EFFECTS OF FLYWHEEL STRENGTH TRAINING ON NEUROMUSCULAR PERFORMANCE OF ELITE ICE-
HOCKEY PLAYERS
Jari Puustinen Master’s thesis
Exercise and Sports Medicine School of Medicine
University of Eastern Finland April 2020
Exercise and Sports Medicine
PUUSTINEN, JARI: Effects of flywheel strength training on neuromuscular performance of elite ice-hockey players
Master’s thesis, 38 pages
Supervisors: Mika Venojärvi Ph.D, Tommy Lundberg Ph.D April 2020
Keywords: strength training, performance, ice-hockey
Traditionally, strength training has been one crucial part of ice-hockey players physical preparation. The aim of this thesis was to analyze the effects of 8 weeks (14 sessions) of flywheel strength training on strength and power performance of elite ice-hockey players.
A total of 18 male Finnish elite players (U-18 to U-21) were recruited and assigned to an experimental (EG) or control (CG) group. All the player had several years of experience from strength training but no earlier experience of flywheel strength training. The experimental group (n=9) carried out 14 sessions of flywheel training with four different exercises, 3-4 sets of 6-7 repetitions, with progressive increase in volume. The control group (n=9) carried out the same amount of training sessions using conventional strength training (barbell and free weights; 4 sets of 4-12 repetitions), employing linear progression.
Explosive power was assessed by measuring jump height in the countermovement jump (CMJ) with (20 kg and 40 kg) and without external weights. Change of direction ability and anaerobic performance were assessed by a maximal 200-m sprint test with a change of direction every 20 m. Power development in the EG was assessed by measuring the average concentric power in the 5RM bilateral squat using the flywheel machine. Results were analyzed using 2-way ANOVA (factors group and time).
Both groups improved performance and the magnitude of improvements were very similar across groups (no group x time interactions). Within-group changes for the CMJ were 5.7 ± 6.9% (p = 0.057) in EG vs. 4.8 ± 12.2% (p = 0.265) in CG. Changes for the CMJ +20 kg were 7.3 ± 7.4% (p = 0.027) in EG vs. 11.1 ± 12.9% (p = 0.032) in CG.
Changes for the CMJ +40 kg were 6.1 ± 3.1% (p = 0.002) in EG vs. 9.2 ± 16.1% (p = 0.134) in CG. In the 20 m sprint test, there were significant within-group changes in both groups (EG: -3.2 ± 1.7%, p = 0.001 vs. CG: -2.6 ± 2.2%, p = 0.008). Similarly, 40 m sprint time improved in the EG (-3.0 ± 1.7%; p < 0.001) and in the CG group (-2.6 ± 1.9%; p = 0.003). Also, 200 m total sprint time showed significant within-group changes in the EG (-1.8 ± 1.6%; p = 0.013) and CG (-1.5 ± 1.0%; p = 0.002). In the maximal power test for the EG, there was a significant training effect in 5RM (33 ± 23%; p = 0.012) and peak average power development (31 ± 25%; p = 0.018).
In conclusion, flywheel training seems to be an effective training method to improve neuromuscular performance in elite ice-hockey players. However, flywheel training does not elicit superior improvements in performance compared with traditional strength training in players with no prior experience of this training method.
Liikunta- ja urheilulääketiede
PUUSTINEN, JARI: Flywheel vastusharjoittelun vaikutukset hermo-lihasjärjestelmän suorituskykyyn ammattijääkiekkoilijoilla
Pro Gradu-tutkielma, 38 sivua
Ohjaajat: FT Mika Venojärvi, FT Tommy Lundberg Huhtikuu 2020
Avainsanat: voimaharjoittelu, suorituskyky, jääkiekko
Perinteisesti jääkiekkoilijoiden harjoitteluun kuuluu yhtenä osana voimaharjoittelu.
Tämän pro gradu tutkielman tarkoituksena oli tutkia vauhtipyörä vastusharjoittelun vaikutuksia ammattijääkiekkoilijoiden suorituskykyyn ja verrata sitä perinteisen voimaharjoittelun aikaansaamiin muutoksiin.
Tutkimukseen osallistui 18 suomalaista ammattipelaajaa, jotka jaettiin joko koeryhmään (EG, n=9) tai kontrolliryhmään (CG, n=9). Ryhmien välillä ei ollut eroja lähtötilanteessa minkään muuttujan absoluuttisen tuloksen suhteen. Pelaajien ikä oli tutkimuksen alkaessa 17-20-vuotta, pituus 163-197 cm ja paino 60-100,0 kg.
Koeryhmän kahdeksan viikon harjoittelu koostui vauhtipyörälaitteilla toteutetusta harjoittelusta kolmella erilaisella laitteella. Harjoitusinterventioon sisältyi neljä eri liikettä ja harjoitusvolyymi kasvoi progressiivisesti kolmesta sarjasta ja kuudesta toistosta neljään sarjaan ja seitsemään toistoon. Kontrolliryhmä harjoitteli vastaavan harjoituskertojen ja liikkeiden määrän perinteisellä voimaharjoittelulla lineaarisella progressiolla ja kertavolyymin ollen neljä sarjaa ja 4-12 toistoa. Suorituskyvyn testeihin kuului kevennyshyppy omalla painolla sekä 20kg ja 40kg lisäpainolla. Lisäksi suunnanmuutoskykyä ja anaerobista suorituskykyä arvioitiin 200 metrin juoksuratatestillä ja tehontuoton kehittymistä EG ryhmällä 2 jalan 5RM kyykkytestissä vauhtipyörälaitteella.
Kummallakin ryhmällä tapahtui tilastollisesti merkitsevää parannusta useimmissa mitatuissa testeissä alkutilanteeseen nähden. Ryhmien välille ei saatu tilastollisesti merkitseviä eroja lopputilanteessa minkään muuttujan absoluuttisen tuloksen tai muutosprosentin suhteen. Ryhmien muutosprosentti kevennyshypyssä oli EG (5.7 ± 6.9%; p = 0.057) vs. CG (4.8 ± 12.2%; p = 0.265). 20 kilon lisäpainolla muutosprosentti oli EG (7.3 ± 7.4%; p = 0.027) vs. CG (11.1 ± 12.9%; p = 0.032). Lisäksi 40 kilon lisäpainolla muutosprosentti oli EG (6.1 ± 3.1%; p = 0.002) vs. CG (9.2 ± 16.1%; p = 0.134). Juoksutestissä 20 metrin matkalla tapahtui tilastollisesti merkittävää kehittymistä kummassakin ryhmässä (EG: -3.2 ± 1.7%, p = 0.001 vs. CG: - 2.6 ± 2.2%, p = 0.008).
Samoin 40 metrin matkalla EG (-3.0 ± 1.3%; p < 0.001) vs. CG (-2.6 ± 1.9%; p = 0.003) sekä 200 metrin totaaliajassa EG (-1.8 ± 1.6%; p = 0.013) vs. CG (-1.5 ± 1.0%; p = 0.002).
Koeryhmän tehontuotto kehittyi merkittävästi 5RM kyykkytestissä (33 ± 23%; p = 0.012) sekä 1RM keskitehossa (31 ± 25%; p = 0.018).
Vauhtipyörä vastusharjoittelu on tehokas tapa parantaa jääkiekkoilijoiden suorituskykyä.
