Submitted by: S. Rokka, G. Mavridis, E. Bebetsos, K. Mavridis – Department of Physical Education & Sport Science – Democritus University of Thrace, 69100 Komotini
Abstract
The aim of the present study was to evaluate the levels of intensity and direction of the competitive state anxiety in junior handball players prior to a competition and to investigate any possible differences between male and female players, as well as in relation to their athletic experience. The sample of the study consisted of 115 handball players, members of eight handball teams (four male and four female), which participated in the Greek Junior Handball Championships finals held in Athens in 2008. For the data collection, the model used was the Competitive State Anxiety Inventory-II (CSAI-II, Martens, Burton, Vealey, Bump & Smith, 1983; Martens et al., 1990; Jones & Swain, 1992), which was modified for the Greek population by Stavrou, Zervas, Kakkos & Phychoudaki (1998). All players filled in the questionnaire 30 minutes before the competition. The results showed that male junior handball players reported lower scores of cognitive anxiety, which was facilitative to performance. On the other hand, females displayed a higher score in cognitive anxiety, which was rather debilitative to performance. Furthermore, junior male handball players displayed higher self-confidence, with positive effects on their performance, while female handball players stated lower self-confidence, which was neither facilitative nor debilitative to performance. In relation to years of experience, the results revealed that players with four to six years of experience showed higher self-confidence with facilitating direction, while players with less years of experience displayed lower self-confidence, with neither facilitative nor debilitative effects on their performance. In conclusion, the psychological preparation of junior handball players must be taken into serious consideration, during the coaching procedure. Nonetheless, further investigation is needed for the generalisation of the results in Greek handball.
Introduction
It is generally recognized that psychological factors are of crucial importance in high-level competitive sports. The relation between anxiety and performance has been the subject of many thorough researches (Craft, Magyar, Becker & Feltz, 2003; Parfitt & Pates, 1999; Martens, Vealey & Burton, 1990). Cognitive anxiety is characterised by negative concerns and worries about performance, inability to concentrate, and disrupted attention (Krane, 1994). Somatic anxiety consists of an individual’s perceptions, which are characterised by indications such as sweaty palms, butterflies, and shakiness (Martens, Burton, Vealey, Bump & Smith, 1990). Research has also been done on the gender differences concerning state anxiety levels. Self-confidence tends to decrease in females on the day a competitive event is to occur. Male young athletes typically display lower levels of anxiety and higher self-confidence than female athletes (Scanlan & Passer, 1979; Wark &Witting, 1979). Krane and Williams (1994) found no gender differences for cognitive anxiety. They also demonstrated that the more experienced college player would show lower levels of cognitive and somatic anxiety than the less experienced player. As far as handball is concerned, Roguli, Nazor, Srhoj and Bozin (2006) supported that it is a sport, which includes complex and accurate motor skills, and they suggested that psychological factors play an even more decisive role in a competition, differentiating between successful and less successful teams. The purpose of the present study was to evaluate the levels of intensity and direction of the competitive state anxiety in junior handball players prior to a competition and to investigate any possible differences between male and female players, as well as in relation to their athletic experience.
Methods
Participants
The sample of the study consisted of 115 handball players, members of eight handball teams (four male and four female), which participated in the Greek Junior Handball Championships finals held in Athens in 2008. The age of the participants was between 14 and 16 years (M. = 14.85, S.D. = 1.14). The participants voluntarily and anonymously took part in the research, with the consent of their coaches and clubs’ managements, as well as with the parents’ informed consent for the players younger than 14 years of age. For functional needs, 61 of the players were males and 54 females. For the needs of the research, the sample was divided according to athletic experience: (a) up to 3 years (n = 55) and (b) 4 to 6 years (n = 60).
Data collection
For the data collection, the model used was the Competitive State Anxiety Inventory-II (CSAI-II, Martens, Burton, Vealey, Bump and Smith, 1983; Martens et al., 1990; Jones & Swain, 1992), which was modified for the Greek population by Stavrou, Zervas, Kakkos & Phychoudaki (1998). The specific instrument measures cognitive, somatic anxiety and self-confidence, as well as the direction of this state anxiety. The scale consists of 15 items (three 5-item subscales arranged on a 4-point Likert-type scale ranging from 1 (none) to 4 (very much) for intensity. Also, it includes a 7-point Likert-type bipolar scale ranging from –3 (hinders performance) to +3 (facilitates performance), which was used to evaluate intensity symptoms as either debilitative or facilitative. All players filled in the questionnaire just prior to the warm-up phase, approximately 30 minutes before the competition.
