterns selected for illustrative comparison. (Note: images modiﬁed to anonymise manufacturer; holes added for mounting; images to scale).

In contrast, with the artiﬁcial playing surface all but one of the cleat patterns resulted in a degree of engagement of the surface that caused forces and torques reaching the limit of the

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 test device (approximately 4.8 kN force and 200 Nm torque). In those cases the cleats engaged the surface, but there was no force-limiting mechanism such as divoting or other failure of the surface. Only one cleat pattern (cleat ID #6) exhibited a force-limiting release mechanism (sliding) on the artificial surface. This resulted in forces and torques lower than the limit of the machine with this, but no other, cleat pattern on the artificial surface. The five cleats selected for further illustrative comparison are noted in Figures 2 and 3. Photos and details on these cleat patterns are shown in Figure 4 and Table I. These cleats

Table I. Cleat samples selected for illustrative comparison.

ID Cleat type Number of cleats (primary) Typical cleat depth (primary; mm)

6 Moulded rubber nubs NA* 3.5 3 Moulded plastic 9 13 4 Moulded plastic 8 12 16 Detachable plastic w/ aftermarket cleats 5 12.5 12 Detachable plastic 5 15.5

Note: *NA, Not applicable. 8 R. Kent et al.

were selected for examination because they encapsulate the range of force and moment responses observed in the tests on natural grass.

Discussion and implications When a lateral force or a torque is applied to the cleat in these tests, three things can occur: The cleat can hold with minimal displacement relative to earth (hold), the cleat can tear a portion of surface free (divot) and thus displace relative to earth, or the cleat can displace relative to the surface without tearing a portion away (slide). The second two are force-limiting mechanisms, and the magnitude of the load experienced by the foot depends on the particulars of the cleat. The first is not a force-limiting mechanism, and the magnitude of loading to the foot is dictated by the magnitude of loading applied as an input to the test. A key characteristic of artificial surfaces is that they do not, during normal use, experience acute tearing of the surface away from the substrate (i.e. the ‘divoting’ mechanism never occurred on the artificial surface tested here and would not be expected or desired with contemporary artificial surfaces). Thus, artificial surfaces like the one tested here are lacking a potentially important injury-mitigating characteristic of natural turf. Non-contact lower-limb injuries can occur on both natural and artificial surfaces (Hershman et al., 2012; Lievers, Frimenko, Crandall, Kent, & Park, 2012), but Hershman et al. showed that elite athletes have a significantly greater rate of some foot, ankle, and knee injuries on an infill artificial surface than they do on natural grass. The inability of infill artificial surfaces to divot as a load-limiting (and possibly injury mitigating) mechanism suggests that the cleat plays a more important role on artificial surfaces than on natural, at least from the standpoint of injury risk. Our results indicate that the release mechanics of cleats on artificial surfaces must be dictated by the characteristics of the cleat itself and the way it interacts with the surface, simply because the surface lacks the ability to fail and thus limit loading on the foot via a divoting mechanism. From the standpoint of cleat design, it is not unreasonable to suggest that a cleat intended for use on an infill, artificial surface should be designed to release from the surface at a lateral force on the order of the force required to divot a natural grass surface. Of course, there may be tradeoffs within the non- injurious, performance regime of the cleat-surface interaction, but ideally these regimes would be separated and cleat designs would be sought that mitigate loading in a potentially injurious regime while simultaneously maintaining performance levels of loading. When tested on natural grass, a positive relationship was observed between the peak