Tämän tutkimuksen perusteella vauhtipyörä vastusharjoittelulla saadaan aikaiseksi samankaltainen hermo-lihasjärjestelmän kehittyminen, mutta ei parempaa kuin perinteisellä voimaharjoittelulla ilman aiempaa kokemusta vauhtipyörälaitteiden käytöstä.
LIST OF ABBREVIATIONS AND DICTIONARY ... 3
1 INTRODUCTION ... 4
2 AIMS AND HYPOTHESES ... 9
3 METHODS ... 10
3.1 Subjects ... 10
3.2 Study design ... 10
3.3 Equipment ... 13
3.4 Training interventions ... 15
3.5 Performance tests ... 19
3.5.1 Jumps ... 19
3.5.2 Runs ... 19
3.5.3 Power output ... 20
3.6 Statistical analysis ... 20
4 RESULTS ... 22
4.1 Jumps ... 23
4.2 Runs ... 25
4.3 Power output ... 27
5 DISCUSSION ... 28
6 CONCLUSION ... 33
REFERENCES ... 34
LIST OF ABBREVIATIONS AND DICTIONARY
CG Control group
CMJ Countermovement jump
CON Concentric, shortening muscle action ECC Eccentric, lengthening muscle action
EG Experimental group
FW Flywheel
GRF Ground reaction force
RM Repetition maximum
1 INTRODUCTION
Ice-hockey is a major international sport with over 1.8 million registered players in the International Ice Hockey Federation’s 81 member countries. In Finland, every 85th person is a registered ice-hockey player while in Canada every 60th person is a registered player (IIHF, 2020). Ice-hockey is one of the fastest sport in the world and is primarily considered “anaerobic” by nature (Cox et al. 1995). Typically, there are 30-80 s long shifts during the game distributed over 3 periods, corresponding typically to a total playing time of 15–35 min per individual players (Roczniok et al. 2016). In single NHL games, the skating distance has been reported to be 4606±219 m (2260–6749 m), of which 2042±97 m (757–3026 m) is covered by high-intensity skating. The pattern of the game include multiple repeated-sprints where fatigue increases in the latter half of the game, which exemplifies the extraordinarily high demands on the ability to perform and recover from short, explosive, high-intensity actions (Lignell et al. 2018).
Playing position has an influence on the overall strain, where forwards usually perform more high-intensity game activities than defensemen. Top-level players cover nearly half of the total distance in a game by high-intensity skating, of which one fourth is performed at sprint speeds (>24 km/h) (Lignell et al. 2018). On the other hand, recently Vigh-Larsen et al. (2019) evaluated fitness profiles of elite and sub-elite Danish male ice hockey players (n = 275) and reported that elite players have a higher muscle mass and possess greater aerobic and anaerobic capacity compared with their sub-elite counterparts.
Muscle strength, power and anaerobic capacity are key physiological characteristics for hockey players on the ice (Roczniok et al. 2016). The intermittent, fast-paced, and high- intensity nature of the game also render it necessary for players to perform explosive accelerations, decelerations and change of directions using different skating speeds forward, backward and even sideway. Moreover, there are frequent high-impact body contacts and a need to execute skilled actions in limited time under pressure (Lignell et al. 2018). Thus, playing continuously at the elite level of ice hockey requires players to develop anaerobic sprint ability, strength, power and endurance at the same time. Hence off-ice training to optimally simulate on-ice requirements might be challenging and varies widely even at the professional level.
High-caliber hockey players achieve faster skating velocities than low-caliber players even though their stride rate might be similar. Thus, greater skating speed can be explained by both greater stride length and stride width, which are associated with greater leg strength and power (Upjohn et al. 2008). In contrast, Chang et al. (2009) observed that from a slow to medium speed, players are responding predominantly by increasing their stride length but from a medium to fast speed, stride rate becomes the predominant variable. Increased stride rate is related to lower-limb recovery rates (Mascaro et al.
1992). Indeed, Kaartinen (2019) reported that the recovery phase of the skating cycle seems to play an important role in maximal forward skating speed, where larger extension of the knee joint at the beginning of the recovery phase predicted better skating performance. For a proficient skating stride, players should focus on both returning the foot directly under their body and maximizing their lateral push together with upper body and upper limbs efficient use (Hayward-Ellis et al. 2017).
In change-of-direction (COD) maneuvers, the players must tolerate significantly greater forces on the outside skate (50–70% body weight) in comparison to the inside skate (12–
24% body weight). Furthermore, in COD tasks, the skate to be first placed on the outside is followed rapidly parallel next to the other (0,40 ms), continuing with both skates in full contact with the ice around 1 second and stopping the movement and skates under the player before the inside skate is lifted and stepped fast towards the turn of direction before the end of COD task (Fortier et al. 2014). This highlights the fact that a COD task is a fairly slow action where pressure is dominant under the outside skate. COD task in ice- hockey differs completely from running and should therefore be considered during off- ice training to prepare players to tolerate unilateral eccentric (ECC) forces, especially on the outer leg. One of the FW exercises used in this study was aiming to mimic outer leg braking and lateral push-off actions together with mediolateral arm swing. Coratella et al.
(2019) compared FW squat training to barbell back squat weight training (80% 1 RM) once a week in-season and its effects on COD. The authors reported that FW training induced different adaptations such that FW caused moderate-to-large improvements in COD ability but non-significant trivial-to-small changes with the barbell training. Thus, single weekly FW session improved COD.
Based on hockey skating kinematics, it is evident that all the major muscle groups from the lower quadrants are used in skating. Therefore, off-ice strength training should also
load and develop musculature relevant for ice-hockey, which was one the goals when planning the FW exercises for this thesis. Hockey blades are naturally unstable and impose a balance problem which necessitates participation of the medio-lateral muscles of the hip, knee and ankle. Also, in forward skating, muscles crossing the hip joint are coordinated such that the hip joint movements are triplanar, hence demanding high activation, strength and motor control from the lumbo-pelvic-hip complex (Chang et al.
2009).
The muscles responsible for knee flexion and extension play an important role in skating performance together with the powerful hip extensors. At the forward skating propulsion phase, hip adductors, abductors, external rotators and extensors from the gluteus and hamstring muscles must work in concert together with quadriceps which develop the largest contractile forces when extending the knee joint in the skating thrust (Mascaro et al. 1992, Pearsall et al. 2000). The hamstring muscles are working as a knee stabilizer during the weight shift and push off during the skating thrust. First, the quadriceps must develop torque quickly during push-off to obtain maximal velocity and during the powerful skating stride the hip extensors and abductors are the prime movers, while the hip flexors and adductors act to stabilize the hip and decelerate the limb (Mascaro et al.
1992, Tyler et al. 2001). To accommodate for the increased stride rate with higher skating speeds, the rate of hip abduction increases significantly in concert with activations of M.
adductor magnus, indicating a substantial ECC contraction. These findings highlight the functional importance of the adductor muscle group and hip abduction–adduction in skating performance as well as indirectly support the notion that groin strain injury potential increases with the skating speed, hence hip abduction–adduction training cannot be neglected (Chang et al. 2009).