Statistics
For the statistical analysis of the data, from the SPSS 11.0 statistical package, the methods used were the Factorial analysis, the Reliability analysis and the one-way ANOVA analysis, which was also used in order to determine whether any of the factors were related to gender (male-female) and athletic experience a) up to 3 years (n= 55), b) 4 to six years (n=60). The level of statistical significance was set at p< .05.
Results
The factor analysis indicated three factors, which interpreted 57.19% of the total fluctuation on the intensity scale and three factors interpreting 61.87% of the direction of this intensity. The Cronbach’s alpha internal cohesion indicator of the questionnaire was satisfactory. The values that came out were .79 for the cognitive anxiety, .81 for the somatic anxiety and .80 for the self-confidence. For the direction of anxiety, the values were .84, .86, and .91 correspondingly (see Table 1). The one-way ANOVA analysis showed statistically important differences concerning cognitive anxiety and self-confidence and its direction, between the male and female players (F1, 114 = 9.78; p < .01, F1, 114 = 30.28; p < .001, F1, 114 = 42.05; p < .001, F1, 114 = 37.07; p < .001). Male players presented lower scores on cognitive anxiety. They also had higher scores on self-confidence and its direction, which facilitated their performance. What is more, there were statistically important differences concerning self-confidence and its direction (F1, 114 =19.09; p<.001, F1, 114 =26.21; p<.001), between players of different years of experience. Players with four to six years of experience reported higher scores on self-confidence and its direction, which facilitated their performance (See Table 1).
Table 1
Descriptive statistics and important differences among the factors of the questionnaire
Handball Players
Athletic Experience
Cronbach’s Alfa
male
female
Up 3 years
4 to six years
Intensity
M. (S.D.)
M. (S.D.)
M. (S.D.)
M. (S.D.)
Cognitive
.79
2.10 (.48)**
2.78 (.57)
2.63 (.68)
2.19 (.55)
Somatic
.81
1.95 (.53)
2.05 (.74)
2.08 (.71)
1.98 (.57)
Self-confidence
.80
3.25 (.52)***
2.63 (.67)
2.69 (.65)
3.20 (.55)***
Direction of intensity
Cognitive
.84
4.26 (.66)***
3.20 (.71)
3.62 (.92)
3.98 (.84)
Somatic
.86
4.12 (.69)
4.06 (.86)
3.98 (.85)
4.16 (.75)
Self-confidence
.91
5.72 (.72)***
4.21 (.93)
4.69 (.62)
5.78 (.57)***
Note 1: Μ = Mean Prices, S.D. = Standard Deviations of factors in relation to the gender and athletic experience Note 2: Significant *** p < .001, ** p < .01, * p < .05.
Discussion/Conclusions
The results of the research showed that male junior handball players reported lower scores of cognitive anxiety, which was facilitative to performance. On the other hand, females displayed a higher score in cognitive anxiety, which was rather debilitative to performance. Furthermore, junior male handball players displayed higher self-confidence, with positive effects on their performance, while females stated lower self-confidence, which was neither facilitative nor debilitative to performance. In relation to years of experience, the results revealed that players with 4 to 6 years of experience showed higher self-confidence with facilitating direction, while players with less years of experience displayed lower self-confidence, with neither facilitative nor debilitative effects on their performance. These results are consistent with the findings of similar studies (Scanlan et al., 1979; Wark et al., 1979) which indicates that male athletes typically display lower levels of anxiety and higher self-confidence than female athletes. The above findings seem to support the existing theories on intensity (Mellalieu, Neil & Hanton, 2006; Parfitt & Pates, 1999; Stavrou, Psychoudaki, Zervaς, 2006; Woodman & Hardy, 2003; Wilson, & Raglin, 1997) which demonstrates that the more experienced player will show lower levels of cognitive and somatic anxiety than the less experienced player.
In conclusion, the psychological preparation of junior handball players must be taken into serious consideration during the coaching procedure. Professional help and programming of the psychological preparation of the athletes and observation of their emotional condition before and during a game is necessary to reduce competitive anxiety and contribute to the high effectiveness of handball players. Nonetheless, further investigation is needed for the generalization of the results in Greek handball.