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 horizontal forces generated in the pre-loaded translation tests and the peak torques observed in the rotation tests. All cleat patterns were able to move relative to the natural grass surface in both translation and rotation due either to sliding (i.e. limited engagement by the cleats) or due to divoting (i.e. tearing of the surface). These results illustrate, however, that even when tearing of the surface occurs the forces generated can vary depending on the cleats. Some cleats generate higher forces and torques than others, even when the force-limiting mechanism is failure of the natural surface itself. This could be caused either by a difference in the area of the shearing interface within the soil engaged by the cleats, or by a difference in the interaction of the cleats with the substrate of the surface. In addition, the positive relationship indicates that in general the difference in engagement of cleats with natural grass surfaces tends to affect the responses under translation and rotation in similar manners. Even with the positive relationship between translation force and rotation torque on natural grass, some residual variance about the relationship was observed. Some cleats (e.g. cleat #4) resulted in a greater than average peak torque relative to their peak translation force. Others (e.g. cleat #3) exhibited the opposite. Researchers have posited that it may be beneﬁcial to reduce the amount of torque capable of being generated at the shoe-surface The mechanics of American football cleats 9

interface to reduce the risk of injury to the ankle or knee. If it is assumed that a player’s performance is related to the amount of force that can be generated under translation (e.g. in pushing off), then it may be desired to increase the force in translation while avoiding a corresponding increase in torque under rotation. Our data show that the ratio of translational to rotational traction on natural grass does depend on cleat design. For example, consider the cleats selected for illustrative comparison. Cleat #12 generated peak translational forces 19% greater than cleat #3, with only a 9% increase in rotational torque. In contrast, cleat #4 generated peak rotational torques 21% greater than cleat #3, despite exhibiting 2% less peak translational force. Since the artificial surface did not tear in the same manner as natural grass, engagement of the cleats generated forces and torques at the limit of the capabilities of the test device (Kent et al., 2012, Kent, Forman, Crandall, & Lessley, 2015) regardless of cleat pattern, with one exception. Cleat pattern #6 resulted in forces and torques with the artificial turf comparable to the range of values observed with the natural grass. This cleat pattern consisted of a large number of small (approx. 3.5 mm height) deformable nubs. Unlike the other cleat patterns, cleat #6 tended not to stay completely engaged with the artificial surface, instead sliding over the surface after a certain level of force or torque was applied. This limited the load on the foot in a manner similar functionally to the divoting behaviour of the natural grass. Neither the natural grass nor the artificial surface exhibited relationships between the peak horizontal forces generated in the translation tests and the peak horizontal forces generated in the translation/drop tests. This suggests that the peak forces generated during the translation/drop tests are more dependent on the surface and the test conditions than on the type of cleat employed. No divoting occurred in any translation/drop test, so that force- limiting mechanism was not exercised. It is possible that greater differences between cleat patterns would be observed if the translation/drop tests were performed at a greater severity, closer to the threshold where divoting affects the forces generated.

Comparison of cleat patterns This study considers the range of forces and torques generated by a sample of contemporary American football cleats. Defining all of the relationships between forces, torques, and the specific details of cleat design in a parametric, quantitative manner would be very difficult given the large number of potential cleat design parameters (e.g. number, shape, size,

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 pattern, distribution of cleats, and combinations thereof), many of which are continuously variable in multiple dimensions. The data from this study can, however, give insight on the performance of speciﬁc cleat patterns relative to the group of cleats tested. For example, on natural grass cleat #16 generated the greatest peak forces in the pre-loaded translation tests and the greatest peak torques in the pre-loaded rotation tests, so could be considered the most ‘aggressive’ pattern. By the same comparison metrics, cleat #6 was the least aggressive overall. Other cleat patterns resulted in not only different degrees of aggressivity, but also in different relationships between the translation forces and rotation torques capable of being generated. In short, these data may provide a baseline for comparison for the evaluation of future cleat designs, either in an overall sense or in targeting speciﬁc desired outcomes (e.g. increased translation force but reduced rotation torque).