The use of flywheels to provide resistance dates back to 1796 when Francis Lowndes described a whole-body exercise device called Gymnasticon, which was based on the FW principle and designed for working the muscles of the legs and upper body at the same time (Bakewell 1997). Then, in the first decades of the twentieth century A.V. Hill, August Krogh, Hansen and Lindhard and associates independently conducted experiments based on this mechanical FW principle (Krogh and Skand 1913, Hill 1920, 1922, Hansen and Lindhard 1923). Since then, the first commercial use FW machines evolved from several research findings and challenges in the 1990's when researchers at
the Karolinska Institutet in Stockholm introduced iso-inertial resistance (YoYoTM) exercise using spinning flywheels (Berg and Tesch 1992, 1994). After being tested in individuals subjected to various established spaceflight analogs, a multi-mode YoYoTM exercise machine was then installed on the International Space Station in 2009. Since then this exercise intervention has found several terrestrial applications and shown increasing evidence in enhancing sports performance, preventing sports injuries and aiding neurological or orthopedic rehabilitation as well providing evidence to promote skeletal muscle adaptations in terms of strength, power and size in healthy subjects and athletes (Tesch et al. 2017).
Recently, 4 different meta-analysis were published between the years 2017-2018 concerning FW training effects (Maroto-Izquierdo et al. 2017a, Vicens-Bordas et al.
2017, Núñez Sanchez & Sáez de Villarreal 2017, Petré et al. 2018). Petre et al. (2018) identified the effects of FW training on multiple strength-related variables affecting athletic performance and concluded that FW training is an effective method for improving several aspects of strength and power with importance for sports performance especially for trained younger individuals. Moreover, when using this training modality in shorter more intensive blocks, for a period of 4–24 weeks, FW training showed statistically significant increases in all strength aspects. Furthermore Maroto-Izquierdo et al. (2017a) examined the effects of FW resistance training with eccentric overload vs. traditional resistance exercise interventions on muscle size and functional capacities (i.e. strength and power) in athletes and healthy subjects. Based on their meta-analysis, it was indicated that brief episodes of eccentric overload induced by FW devices and performed at high intensity are associated with greater improvements in both concentric (CON) and ECC force, muscle power and muscle hypertrophy in healthy and well-trained subjects. In addition, eccentric overload training appeared to be more effective than traditional resistance exercise in promoting increases in capacities highly related to athletic performance, such as vertical jump height and running speed which were also main performance measurements in this thesis. Also, they reported that muscle power is the training outcome that experienced the greatest increase after a period of FW training with eccentric overload in healthy subjects and athletes. On the other hand, Vicens-Bordas et al. (2017) concluded, based on their meta-analysis, that inertial FW resistance training is not superior to gravity-dependent resistance training on enhancing muscle strength. The conclusion was based upon the current available literature where only three RCT´s
qualified for their meta-analysis. Data for other strength variables and other muscular adaptations were insufficient to draw firm conclusions. The purpose of the Núñez Sanchez & Sáez de Villarreal (2017) meta-analysis was to determine if chronic training using a FW paradigm in healthy people increase muscle volume and force. They found out that FW training increases muscle mass after short periods of training, which was not influenced by the existence of eccentric overload during the exercise, but previous familiarization with FW training facilitated the improvements even more. In summary, there is evidence that FW training is an effective method for improving several aspects of strength and power hence development of athletic performance in well-trained athletes but evidence comparing FW training to other modalities like free weights in well- designed training studies is lacking. Based on the physiological demands of ice-hockey and the scientific evidence of FW strength training, it is evident that FW could be effective in increasing physical performance of ice-hockey players during off-season training, or could be used as a time-efficient high-load exercise stimulus for in-season training to maintain muscle performance and capacity.
2 AIMS AND HYPOTHESES
The purpose of this study was to examine the effects of FW strength training, using YoYoTM Technology machines, on strength, speed, change of direction ability, anaerobic performance, and power development of elite ice-hockey players. Based on earlier findings, it was hypothesized that FW training would lead to greater performance gains than traditional strength training using free weights.
3 METHODS 3.1 Subjects
Eighteen male Finnish elite ice-hockey players (U-18 to U-21, age 18.6 ± 0.8 years, height 181 ± 9 cm, weight 78.9 ± 9.6 kg, body fat 12.3 ± 2.4%, body-mass index 24.6 ± 2.4 kg/m2 ) voluntarily accepted to participate in this study (TABLE 1.). Data collection took place during the first pretest sessions right before the off-season training period and during the posttest sessions after the training period. All the tests were executed and registered by the director of Varala Testing Lab, M.Sc Marko Haverinen in Varala Sports Institute, Tampere, Finland. Players belonged to a highest Finnish ice-hockey Championships League (ie, Liiga) or A-junior (U-20) ice-hockey club squad. None of the participants had previously used FW devices but had several years of strength training background with free weights. After a detailed explanation about the aims, benefits, and risks involved in this study, all participants gave their informed written consent to participate. The study protocol was approved by the University of Eastern Finland and conformed to the standards set by the Declaration of Helsinki.
TABLE 1. Descriptive Data of the Participants, Mean and standard deviations (SD).
Age (years) Height (cm) Weight (kg) Body fat (%) BMI (kg/m2) EG group 18.9 ± 1.0 180.0 ± 7.8 80.0 ± 9.1 12.9 ± 2.6 24,7 ± 2.0 CG group 18.3 ± 0.5 182.7 ± 11.1 77.7 ± 10.7 11.7 ± 2.7 23,3 ± 3.1
3.2 Study design
Using a controlled nonrandomized study design, eighteen professional male ice-hockey players were recruited and assigned to an experimental group (EG, n = 9) or control group (CG, n = 9). Recruitment and assignment were done by the team coaches and director of Varala Testing Lab, M.Sc Marko Haverinen. The players could not be randomized because the flywheel training intervention required commitment in the facility where the FW machines were located for the duration of the training period. The training period also included a player self-training phase, which excluded some of the players who were training in other geographical location during the intervention. Other determinants were
health status which allowed FW training and homogeneity with the other participants in the study. Some of the players were not able to train in a team setting because of operated or non-operated injury but they were able to participate in FW training, so they were assigned into the experimental group. Injuries were located on the shoulder or low back and therefore using barbell exercises like power clean or back squat were not tolerable.
Exclusion criteria were:
- Training facility not available during the intervention - Injuries or illness that interfere with strength training - Earlier experience of FW training
Both groups were otherwise involved in normal training regimen with a similar weekly training volume and methodology. Furthermore, familiarization sessions (2) with the flywheel devices and exercises used in the study were allowed and supervised in the EG group. In these sessions, a full explanation of the experimental protocol and recommendations were given to the participants, and they were permitted to practice power tests with the squat machine. In the first training week session, the EG group performed a test where the inertia used during the intervention (first 4 training weeks) with the FW squat machine was selected based on best average CON power output of the set. This inertia was then readjusted after 4 weeks of training for optimizing individual maximal power for the squat exercises. Also, bilateral squat power test was measured and registered for later analysis. Thereafter, each player was responsible for his own training and keeping track of times and days and marking those in an exercise diary in both groups.
Only players who participated in 3/4 of the training sessions and both pre- and post-test sessions were included in the final analyses.
When designing FW training protocol for this study earlier research findings were exploited. After FW and eccentric overload training, the metabolic and perceptual demands are high to 48 hours post exercise. During the off- or pre-season as FW training was completed in this study per se, it is recommended to perform a maximum of 2 sessions per week of lower-limb eccentric overload training separated by at least 2 days of rest to enable adequate recovery time. Then during the season or competitive phase lower-limb eccentric overload training could be wise to reduce to 1 session at the beginning of the week or programmed by the game calendar (Raeder et al. 2016).