References
Craft, L.L., Magyar, T.M., Becker, B.J., & Feltz, D. L. (2003). The relation between the competitive state anxiety inventory-ΙΙ and sport performance: A meta-analysis. Journal of Sport and Exercise Psychology, 25, 44-65.
Jones, G., & Swain, A. (1992). Intensity and direction of competitive state anxiety and relationships with competitiveness. Perceptual and Motor Skills, 74, 464-472.
Krane, V., & Williams, J.M. (1994). Cognitive anxiety, somatic anxiety and confidence in track and field athletes: the impact of gender, competitive level and characteristics. International Journal of Sport Psychology, 25, 203-217.
Martens, R., Vealey, R.S. & Burton, D. (1990). Competitive anxiety in sport (pp.117-173). Champaign, IL: Human Kinetics.
Martens, R., Burton, D., Vealey, R. S., Bump, L. A., & Smith, D. E. (1990).
Development and validation of the competitive state anxiety inventory-2 (CSAI-2). In R. Martens, R. S. Vealey, & D. Burton (Eds.), Competitive anxiety in sport (pp. 117-173). Champaign, IL: Human Kinetics.
Mellalieu, S.D., Neil, R., & Hanton, S. (2006). Self-confidence as a medicator of the relationship between competitive anxiety intensity and interpretation. Res Q Exercise Sport, 77(2), 263-270.
Parfitt, G., Hardy, L., & Pates, J. (1995). Somatic anxiety and physiological arousal: Their effects upon a high anaerobic, low memory demand task. International Journal of Sport Psychology, 26, 196-213.
Parfitt, G., & Pates, J. (1999). The effects of cognitive and somatic anxiety and self-confidence on components of performance during competition. Journal Sports Science, 17(5), 351-356.
Rogulj, N., Nazor M., Srhoj, V., & Bozin, D. (2006). Differences between competitively efficient and less efficient junior handball players according to their personality traits. Kinesiology Journal, 2, 158-163.
Scanlan, T.K., & Passer, M.W. (1979). Sources of competitive stress in young female athletes. Journal of Sport Psychology, 1, 248-250.
Stavrou, N.A., Psychountaki, M., Zervas, Y. (2004). Intensity and direction dimensions of competitive state anxiety: a time-to-event approach. Perceptual Motor Skills, 98(2), 439-449.
Stavrou, N.A., Zervas, Y., Kakkos & Phychoudaki, M. (1998). Intensity and direction of competitive state anxiety. Proceedings of the 2nd International /5th Panhellenic Congress of Athletic Psychology (pp. 139-141), Trikala.
Stavrou, N.A., Psychoudaki, M., Zevraς, Y. (2006). Intensity and direction dimensions of competitive state anxiety: a time-to-event approach. Laboratory of Motor Behavior and Sport Phycology, Department of Physical Education and Sport Science, University of Athens, 103(1), 91-98.
Wark, K.A., & Witting, A.F. (1979). Sex role and sport competition anxiety. Journal of Sport Psychology, 1, 248-250.
Woodman, T., & Hardy L. (2003). The relative impact of cognitive anxiety and self-confidence upon sport performance: A meta-analysis. Journal Sports Sciences, 21(6), 443-457.
Wilson, G.S., & Raglin, J.S. (1997). Optimal and predicted anxiety in 9-12 year old track and field athletes. Scadinavian Journal of Medicine and Science in Sports, 2, 148-152.