Limitations and continuing work In the pre-loaded translation and rotation tests on the artiﬁcial surfaces, the forces and torques generated for most cleat patterns were limited by the inputs speciﬁed for the test 10 R. Kent et al.

device (similar to Kent et al., 2012, 2015). As a result, in those tests the peak forces and torques for most cleats tended to clump near the force-generation limit of the machine. A larger variance in the peak forces and torques would undoubtedly be observed among the different cleats if tests were performed at a greater severity (i.e. a pulling force where the cleat was forced to release from the surface). While pursuing such responses under greater forces is desirable to some extent, there comes a point where increasing the test severity becomes academic and beyond the scope of forces expected to occur during play. The maximum pulling force capable of being generated by the test device was approximately 4.8 kN, or 5.2 times the body weight of a 95 kg athlete. While this is well above the ground reaction forces typically generated by athletes during running and cutting manoeuvres (e.g. Valiant, 1987), there are limited data on the maximum ground forces capable of being generated specifically by American football players, especially during high-force tasks such as blocking. Future work should include studying such scenarios to determine if it is necessary to investigate cleat-surface interactions at greater force levels. Furthermore, this study has not evaluated the relationship between lateral force or torque and vertical load. This study considered a single value for the vertical preload (2.8 kN) applied during the translation and rotation tests. This was chosen based on previous observations that the peak vertical forces generated by an athlete performing play-relevant manoeuvres typically reached 2.5–3.5 the body weight of the athletes (Cavanagh & Lafortune, 1980; Hunter, Marshall, & McNair, 2005; Kaila, 2007; McClay et al., 1994; Stafilidis & Arampatzis, 2007; Valiant, 1987). In an actual play or practice situation there are undoubtedly many instances where different peak vertical forces are generated during manoeuvres, including vertical loads that may exceed this value (Cavanagh & Lafortune, 1980). The applied vertical load may affect the breakaway mechanics during these types of tests, including the forces and torques experienced by the foot-form (Livesay et al., 2006; Torg et al., 1974). For maximum performance during play, an athlete may desire a significant lateral force regardless of the magnitude of vertical force. Ideally, a cleat would generate the lateral force desired for performance, regardless of the magnitude of vertical force being applied, while simultaneously preventing the generation of an injurious level of lateral force or torque. As a step toward that ideal, future research should explore the sensitivity to applied vertical load in the force-generation and breakaway mechanics of cleats, particularly on artificial surfaces. This study performed tests under relatively dry and temperate environmental conditions and therefore does not contemplate the potential interactions among release mechanics,

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 cleat loads, and environmental conditions. These tests were also performed on fresh, well- maintained, intact surfaces. It is unknown how the various cleats may perform on playing surfaces that are more worn or otherwise damaged. Future work could include studying how environmental conditions and surface condition inﬂuence the responses generated with various cleat designs.

Conclusions This study investigated the interaction mechanics of a group of nineteen contemporary American football cleats tested on natural grass and in-fill type artificial turf playing surfaces. In translation and rotation tests, a wide variation in peak horizontal forces and torques was observed among the various cleats when tested on natural grass (range of average values from 2.8 kN to 4.2 kN in translation, 120 Nm to 174 Nm in rotation). In these tests on natural grass, a statistically significant positive relationship was observed between the horizontal forces generated in the pre-loaded translation tests and the torques generated in the rotation tests for the various cleats. Almost all of the cleats resulted in the natural surface divoting, The mechanics of American football cleats 11

which effectively modulated the forces generated. In contrast, almost all of the cleats held fast in the artiﬁcial turf, resulting in peak forces and torques limited only by the pressure input to the test device (up to approximately 4.8 kN and 200 Nm). Only one cleat pattern, consisting of small deformable nubs, resulted in translation-test forces with the artiﬁcial surface (3.9 kN) comparable to the range observed with natural grass.

Disclosure statement No potential conﬂict of interest was reported by the authors.