The subjects completed a conventional strength and power training period lasting for 8 weeks from June to July in 2016 or 2017 during their off-seasonal training phase.
Performance tests included jumps, runs and power output tests and were executed immediately (5-7 days) before and after the training period (Figure 1). Force-velocity characteristics, change of direction ability and anaerobic performance was analyzed from the test results. The subjects performed identical strength and power training period with the exception that EG strength and power training were carried out using YoYoTM Technologymachines.
Figure 1. Overview of performance tests
3.3 Equipment
In this experimental study, 3 different YoYoTM Technologymachines were used to target the main muscle groups which are used in ice-hockey. The machines used were the squat, leg curl and leg press (Picture 1). With the squat machine, it is possible to train uni- and bilateral squats which both were implemented in this study.
Picture 1. Machines used in the study (with the permission of YoYo Technology AB, Stockholm, Sweden).
When using the YoYoTM training principle, the energy which is generated in the CON action is converted into kinetic energy, which then accumulates in the flywheel and is removed in the following braking action. Any FW machine generates resistance by opposing to the trainee’s effort with the inertial force generated by a rotating FW. The same inertia must be overcome during each repetition by means of accommodated loading. The more energy that is transferred to the FW during the CON action, the faster will the FW spin and more power is produced. When this kinetic energy is decelerated in a restricted part of the ECC action, force must be produced that exceeds the force generated in the CON phase. Definition of eccentric overload exercise, which is characteristic of the YoYoTM training principle, comes from this breaking action. Every repetition must be stopped by the trainee before the machine’s mechanical stop to achieve the desired training effect (Tesch et al. 2017).
The YoYoTM Leg Press ergometer (Leg Press, YoYo Technology AB, Stockholm, Sweden) was originally introduced by Berg and Tesch (1994). The machine is designed to load the knee and ankle extensor muscles. In this experimental study, the FW device used by the EG was equipped during the whole training period with two flywheels
(material: aluminum; density: 2,8 g/cm3, diameter: 370 mm; thickness: 7 mm) with moment inertia of 0.05 kg·m2 resulting total inertia of 0.1 kg·m2
The YoYoTM Leg Curl machine (Leg Curl, YoYo Technology AB, Stockholm, Sweden) was first investigated in a study which evaluated preseason training effects for injury prevention, and strength and speed performance in elite soccer players (Askling et al.
2003). The FW device used by the EG was also equipped the whole training period with two flywheels (material: aluminum; density: 2,8 g/cm3, diameter: 370 mm; thickness: 7 mm), with moment inertia of 0.05 kg·m2 resulting total inertia of 0.1 kg·m2. The machine is designed to load all the hamstring muscles as it was showed in a study by Mendez- Villanueva et al. (2016).
The YoYoTM Squat machine (Squat Ultimate, YoYo Technology AB, Stockholm, Sweden) was originally introduced by Berg and Tesch (1998) when they highlighted this lightweight and compact multi-exercise configuration. With this machine, when doing uni- or bilateral squat exercises, the cord is anchored to a harness which users wear during the exercise. The machine is designed to load mostly knee and hip extensor muscles, hip adductor and abductor muscles and supporting trunk and core muscles. The FW device used by the EG was equipped with three flywheels, which of two flywheels had moment inertia of 0.05 kg·m2 (material: aluminum; density: 2,8 g/cm3, diameter: 370 mm;
thickness: 7 mm) and one with moment inertia of 0.025 kg·m2 (material: aluminum;
density: 2,8 g/cm3, diameter: 185 mm; thickness: 3,5 mm). During the first training session, the inertia that was used for the first 4 weeks was chosen between inertia 0.025 kg·m2, 0.05 kg·m2 and 0.075 kg·m2. The inertia that achieved highest average power output per 5 repetition set was selected. Similar assessment was carried out during the fourth training week and inertia was chosen similarly between inertia 0.05 kg·m2, 0.075 kg·m2 and 0.1 kg·m2. The protocol was performed using a specific sensor technology measurement system designed for the YoYoTM FW machines (BlueBrainTM Power Encoder, YoYo Technology AB, Stockholm, Sweden) with associated BlueBrain software (v 1.5). This power encoder was attached only to the squat machine and therefore optimizing individual maximal power was possible only with the squat exercises.
3.4 Training interventions
The CG performed the training program using the linear periodization model with possible to daily undulate the resistance as high as tolerated. The training program consisted of 4 different lower body exercises which were back squat, power clean, pistol squat and walking lunge with long step. Back squat and power clean weights were chosen after the pretest session where 1 RM of those lifts were tested. During the first two weeks of the intervention, the players performed back squat and power clean exercises so that the first set was 12 RM, second 10 RM, third 8 RM and fourth 6 RM. The Pistol squat exercise consisted of 4 sets of 8 repetitions with each leg and it was done with 10-15 kg extra weights or as high as tolerated. The walking lunge exercise consisted of 4 sets of 10 repetitions with each leg and was done with long forward steps and added 30-40 kilograms extra weights or as high as tolerated. The above-mentioned program was performed identically twice a week by both EG and CG from 4 weeks before the intervention started. After first 2 weeks of the intervention, the program was modified in the CG as shown in Figure 2.
The EG training program consisted of 4 YoYoTM exercises which were bilateral squat, unilateral squat, leg curl and leg press. All the exercises were carried out with same incremental volume methodology and program was advised to do in this order (1.
Bilateral squat, 2. Unilateral squat, 3. Leg curl and 4. Leg press). The first two weeks of exercises were performed once a week and next 6 weeks two times per week. Volume was increased so that during the first four weeks were 3 times 6 repetitions, fifth and sixth weeks were 4 times 6 repetitions and seventh and eight weeks were 4 times 7 repetitions.
Break between the sets was 3 minutes and 3-5 minutes between the exercises.
Figure 2. Study design – timeline
In the bilateral squat exercise, participants were advised to make the transition phase from ECC to CON between 90-110 knee angle and then perform the CON phase as fast as possible. The strap was adjusted so that it was possible to achieve full extension from knee and hip, so the starting position was fully extended standing position. It was advised to repeat the first couple of controlled repetitions and once good rhythm and position was achieved, applying maximal CON effort for the first counted repetition. This starting instruction was the same for all four exercises.
The unilateral squat (Picture 2.) was advised in a similar manner, pushing knee and hip in almost full extension and breaking the ECC action around 90 degrees of knee flexion.
Movement direction was supero-laterally, pushing leg against the foot stance and the
other leg on the lateral foot platform stabilizing the machine to the floor and allowing maximal effort for unilateral squat. The unilateral exercise was developed to mimic skating push off action during skating. Together with the foot work, mediolateral arm swing was executed to maximize the outer leg ground reaction forces (GRF´s) comparable to each skating stride (Hayward-Ellis et al. 2017).
Picture 2. Start and turning point position of the unilateral skating squat exercise.
With the leg curl machine, participants performed bilateral knee flexion exercise (Picture 3.) where they were advised to emphasize maximal effort through the entire CON action until the turning point. Then, after the strap started to rewind in the ECC phase, it was advised to resist gently first 20-30 degrees and then breaking the action before the mechanical stop at around 170 degrees knee angle and then continuing to the next repetition with maximal effort without pausing the movement. In the ECC phase, the instruction was to keep the ankles in plantar flexion to minimize calf musculature involvement and the strap was adjusted so that maximal knee flexion was possible to achieve.