Submitted by: A. Bosak – Dept. of Physical Education & Kinesiology, Brock University, P. Bishop – Dept. of Kinesiology, University of Alabama, J. Green – Dept. of Health, PE, and Recreation, University of North Alabama and G. Hawver – Dept of Health and Human Performance, Georgia Southwestern State University
Abstract
Much effort over the past 50 years has been devoted to research on training, but little is known about recovery after intense running efforts. Insufficient recovery impedes training and performance. Anecdotal evidence suggests that cold water immersion immediately following intense distance running efforts aids in next day performance perhaps by decreasing injury or increasing recovery. The purpose of this study was to compare 5 km racing performance after 24 hrs with and without cold water immersion. Twelve well-trained runners (9 males, 3 females) completed successive (within 24 hours) 5 km performance trials on two separate occasions. Immediately following the first baseline 5 km trial, runners were treated with ice water immersion for 12 minutes followed by 24 hrs of passive recovery (ICE). Another session involved two 5 km time trials: a baseline trial and another trial after 24 hrs of passive recovery (CON). Treatments occurred in a counterbalanced order and were separated by 6-7 days of normal training. ICE (20:08 ± 2.0 min) was not significantly different (p = 0.09) from baseline (19:59 ± 2.0 min). CON (19:59 ± 1.9 min) was significantly (p = 0.03) slower than baseline (19:49 ± 1.9 min). ICE heart rate (175.3 ± 7.6 b/min) was significantly (p = 0.02) less than baseline (178.3 ± 9.8 b/min), yet CON heart rate (177.3 ± 6.3 b/min) was the same as baseline (177.3 ± 7.3 b/min). ICE rate of perceived exertion (19.2 + 1.0) was significantly less (p = 0.03) than baseline (19.8 ± 0.5) while CON rate of perceived exertion (19.5 ± 0.8) was not significantly different (p = 0.39) from baseline (19.6 ± 0.8). Seven individuals responded negatively to ICE running a mean 24.0 ± 13.9 seconds slower than baseline. Nine individuals responded negatively to CON by running a mean 17.4 ± 12.1 seconds slower than baseline. Three individuals responded positively to ICE running a mean 20.33 ± 6.7 seconds faster during second day performance. Three individuals responded positively to CON by running a mean 13.3 ± 6.8 seconds faster than baseline. In general, cold water immersion minutely reduced the decline of next day performance, yet individual variability existed. Efficacy of longer durations of cold water immersion impact after 48 hrs and on distances greater than 5 km appear to be individual and need to be further explored.
Key words: cryotherapy, ice water immersion, passive recovery, running
Introduction
Recovery from hard running efforts plays a vital role in determining when a runner can run at an intense level again (Fitzgerald, 2007). Hard training, followed by adequate recovery, allows the body to adapt to the unusual stress and become better accustomed and more prepared for the same stress, should it occur again (Fitzgerald, 2007; Sinclair, Olgesby, & Piepenberg, 2003). Balancing hard efforts with periods of rest is essential in improving performance during endurance efforts.
The recovery process from endurance efforts tends to revolve around repairing damaged muscle fibers and replenishing glycogen stores (Gomez et al., 2002; Nicholas et al., 1997). Methods proposed to enhance recovery, such as cold water immersion, potentially decrease swelling and the severity of delayed onset of muscle soreness (DOMS), which possibly benefits endurance (i.e. running) and anaerobic performance (Higdon, 1998; Vaile, Gill, & Blazevich, 2007).
Cold water immersion is a common practice among collegiate and professional athletes following intense physical efforts. Anecdotal evidence from several National Athletic Trainers’ Association (NATA) collegiate head athletic trainers suggests that cooling the legs after a hard training effort may benefit the next day’s performance. Popular running and athletic magazines (e.g., Runner’s World, Running Times, etc.) have continually suggested that applying cold water to the legs of a runner facilitates a better perceived feeling for the next run on the following day. Yet, despite its widespread use there is no scientific data supporting the notion that cooling the legs after a hard distance running effort will improve performance 24 hrs later.
The use of cold as a treatment is as ancient as the practice of medicine, dating back to Hippocrates (Stamford, 1996). The therapeutic use of cold is the most commonly used modality in the acute management of musculoskeletal injuries. Running is a catabolic process, with eccentric muscle contractions leading to muscle damage. Applying cold to an injured site decreases pain sensation, improves the metabolic rate of tissue, and allows uninjured tissue to survive a post-injury period of ischemia, or perhaps allows the tissue to be protected from the damaging enzymatic reactions that may accompany injury (Arnheim and Prentice, 1999; Merrick, Jutte, & Smith, 2003). The use of cryotherapy, between sets of “pulley exercises” (similar to a seated pulley row), decreased the feelings of fatigue of the arm and shoulder muscles of 10 male weight lifters (Verducci, 2000), while other cryotherapy research involving recovery from intense anaerobic efforts has yielded equivocal results (Barnett, 2006; Cheung, Hume, & Maxwell, 2003; Crowe, O’Connor, & Rudd, 2007; Howatson, Gaze, & Van Someren, 2005; Howatson and Van Someren, 2003; Isabell et al., 1992; Paddon-Jones and Quigley, 1997; Sellwood et al., 2007; Vaile, Gill, & Blazevich, 2007; Vaile et al., 2008; Yackzan, Adams, and Francis, 1984). However, methods of cryotherapy effective for enhancing recovery from distance running efforts have not been examined.