References Agel, J., Arendt, E. A., & Bershadsky, B. (2005). Anterior cruciate ligament injury in national collegiate athletic association basketball and soccer: A 13-year review. American Journal of Sports Medicine, 33, 524–531. doi:10.1177/0363546504269937 American Society for Testing and Materials (ASTM). (2006). Standard test method for traction characteristics of the athletic shoe-sports surface interface: F2333-04. In Annual book of ASTM standards (pp. 1412–1420). West Conshohocken, PA: American Society for Testing and Materials. Andreasson, G., Lindenberger, U., Renstrom, P., & Peterson, L. (1986). Torque developed at simulated sliding between sport shoes and an artiﬁcial turf. American Journal of Sports Medicine, 14, 225–230. doi:10.1177/ 036354658601400308 Bloemers, F. W., & Bakker, F. C. (2006). Acute ankle syndesmosis injury in athletes. European Journal of Trauma, 32, 350–356. doi:10.1007/s00068-006-1001-3 Bonstingl, R. W., Morehouse, C. A., & Niebel, B. W. (1975). Torques developed by different types of shoes on various playing surfaces. Medicine and Science in Sports, 7, 127–131. Bowers, Jr, K. D., & Martin, R. B. (1975). Cleat-surface friction on new and old AstroTurf. Medicine and Science in Sports, 7, 132–135. Bradley, J. P., Klimkiewicz, J. J., Rytel, M. J., & Powell, J. W. (2002). Anterior cruciate ligament injuries in the National Football League: Epidemiology and current treatment trends among team physicians. Arthroscopy, 18, 502–509. doi:10.1053/jars.2002.30649 Cavanagh, P. R., & Lafortune, M. A. (1980). Ground reaction forces in distance running. Journal of Biomechanics, 13, 397–406. doi:10.1016/0021-9290(80)90033-0 Dura, J. V., Hoyos, J. V., Martinez, A., & Lozano, L. (1999). The inﬂuence of friction on sports surfaces in turning movements. Sports Engineering, 2, 97–102. doi:10.1046/j.1460-2687.1999.00024.x Heidt, Jr, R. S., Dormer, S. G., Cawley, P. W., Scranton, Jr, P. E., Losse, G., & Howard, M. (1996). Differences in friction and torsional resistance in athletic shoe-turf surface interfaces. American Journal of Sports Medicine, 24, 834–842. doi:10.1177/036354659602400621

Downloaded by [Mr Richard Kent] at 09:17 26 June 2015 Hershman, E. B., Anderson, R. R., Bergfeld, J. A., Bradley, J. P., Coughlin, M. J., Johnson, R. J., & Tucker, A. (2012). An analysis of speciﬁc lower extremity injury rates on grass and FieldTurf playing surfaces in National Football League Games: 2000–2009 seasons. American Journal of Sports Medicine, 40, 2200–2205. doi:10.1177/ 0363546512458888 Hunter, J. P., Marshall, R. N., & McNair, P. J. (2005). Relationships between ground reaction force impulse and kinematics of sprint-running acceleration. Journal of Applied Biomechanics, 21, 31–43. Kaila, R. (2007). Inﬂuence of modern studded and bladed soccer boots and sidestep cutting on knee loading during match play conditions. American Journal of Sports Medicine, 35, 1528–1536. doi:10.1177/0363546507300257 Kent, R., Crandall, J., Forman, J., Lessley, D., Lau, A., & Garson, C. (2012). Development and assessment of a device and method for studying the mechanical interactions between shoes and playing surfaces in situ at loads and rates generated by elite athletes. Sports Biomechanics, 11, 414–429. doi:10.1080/14763141.2011.650188 Kent, R., Forman, J. L., Crandall, J., & Lessley, D. (2015). The mechanical interactions between an American football cleat and playing surfaces in situ at loads and rates generated by elite athletes: A comparison of playing surfaces. Sports Biomechanics, 14(1), 1–17. doi:10.1080/14763141.2015.1024277 Lambson, R. B., Barnhill, B. S., & Higgins, R. W. (1996). Football cleat design and its effect on anterior cruciate ligament injuries: A three-year prospective study. American Journal of Sports Medicine, 24, 155–159. Lievers, W. B., Frimenko, R. E., Crandall, J. R., Kent, R. W., & Park, J. S. (2012). Age, sex, causal and injury patterns in tarsometatarsal dislocations: A literature review of over 2000 cases. Foot, 22, 117–124. doi:10.1016/j. foot.2012.03.003 12 R. Kent et al.