Picture 3. Start and turning point position of the bilateral knee flexion exercise.
In the leg press machine, seated bilateral ankle, knee and hip extension exercise (Picture 4.) the subjects were advised to execute the motion from around 90 degrees knee angle to almost full knee and ankle extension in the CON phase until the turning point when flywheel strap starts to rewind. Then in the ECC phase, the first 20-30 degrees should be gently resisted and then breaking the action around 90 degrees knee angle before hitting the mechanical stop and then continuing to the next repetition with maximal effort and without pausing the movement.
Picture 4. Start and turning point position of the bilateral leg press exercise.
Both CG and EG performed the same amount of exercise sessions with strength training and both groups were otherwise involved in the normal training regimen with an identical weekly training volume and methodology. Lower limb exercise sessions were separated in both groups by 48-96 hours. Warm-up and supplementary exercises were the same in both groups in the strength training sessions. Warm-up consisted of biking/running/rowing, followed by 8 complementary warm-up exercises before the main exercises. Sessions ended with to 3x15 repetitions of toes to bar exercise and 3x5 repetitions of hurdle jumps followed by short sprints.
3.5 Performance tests 3.5.1 Jumps
Explosive power was assessed by measuring the height of the center of mass during countermovement jump (CMJ) using an infrared-lighting mat (SpintestTM, Spin Test LCC, Tallinn, Estonia). The CMJ starts in an upright position - subjects were asked to squat and change direction from 90 degrees knee angle and jump upwards as high as possible. The change of direction was done as fast as possible without a pause and hands were kept on the hips during the whole movement. Furthermore, landing was done with straight legs on the ball of the feet and after the first ground contact subjects were allowed to bend the knees. The CMJ was also performed with extra weights of 20 kg and 40 kg loads on the barbell. The subjects performed 3 jumps in each test and the highest result was used for later analyses. Participants were allowed to recover for 30 seconds between repetitions and 5 minutes between individual tests.
3.5.2 Runs
Change of direction ability and anaerobic performance (Figure 3.) were assessed by performing a maximal 200-m sprint with a change of direction every 20 m. A dual-beam electronic timing gate (SpintestTM, Spin Test LCC, Tallinn, Estonia) was placed in the middle of the 20-m section which allowed to record split times in every 20-m shuttle during the whole 200-m distance. This way every 20-m shuttle consisted of two 10-m accelerations divided by a change of direction. The starting position was standardized by keeping another foot behind the starting line and subjects were asked to accelerate maximally for 10 meters before the timing started. During the change of direction, self- selected foot and hand were instructed to touch to the ground on the line. The change of direction was done on the same side during the whole test. The test was performed on an indoor hall plastic surface, and subjects wore running shoes.
The following variables were used for posterior analysis:
Speed (anaerobic alactic power):
- Time of the first shuttle (10+10 m) including a change of direction
- Time of the first two shuttles (2*10+10m) including a change of direction Speed endurance (anaerobic lactic capacity):
- total time of ten shuttles (10*10+10 m) including a change of directions - profile of the anaerobic performance (split times of each 20-m shuttles)
Figure 3. Change of direction ability and anaerobic performance measurements
3.5.3 Power output
CG power measurements were done with the power encoder (BlueBrainTM Power Encoder, YoYo Technology AB, Stockholm, Sweden) with associated BlueBrain software (v 1.5) which is developed for the use with any YoYoTM Technology machine.
Average power was assessed ~3 days before and ~5 days after the intervention. The players completed 3 sets of 5 maximal coupled CON-ECC bilateral squats using a FW squat machine (YoYo Technology AB, Stockholm, Sweden). The power test was done with the moment inertia of 0.05 kg·m2 (material: aluminum; density: 2,8 g/cm3, diameter: 370 mm; thickness: 7 mm) before and after the intervention. Results were registered as CON average power per set and reps and was saved for later analyses.
3.6 Statistical analysis
Data are presented as mean ± SD. Mean and standard deviation (SD) were calculated for all variables. Statistical analysis was performed using SPSS 25.0 for Windows (SPSS Inc, Chicago, IL, USA). Variables were analyzed by a 2-way analysis of variance (ANOVA) with repeated measurements: group (EG and CG) and time (pre- and post-training).
Follow-up comparisons were made with paired t-tests within each group. Statistical significance was set at p ≤ 0.05.
4 RESULTS
The key performance data are shown TABLE 2. No substantial improvements were found in the CMJ in any group compared with the pretest but EG showed substantial improvements in CMJ +20kg and CMJ +40kg and CG in CMJ +20kg, despite no overall interaction effect in the 2-way ANOVA. The 200 meters running test results within-group analysis revealed substantial changes from pre- to post-training for most of the total and split times.
Results from the between-groups analysis are illustrated in Figure 4 and Figure 11. No significant group by time interaction was found in any jump test. Furthermore, no significant group by time interaction was observed for any of the 200 meters running test, total or split times.
TABLE 2. Outcome variables before (PRE) and after (POST) 8 weeks of resistance training in the flywheel (EG) and control (CG) group, Mean ± SD
EG CG
VARIABLE PRE POST Δ% PRE POST Δ%
CMJ (cm) 41.58 ± 4.02 43.79 ± 3.18 5.7 42.09 ± 2.56 44.11 ± 5.63 4.8
CMJ 20 kg (cm) 30.23 ± 2.58 32.39 ± 2.93 7.3 29.46 ± 2.75 32.71 ± 4.76 11.1 CMJ 40 kg (cm) 23.44 ± 2.59 24.84 ± 2.65 6.1 22.79 ± 2.54 24.80 ± 4.05 9.2 10+10-m sprint (s) 4.11 ± 0.08 3.99 ± 0.11 -3.2 4.05 ± 0.07 3.94 ± 0.06 -2.6 2*10+10-m sprint (s) 8.26 ± 0.19 8.03 ± 0.20 -3.0 8.13 ± 0.14 7.92 ± 0.10 -2.6 3*10+10-m sprint (s) 12.45 ± 0.25 12.12 ± 0.31 -2.7 12.31 ± 0.17 12.00 ± 0.16 -2.5 4*10+10-m sprint (s) 16.68 ± 0.41 16.30 ± 0.42 -2.3 16.55 ± 0.23 16.13 ± 0.25 -2.5 5*10+10-m sprint (s) 21.04 ± 0.48 20.49 ± 0.58 -2.5 20.82 ± 0.31 20.36 ± 0.30 -2.2 6*10+10-m sprint (s) 25.39 ± 0.58 24.74 ± 0.67 -2.5 25.17 ± 0.36 24.68 ± 0.36 -1.9 7*10+10-m sprint (s) 29.77 ± 0.68 29.10 ± 0.87 -2.0 29.60 ± 0.41 29.04 ± 0.45 -1.9 8*10+10-m sprint (s) 34.27 ± 0.74 33.46 ± 0.99 -2.2 34.06 ± 0.45 33.42 ± 0.52 -1.9 9*10+10-m sprint (s) 38.77 ± 0.80 37.89 ± 1.12 -2.1 38.47 ± 0.52 37.91 ± 0.63 -1.4 10*10+10-m sprint (s) 43.24 ± 0.84 42.33 ± 1.23 -1.8 43.05 ± 0.62 42.41 ± 0.72 -1.5 20-40-m sprint split time (s) 4.15 ± 0.15 4.01 ± 0.12 -3.4 4.08 ± 0.12 3.97 ± 0.07 -2.7 40-60-m sprint split time (s) 4.19 ± 0.10 4.07 ± 0.12 -2.8 4.18 ± 0.07 4.03 ± 0.09 -3.4 60-80-m sprint split time (s) 4.23 ± 0.24 4.13 ± 0.12 -1.8 4.24 ± 0.08 4.12 ± 0.10 -2.8 80-100-m sprint split time (s) 4.36 ± 0.16 4.20 ± 0.18 -3.3 4.27 ± 0.10 4.20 ± 0.10 -1.5 100-120-m sprint split time (s) 4.35 ± 0.12 4.24 ± 0.13 -2.5 4.35 ± 0.08 4.28 ± 0.09 -1.5 120-140-m sprint split time (s) 4.37 ± 0.17 4.32 ± 0.25 -0.6 4.43 ± 0.12 4.35 ± 0.10 -1.8 140-160-m sprint split time (s) 4.50 ± 0.11 4.32 ± 0.16 -3.8 4.46 ± 0.11 4.39 ± 0.09 -1.6 160-180-m sprint split time (s) 4.51 ± 0.16 4.39 ± 0.16 -1.8 4.41 ± 0.22 4.45 ± 0.14 0.9 180-200-m sprint split time (s) 4.47 ± 0.15 4.43 ± 0.13 -0.3 4.58 ± 0.22 4.49 ± 0.14 -1.9
4.1 Jumps
Both protocols showed comparable increases in CMJ, CMJ +20kg and CMJ +40kg height (Figure 4). No significant group by time interaction or within-group differences were observed for the CMJ test (p=0.863). Within-group differences for the CMJ were in the EG 5.7 ± 6.9%; p = 0.057, and in the CG group 4.8 ± 12.2%; p = 0.265 (Figure 5).