Long duration or high intensity running contributes to muscle cell damage (Fitzgerald, 2007; Noakes, 2003). Edema, a by-product of muscle damage can cause reduced range of joint motion. Because cryotherapy has been shown to decrease inflammation (Dolan et al., 1997; O’Conner and Wilder, 2001), it is logical to assume that this treatment may reduce the severity of DOMS. Less pain may permit an athlete to push themselves harder potentially improving performance. Despite the fact that previous research has shown that 24 hrs alone is not sufficient recovery from 5 km running performance (Bosak, Bishop, & Green, 2008), it might be possible that combining cold water immersion with 24 hrs of recovery could potentially hasten the recovery process. Therefore, the purpose of this study was to compare 5 km racing performance after 24 hrs of passive recovery with and without cold water immersion.
Methods
Participants:
Participants for the study were 12 well trained male (n = 9) and female (n = 3) runners currently engaged in rigorous training. Runners from the local road running and track club, local triathlon competitors, as well as former competitive high school and college runners, were recruited by word of mouth. Participant inclusion criteria included the following: 1) Subjects must have been currently involved in a distance running training program; 2) Their 5 km times previously run had to be at least 16-22 min for male runners or 18-24 min for female runners; 3) They had to be currently averaging at least 20-30 miles (running) per week; 4) They had to have previously completed at least five 5 km road or track races; 5) They had to have a VO2max of at least 45 ml/kg/min (females) or 55 ml/kg/min (males); and 6) They had to provide sufficient data (from running history questionnaires, physical activity readiness questionnaires, and health readiness questionnaires) that reflected good health.
Participants completed a short questionnaire regarding their running background, racing history, and current training mileage. All participants were volunteers and signed a written informed consent outlining requirements as well as potential risks and benefits resulting from participating.
Procedures:
Participants were assessed for age, height, body weight, and body fat percentage using a 3-site skinfold technique (Brozek and Hanschel, 1961; Pollock, Schmidt, & Jackson, 1980). Participants were fitted with a Polar heart rate monitor, and then completed a graded exercise test (GXT) to exhaustion lasting approximately 12-18 min. VO2max, heart rate (HR), and ratings of perceived exertion (RPE) were collected every minute.
All GXTs were completed on a Quinton 640 motorized treadmill. The test began with a 2 min warm-up at 2.5 mph. Speed was increased to 5 mph for 2 min, followed by 2 min at 6 mph, 2 min at 7 mph, and 2 min at 7.5 mph. At this point, incline was increased two percent every 2 min thereafter until the participant reached volitional exhaustion (i.e. they felt like they could no longer continue running at the required speed and grade). Once the participant reached volitional exhaustion, they were instructed to cool down until they felt recovered.
Approximately five days later, participants performed their first 5 km race (performance trial) between the hours of 6:30 am to 7:30 am. The time of day for each performance trial was consistent throughout the entire study. All performance trials were completed on a flat hard-surfaced 0.73 mile loop. Prior to each trial, participants completed visual analog scales, before and after a 1.5 mile warm-up run, regarding their feelings of fatigue and soreness within local muscle groups (quadriceps, hamstrings, gastrocnemius), and for lower and total body muscle groups. Visual analog scales were 15 cm lines, where participants placed an “X” on the line indicating their feelings (with 0 = no fatigue or soreness and 15 = extreme fatigue or soreness). The focus of the visual analog scales was to determine if participants felt the same before the start of every time trial. Participants were also required to rate their perceived exertion (RPE) after the warm-up and prior to the start of each 5 km, during each trial, and at the end of each performance trial to determine if feelings of effort remained consistent between each trial, as well as during each lap and at the end of each trial.
Runners underwent a 1.5 mile warm-up prior to every 5 km performance trial (Kaufmann and Ware, 1977). Participants completed four 5 km performance trials within nine days. Two 5 km performance trials (baseline and CON) were separated by 24 hrs of passive recovery. Passive recovery was deemed as no exercise or extensive physical activity during the allotted recovery hours. Two 5 km performance trials (baseline and ICE) were also separated by 24 hrs of passive recovery, but with 12 minutes of 15.5ºC water immersion immediately following the baseline trial. The two sessions of 5 km performance trials were counterbalanced and were separated by 6-7 days of normal training. Each trial session therefore, had a separate baseline preceded by 24 hrs of passive recovery.