Livesay, G. A., Reda, D. R., & Nauman, E. A. (2006). Peak torque and rotational stiffness developed at the shoe- surface interface: The effect of shoe type and playing surface. American Journal of Sports Medicine, 34, 415–422. doi:10.1177/0363546505284182 McClay, I. S., Robinson, J. R., Andriacchi, T. P., Frederick, E. C., Gross, T., Martin, P., et al. (1994). A profile of ground reaction forces in professional basketball. Journal of Applied Biomechanics, 10, 222–236. McHugh, M. P., Tyler, T. F., Mirabella, M. R., Mullaney, M. J., & Nicholas, S. J. (2007). The effectiveness of a balance training intervention in reducing the incidence of noncontact ankle sprains in high school football players. American Journal of Sports Medicine, 35, 1289–1294. doi:10.1177/0363546507300059 Meyers, M. C., & Barnhill, B. S. (2004). Incidence, causes, and severity of high school football injuries on FieldTurf versus natural grass: A 5-year prospective study. American Journal of Sports Medicine, 32, 1626–1638. doi:10. 1177/0363546504266978 Orchard, J., Seward, H., McGivern, J., & Hood, S. (2001). Intrinsic and extrinsic risk factors for anterior cruciate ligament injury in Australian footballers. American Journal of Sports Medicine, 29, 196–200. Queen, R. M., Charnock, B. L., Garrett, W. E., Hardaker, W. M., Sims, E. L., & Moorman, C. T. (2008). A comparison of cleat types during two football-specific tasks on FieldTurf. British Journal of Sports Medicine, 42, 278–284. doi:10.1136/bjsm.2007.036517 Ramirez, M., Schaffer, K. B., Shen, H., Kashani, S., & Kraus, J. F. (2006). Injuries to high school football athletes in California. American Journal of Sports Medicine, 34, 1147–1158. doi:10.1177/0363546505284385 Riley, P., Kent, R., Dierks, T., Lievers, W., Frimenko, R., & Crandall, J. (2013). Foot kinematics and loading of professional athletes in American football-specific tasks. Gait and Posture, 38, 563–569. doi:10.1016/j.gaitpost. 2012.03.034 Society of Automotive Engineers. (1995). J211-1 Instrumentation for impact test – Part 1 – Electronic instrumentation. In Surface vehicle recommended practice. Warrendale, PA: SAE International. Stafilidis, S., & Arampatzis, A. (2007). Track compliance does not affect sprinting performance. Journal of Sports Sciences, 25, 1479–1490. doi:10.1080/02640410601150462 Torg, J. S., Stilwell, G., & Rogers, K. (1996). The effect of ambient temperature on the shoe-surface interface release coefficient. American Journal of Sports Medicine, 24, 79–82. doi:10.1177/036354659602400114 Torg, J. S., Quedenfeld, T. C., & Landau, S. (1974). The shoe-surface interface and its relationship to football knee injuries. The Journal of Sports Medicine, 2, 261–269. doi:10.1177/036354657400200502 Valiant, G. A. (1987). Ground reaction forces developed on artificial turf. In T. Reilly, A. Lees, K. Davids, & W. J. Murphy (Eds.), Science and football: Proceedings of the first world congress of science and football (pp. 406–415). London: Routledge. Villwock, M. R., Meyer, E. G., Powell, J. W., Fouty, A. J., & Haut, R. (2009). Football playing surface and shoe design affect rotational traction. American Journal of Sports Medicine, 37, 518–525. doi:10.1177/ 0363546508328108 Woods, C., Hawkins, R., Hulse, M., & Hodson, A. (2003). The Football Association Medical Research Programme: An audit of injuries in professional football: An analysis of ankle sprains. British Journal of Sports Medicine, 37, 233–238. doi:10.1136/bjsm.37.3.233 Downloaded by [Mr Richard Kent] at 09:17 26 June 2015