Figure 4. CMJ jump tests group mean differences
Figure 5. Pre- and Post-training CMJ height and mean change. No significant group by time interaction (p > 0.05). No significant planned comparisons within-group differences (p > 0.05).
No significant group by time interaction was observed for the CMJ +20kg test (p=0.474).
Significant within-group differences for the CMJ +20kg were observed in the EG (7.3 ± 7.4 %; p = 0.027) and CG group (11.1 ± 12.9 %; p = 0.032). Figure 6 shows the CMJ+20 kg height (cm) results before and after the intervention.
Figure 6. Pre- and Post-training CMJ + 20 kg height and mean change. No significant group by time interaction (p > 0.05). Significant planned comparisons within-group differences (p < 0.05).
There was no significant group by time interaction in CMJ+40kg test (p=0.625).
Significant within-group differences for the CMJ were observed in the EG (6.1 ± 3.1%;
p = 0.002) vs. CG group (9.2 ± 16.1%; p = 0.134). Figure 7 shows the CMJ+40 kg height (cm) results before and after the intervention.
Figure 7. Pre- and Post-training CMJ + 20 kg height and mean change. No significant group by time interaction (p > 0.05). Significant planned comparisons within-group differences in EXP group (p < 0.01).
4.2 Runs
No significant group by time interaction was observed for any of the 200 meters running test total or split times. However, planned comparisons within-group analysis revealed statistically significant changes from pre- to post-training for most of the total and split times.
There was no significant group by time interaction in 20 meters sprint time (p=0.517).
Planned comparisons of the 20 m sprint test revealed significant within-group changes in both groups (EG: -3.2 ± 1.7%, p = 0.001 vs. CG: - 2.6 ± 2.2%, p = 0.008). Figure 8 shows the 20 m sprint times before and after the intervention and mean change.
Figure 8. Pre- and Post-training 20 meters sprint time and mean change. No significant group by time interaction (p > 0.05). Significant planned comparisons within-group differences in CG (p <
0.05). Significant planned comparisons within-group differences in EG (p < 0.01).
There was no significant group by time interaction in 40 meters sprint time (p=0.605).
Significant within-group differences were observed in the EG (-3.0 ± 1.3%; p < 0.001) and CG group (-2.6 ± 1.9%; p = 0.003). Figure 9 shows the 40 meters sprint time (s) results before and after the intervention and mean change.
Figure 9. Pre- and Post-training 40 meters sprint time and mean change. No significant group by time interaction (p > 0.05). Significant planned comparisons within-group differences in both groups (p < 0.01).
There was no significant group by time interaction in 200 meters total sprint time (p=0.572). Significant within-group differences were observed in the EG (-1.8 ± 1.6%; p
= 0.013) and CG group (-1.5 ± 1.0%; p = 0.002). Figure 10 shows results before and after the intervention and mean change.
Figure 10. Pre- and Post-training 200 meters sprint time and mean change. No significant group by time interaction (p > 0.05). Significant planned comparisons within-group differences in EG (p < 0.05). Significant planned comparisons within-group differences CG (p < 0.01).
The fatigue profile was similar between the groups. Furthermore, speed and COD profile i.e. first two shuttles improvements, were similar between the groups without any significant group by time interaction. Similar findings were found in speed endurance profile. In summary, even though EG improved 14/19 of the running test total or split times results more than CG, there were no group x time interactions (Figure 11).
Figure 11. Running test group mean differences
4.3 Power output
In the maximal power output test for the EG, there were significant within-group differences on 5RM (33 ± 23 %; p = 0.012) and peak repetition (31 ± 25 %; p = 0.018) AVG power development (W). Figure 12. shows the results before and after the intervention.
Figure 12. Pre- and Post-training power output (W) 5 repetition and 1 repetition peak average power and mean change. Significant planned comparisons within-group differences (p < 0.05).
5 DISCUSSION
This study aimed to assess the effects of 8 weeks of FW training compared to traditional strength training on neuromuscular performance of elite ice-hockey players. The results showed that both training interventions led to performance improvements of similar magnitude. There were no significant group x time interactions, but several main training effects.
The current study shows that FW training is an effective method for improving several aspects of strength and power with relevance for ice-hockey performance. Indeed, there were substantial improvements in capacities highly related to athletic performance, such as vertical jump height, running speed and endurance as well as power output. These findings are supported by previous meta-analyses (Petre et al. 2018, Maroto-Izquierdo et al. 2017a) where several aspects of neuromuscular performance have been shown to be enhanced by FW training. Thus, elite ice-hockey players could include flywheel training into their training regimen for performance enhancement.
The current study showed that FW training induces significant improvements in power (33%) after 14 sessions. This was an important finding given that muscle strength, power and anaerobic capacity are key physiological characteristics for hockey players on the ice (Roczniok et al. 2016). The magnitude of improvement was in line with Petre et al.
(2018), who reported a significant increase of 25% in power of large effect size from pre- to post-test after 4-24 weeks of flywheel training and 1-3 sessions per week. Similarly, Maroto-Izquierdo et al. (2017a) reported that muscle power is the training outcome that experienced the greatest increase after a period of FW training with eccentric overload in healthy subjects and athletes.
In this study, there were significant improvements (around 3%) in 20 and 40 meters sprint time in both groups. In the 200 meters sprint, the improvement was 1.5%. Comparable results were found by Petre et al. (2018) who found significant improvement of 2.4% in horizontal displacement with a large effect size from pre- to post-test after 6–10 weeks of flywheel training 1–3 times per week. For vertical displacement, the current study report performance improvements of 5-7% in CMJ jumps. Similar findings were found by Petre
et al. (2018) who showed that there was a significant increase of 6.8% in vertical displacement with a large effect size from pre- to post-test.