Ideal cryotherapeutic water temperature has not been determined, yet various head collegiate athletic trainers prefer that the water temperature does not dip below 13ºC (55.5ºF) since many people find water temperatures below 13ºC uncomfortable (O’Connor and Wilder, 2001). Also, the duration of ice baths generally lasts 10-15 minutes and is usually applied immediately after a hard training session (Crowe, O’Connor, & Rudd, 2007; Schniepp et al., 2002; Vaile et al., 2008). Hence, in this study, 15.5ºC (60ºF) was the temperature for the cold water and the athletes were immersed for 12 min.
During each time trial, average heart rate and ending RPE were recorded in order to determine if effort for each 5 km was consistent. All participants competed with runners of similar ability to simulate race day and hard training conditions, while verbal encouragement was provided often and equally to each participant. At the end of every performance trial, each runner was instructed to complete a low intensity 1.5 mile cool-down. Each total testing trial required approximately 60 min.
Statistical Analysis:
Basic descriptive statistics were computed. Repeated measures of analysis of variance (ANOVA) were employed for making comparisons between CON and baseline and PAS and baseline performance trials for the following variables: finishing times, HR, RPE, and fatigue or soreness responses. All statistical comparisons were made at an a priori p < .05 level of significance. Data were expressed as group mean + standard deviation and individual results.
In order to evaluate individual responses, data from each participant’s first run was compared to the second run using a paired T-test. The least significance group mean difference (p < 0.05) was determined and group mean finishing time was adjusted to determine the amount of change in seconds needed for significance to occur. The time change between the first trial run and the adjusted trial run baseline was divided by the first trial run and expressed as mean number of seconds or percent for both the ICE (9.3 seconds or 0.8%) and CON (9.5 seconds or 0.8%) trials. The percent values were applied to each individual baseline time in order to determine how many seconds (positive or negative) the second performance trial time had to be over or under the first performance trial, in both CON and ICE conditions, to quantify as a response. Participants were then labeled as non-responders, positive-responders (faster after treatment), and negative-responders (slower after treatment).
Results
Descriptive characteristics are found in Table 1. The participants were between the ages of 18 and 35 (the majority of subjects were between ages 20-28) years. All participants were trained runners or triathletes (where running was their specialty event).
Mean finishing times, HR, and RPE for CON and ICE trials are found in Table 2. CON was significantly (p = 0.03) slower (10 seconds) than baseline, where as ICE was not significantly different (p = 0.09) from baseline. No significant differences were found between CON HR vs. baseline, but ICE HR was significantly (p = 0.01) less than baseline. No significant differences (p = 0.39) were found between CON RPE and baseline, yet ICE RPE was significantly (p = 0.03) less than baseline.
Figure 1 shows individual changes in finishing times for all CON and ICE performance trials. To be considered a non-responder, the individual time change had to fall within 0.8% of baseline performance for ICE and CON. Positive and negative responders (Table 3) were identified when individual time change was greater than 0.8% for CON and ICE trials, with a positive responder being one whose second performance trial time improved (expressed as a negative value) and a negative responder being one whose second performance trial time slowed (expressed as a positive value).
Seven individuals responded negatively to ICE by running a mean 24.0 ± 13.9 seconds slower during the second trial (Table 3). Three individuals responded positively to ICE by running a mean 20.3 ± 6.7 seconds faster than baseline. Two individuals were considered non-responders to ICE with a mean time change of 2.5 ± 0.7secs.
Seven individuals responded negatively to CON by running a mean 20.6 ± 9.0 seconds slower than baseline (Table 3). Three individuals responded positively to CON by running a mean 13.3 ± 6.8 seconds faster than baseline. Two individuals were non-responders to the CON trials with a mean time change of 6.5 ± 0.7 seconds. It is important to note that the seven individuals who were negative responders to ICE were not the same seven participants who responded negatively to CON. Also, the three participants who responded positively to ICE were not the same three individuals who responded positively to CON. Finally, the non-responders to ICE were not the same non-responders to CON.
Soreness and fatigue scores (Table 4) on the pre-and post-warm-up fatigue or soreness visual analog scales were not significantly different between CON and baseline versus ICE and baseline.