This study showed similar magnitudes of improvements of performance in both training groups. This might be due the small sample size and the fact that the training was programmed in a similar manner in both groups and was based on strength training methods already proven effective. Furthermore, since all the players were trained athletes with background of several years strength training, small improvements are expected.
It is evident that strength training with free weights is an efficient way to gain performance enhancement for elite athletes. This study supports that conclusion. Well programmed general strength training interventions are effective and experienced younger athletes tend to develop significant improvements in performance in a short period of time. Most of the earlier studies have compared FW training to conventional strength training in one single exercise to other single exercises like FW leg press exercise vs. weight-stack leg press exercise but not the multi-movement exercise program as in this study (Maroto-Izquierdo et al. 2017b). Núñez Sanchez & Sáez de Villarreal (2017) earlier reported that the ability to produce an eccentric overload in the FW system appears to require some earlier experience of the training method. The ability to produce an eccentric overload is indicative of greater mechanical loading, potentially resulting in more robust training stimulus. None of the players in this study had earlier experience of FW training which can influence of the capacity to achieve eccentric overload during the exercise. Thus, it is possible that performance could have been even better if the players had earlier experience of the training method. In contrast, the players were experienced to training with free weights, which can potentially favor improvements in the control group. Ideally, future studies should compare FW training to other conventional strength training methods with experienced users to the training method per se.
In the research literature, there is some evidence that FW and eccentric overload training is superior to conventional strength training in terms of performance enhancement as well as muscle anabolic response and adaptations in well-trained athletes (Friedmann-Bette et al. 2010, Norrbrand et al. 2011, Maroto-Izquierdo et al. 2017b). This study did not support that. Well-trained athletes familiar to free weights but not to FW might benefit more from free weights. Conversely, more experienced athletes using eccentric overload training
could perhaps reach better performance gains and muscle adaptations as shown by Friedmann-Bette et al. (2010).
One of the goals of this study was to load all the main leg muscles used in ice-hockey with large forces. As Kaartinen (2019) presented, hockey players would gain from training programs involving motions of large forces and with maximal range of motion as extension of the knee joint at the beginning of the recovery phase predicted better skating performance. Unilateral squat with FW machine mimics this propulsion action.
On the other hand, unilateral training with FW squat machine mimics COD, skating thrust mechanics and SSC stimulus on-ice and leaves opportunity to use the medio-lateral arm swing to increase GRF’s on the outer leg which is almost impossible with barbell training.
These are all important and relevant aspects when designing general and sports specific strength training exercises to mimic the demands of ice-hockey. Deep squats, full-range knee extensions, and full-range leg presses are exercises that need to be incorporated into a hockey training program, since these exercises mobilize the lower extremities through a range of motion similar to that for hockey skating (Lafontaine 2007). Behm et al. (2005) presented that either skating-specific resistance exercises or exercises that can specifically isolate the hamstrings are required to provide the necessary level of activation for hockey skating. In that regard, the FW leg curl is a viable option and is complemented by the FW squat since it increases the hamstrings-to-quadriceps ratio alone but not with the traditional squat training using barbell. In addition, the specific FW squat training alone increases hamstring ECC & quadriceps CON ratio which could be beneficial for injury prevention (Coratella et al. 2019).
Running force is generally generated and applied to the ground in a single sagittal plane.
Skating force, however, is applied to the ice not only sagittally but also through the frontal plane. This difference in the application of force requires differences in joint mechanics, particularly in the hip. Therefore, testing and evaluating linear skating speed should also ideally be done on-ice. Off-ice testing measures that will accurately measure progress are necessary to determine the effectiveness of prescribed training regimens, especially during the off season. These tests should be relatively easy to administer yet highly correlated with skating speed and hockey performance (Runner et al. 2015). During the off-season, the availability of ice time is reduced, and thus, for adequate monitoring of off-season training, testing would be appropriate to take place off-ice also. Farlinger et
al. (2007) presented that measures of horizontal power are the strongest predictors of on- ice skating performance that includes an acceleration component typical of normal game play. Mascaro et al. (1992) reported that vertical jump anaerobic power was the single best predictor of skating speed per se. Behm et al. (2005) found significant correlations between skating performance and the sprint and balance tests but not in CON squat, jump, drop jump, 1 repetition maximum leg press or flexibility tests. The off-ice tests used in this thesis were pre-determined by the ice-hockey team and Varala Sports Institute.
FW training is a safe form of exercise to achieve short bouts of eccentric overload controlled by the user and the current study confirmed that there is no need for prior experience to reach some benefit from FW training. There are also several methodological options to manipulate the ECC load when players can tolerate more ECC force and get customized with the traditional FW training resistance paradigm. ECC training has also been shown to be efficient to prevent lower limb muscle injuries and muscle problems in elite athletes (Goode et al. 2015, Harøy et al. 2018). Therefore, high priority should be placed on hockey players lower limb muscles for injury prevention. Douglas et al. (2016) have determined that ECC training is also a potent stimulus for enhancements in muscle mechanical function, and muscle-tendon unit morphological and architectural adaptations associated with a faster (i.e. explosive) phenotype.
Free weights, FW machines and other resistance training modalities have and will have their own role and benefits for athletic development. Thus, coaches and athletes should consider implementing a variety of different training methods to enhance athletic performance.
A limitation of this study was the non-randomization of the players which may have influenced on the player characteristics even though players characteristics were comparable across groups. Also, the strength training sessions were controlled by the players themselves by completing an exercise diary which could possibility give rise to errors compared to supervised training. Furthermore, it would be of interest to analyze the effect of this intervention on the most representative performance tests in ice-hockey on-ice. As earlier training studies have shown, it is difficult to get significant group x time interactions across training interventions in small sample sized studies where the improvements are not expected to be large. Further studies are needed to analyze the
effects of eccentric overload FW training on indices of ice-hockey performance on-ice and compare them to well-designed conventional strength-training programs. This could be done both for reduction of injuries and for improving performance on-ice, in larger sample sized interventions with players experienced in the training method.
6 CONCLUSION
Based on the findings of this experimental training study, it is concluded that both flywheel and conventional resistance training improves performance proxies relevant to ice-hockey players. There were no significant group x time interactions, suggesting similar magnitudes of improvements across groups. Thus, FW training seems to be an effective training method to improve neuromuscular performance in elite ice-hockey players. However, flywheel training does not elicit superior improvements in performance compared with traditional strength training in players with no experience of this training method.
REFERENCES
Askling C, Karlsson J, Thorstensson A. Hamstring injury occurrence in elite soccer players after preseason strength training with eccentric overload. Scand J Med Sci Sports 2003;13:244-250.
Bakewell S. Illustrations from the Wellcome Institute Library: Medical gymnastics and the Cyriax collection. Medical History 1997;41:487-495.
Behm DG, Wahl MJ, Button DC, Power KE, Anderson KE. Relationship between hockey skating speed and selected performance measures. Journal of Strength and Conditioning Research 2005;19(2):326–331.
Berg HE, Tesch PA. Designing methods for musculoskeletal conditioning in weightlessness. Physiologist 1992;35(1 Suppl):S96‐8.
Berg HE, Tesch PA. A gravity‐independent ergometer to be used for resistance training in space. Aviat Space Environ Med 1994;65(8):752‐6.
Berg HE, Tesch PA. Force and power characteristics of a resistive exercise device for use in space. Acta Astronautica 1998;42:219-230.
Chang R, Turcotte R, Pearsall D. Hip adductor muscle function in forward skating, Sports Biomechanics 2009;8:3,212-222.
Coratella G, Beato M, Cè E, Scurati R, Milanese C, Schena F, Esposito F. Effects of in- season enhanced negative work-based vs traditional weight training on change of direction and hamstrings-to-quadriceps ratio in soccer players. Biology of Sport.
2019;36:3,241-248.
Cox MH, Miles DS, Verde TJ, Rhodes EC. Applied physiology of ice hockey. Sports Med. 1995;19(3):184-201.
Douglas J, Simon Pearson S, Ross A, McGuigan M. Chronic Adaptations to Eccentric Training: A Systematic Review. Sports Med. 2016. DOI 10.1007/s40279-016-0628-4 Farlinger CM, Kruisselbrink LD, Fowles JR. Relationships to skating performance in competitive hockey players. Journal of Strength and Conditioning Research 2007;21(3):
915-922.
Fortier A, Turcotte RA, Pearsall DJ. Skating mechanics of change-of-direction manoeuvres in ice hockey players. Sports Biomechanics 2014;13:4,341-350.
Friedmann-Bette B, Bauer T, Kinscherf R, Vorwald S, Klute K, Bischoff D, Müller H, Weber M-A, Metz J, Kauczor H-U, Bärtsch P, Billeter R.. Effects of strength training with eccentric overload on muscle adaptation in male athletes. Eur J Appl Physiol 2010;108:821–836. DOI 10.1007/s00421-009-1292-2.
Goode AP, Reiman MP, Harris L, DeLisa L, Kauffman A, Beltramo D, Poole C, Ledbetter L, Taylor AB. Eccentric training for prevention of hamstring injuries may depend on intervention compliance: a systematic review and meta-analysis. Br J Sports Med. 2015 Mar;49(6):349-56. doi: 10.1136/bjsports-2014-093466.
Hansen TE, Lindhard J. On the maximal work of human muscles especially the flexors of the elbow. The Journal of Physiology 1923;57(5):287-300.
Harøy J, Clarsen B, Wiger EG, Øyen MG, Serner A, Thorborg K, Hölmich P, Andersen TE, Bahr R. The Adductor Strengthening Programme prevents groin problems among male football players: a cluster-randomised controlled trial Br J Sports Med Published Online First. 2018;10. doi: 10.1136/bjsports-2017-098937.
Hayward-Ellis J, Alexander MJL, Glazebrook CM, Leiter J. Ground reaction forces produced by two different hockey skating arm swing techniques. Eur J Sport Sci 2017;17(9):1153-1160.
Hill AV. An instrument for recording the maximum work in muscular contraction. J Physiol 1920;53:88‐90.
Hill AV. The maximum work and mechanical efficiency of human muscles, and their most economical speed. The Journal of Physiology 1922;56:1-2:19–41.
International ice hockey federation 2020. Available at:
https://www.iihf.com/en/associations
Kaartinen S. Effect of skating stride recovery phase on skating speed in ice hockey.
Master´s thesis. University of Eastern Finland 2019.
Krogh A, Skand A. Bicycle ergometer and respiration apparatus for the experimental study of muscle work. Arch. Physiol 1913;33:375-395.
Lignell E, Fransson D, Krustrup P, Mohr M. Analysis of high-intensity skating in top- class ice-hockey match-play in relation to training status and muscle damage. Journal of Strength and Conditioning Research 2018;32(5)/1303–1310
Maroto-Izquierdo S, García-López D, Fernandez-Gonzaloc R, Moreiraa OC, González- Gallegoa J, de Paz JA. Skeletal muscle functional and structural adaptations after eccentric overload flywheel resistance training: a systematic review and meta-analysis. J Sci Med Sport 2017a;20:943–951.
Maroto-Izquierdo S, García-López D, de Paz JA. Functional and Muscle-Size Effects of Flywheel Resistance Training with Eccentric-Overload in Professional Handball Players.
Journal of Human Kinetics volume. 2017b;60:133-143.
Mascaro T, Seaver B, Swanson L. Prediction of Skating Speed with Off-Ice Testing in Professional Hockey Players. JOSPT 1992;15:2:92-98.
Mendez-Villanueva A, Suarez-Arrones L, Rodas G, Fernandez-Gonzalo R, Tesch P, Linnehan R, Kreider R, Di Salvo V. MRI-Based Regional Muscle Use during Hamstring Strengthening Exercises in Elite Soccer Players. PLoS ONE. 2019;11(9):1-15.
Norrbrand L, Tous-Fajardo J, Vargas R, Tesch PA. Quadriceps muscle use in the flywheel and barbell squat. Aviat Space Environ Med 2011;82:13-19.
Núñez Sanchez FJ, Sáez de Villarreal E. Does flywheel paradigm training improve muscle volume and force? A meta-analysis. J Strength Cond Res 2017;31(11):3177–
3186.
Pearsall DJ, Turcotte RA, Murphy SD. Biomechanics of ice hockey. Exercise and Sport Science 2000;675-692.
Petré H, Wernstål F and Mattsson CM. Effects of Flywheel Training on Strength-Related Variables: a Meta-analysis. Sports Medicine – Open 2018;4:55.
Raeder C, Wiewelhove T, Westphal-Martinez MP, Fernandez-Fernandez J, De Paula RA.
Neuromuscular fatique and physiological responses after five dynamic squat exercise protocols. Journal of Strength and Conditioning Research 2016;30:(4)/953–965.
Roczniok R, Stanula A, Maszczyk A, Mostowik A, Kowalczyk M, Fidos-Czuba O, Zając A. Physiological, physical and on-ice performance criteria for selection of elite ice hockey teams. Biol Sport 2016;33:43–48.
Runner AR, Lehnhard RA, Butterfield SA, Tu S, O’neill T. Predictors of speed using off- ice measures of college hockey players. Journal of Strength and Conditioning Research 2015;30(6):1626–1632.
Tesch PA, Fernandez-Gonzalo R, Lundberg TR. Clinical Applications of Iso-Inertial, Eccentric-Overload (YoYoTM) Resistance Exercise. Front. Physiol 2017;8:241.
Tyler TF, Nicholas SJ, Campbell RJ, McHugh MP. The Association of Hip Strength and Flexibility with the Incidence of Adductor Muscle Strains in Professional Ice Hockey Players. The American Journal of Sports Medicine 2001;29:124-128.
Upjohn T, Turcotte T, Persall DJ, Loh J. Three-dimensional kinematics of the lower limbs during forward ice hockey skating, Sports Biomechanics 2008;7:206-221.
Vicens-Bordas J, Estevea E, Fort-Vanmeerhaeghec A, Bandholmd, T, Thorborg K. Is inertial flywheel resistance training superior to gravity-dependent resistance training in improving muscle strength? A systematic review with meta-analyses. J Sci Med 2018;21:75–83.
Vigh-Larsen JF, Beck JH, Daasbjerg A, Knudsen CB, Kvorning T, Overgaard K, Andersen TB, & Mohr M. 2019. Fitness Characteristics of Elite and Subelite Male Ice Hockey Players: A Cross-Sectional Study. Journal of Strength and Conditioning Research 00(00)/1–9.