Discussion
The effects of cold-water immersion on recovery and next day performance in 5 km racing have not been previously evaluated. Therefore, the primary purpose of this study was to compare 5 km running performance after 24 hrs of passive recovery with and without cold water immersion. This study appeared to indicate that cold water immersion does not dramatically help performance (regarding the group of runners as a whole) during second day 5 km trials.
Twenty-four hours of passive recovery may allow for normalization of muscle and liver glycogen, yet muscle function and performance measures may not be fully recovered (Foss and Keteyian, 1998). Hence, 24 hrs of recovery, by itself, may not be sufficient to allow for a return to optimal performance (Bosak, Bishop, & Green, 2008). When racing (e.g., a 5 km distance) on consecutive days, race times may be slower on the second day due to magnified perception of pain and impaired muscle function associated with DOMS (Brown and Henderson, 2002; Fitzgerald, 2007; Galloway, 1984). Since cold water immersion may speed up the recovery process (Arnheim and Prentice, 1999; Vaile et al., 2008) it is logical to assume that cold water immersion immediately after a 5 km race or workout could attenuate soreness potentially minimizing performance decrements on successive days.
There were no significant (p = 0.09) differences in 5 km performance between ICE and baseline, indicating that mean performance during ICE was not significantly slower (9 seconds) than baseline (refer to Table 2). However, CON performance was significantly (p = 0.03) slower (10 seconds) than baseline. Hence, due to significant differences occurring between ICE and baseline, it appears that cold water immersion slightly attenuated the rate of decline on successive 5 km time trial performance. However, the time difference between CON and baseline versus ICE and baseline was a mere second. Therefore, from a practical standpoint, cold water immersion was no more beneficial than CON on successive 5 km performance.
Despite the minimal differences between CON (10 seconds) and ICE (9 seconds) trials regarding mean time change, it is important to focus on the effects of cold water immersion on individual runners (Figure 1). Because some runners ran slower during successive performance trials while other runners ran faster, the mean finishing times do not necessarily give a true impression of the benefits or liabilities of the specific treatments involved in this study. As it is with most ergogenic aids, individual variability suggests what works (e.g., ice) for one person may not work the same for another person. It is possible that the treatment may often not have an effect at all, as similar to what occurred with several prior anaerobic performance studies (Barnett, 2006; Cheung, Hume, & Maxwell, 2003; Crowe, O’Connor, & Rudd, 2007; Howatson, Gaze, & Van Someren, 2005; Howatson and Van Someren, 2003; Isabell et al., 1992; Paddon-Jones and Quigley, 1997; Sellwood et al., 2007; Vaile et al., 2008), which was also the case in this study as two individuals were considered non-responders to ICE with a mean time change of 2.5 ± 0.7 seconds between ICE and baseline, while two other participants were non-responders to CON with a mean time change of 6.5 ± 0.7 seconds between CON and baseline.
Three individuals responded positively (Table 3) to ICE, running a mean 20.33 ± 6.7 seconds faster, indicating that cold water immersion may have actually allowed these individuals to run faster on the second day. However, 3 different individuals responded positively to CON, running a mean 13.3 ± 6.8 seconds faster than baseline. The mechanism by which cold water immersion aids in recovery, from endurance performance, remains somewhat unclear and equivocal (Schniepp et al., 2002; Vaile et al., 2008). Yet, several runners who did run faster during ICE trial, verbally indicated that prior to the second trial, their legs felt better (regarding fatigue and soreness) than they had prior to CON. Thus, the notion of feeling better may have allowed the runners to perform faster.
Seven individuals responded negatively (Table 3) to ICE, running a mean 24.0 ± 13.9 seconds slower. However, they were not the same seven individuals who responded negatively to CON, who ran an average of 20.6 ± 9.0 seconds slower than baseline. As was the case with Schniepp et al. (2002) endurance cycling recovery study and various anaerobic performance studies (Crowe, O’Connor, & Rudd, 2007; Sellwood et al., 2002; Vaile et al., 2008; Yackzan, Adams, & Francis, 1984), it appears ICE may have had a more negative effect, for these individuals, on second day performance compared to CON.
Three individuals responded positively to CON running a mean 13.3 ± 6.8 seconds faster during the second day performance trial. It is unclear why some participants ran faster during CON. There were no consistent patterns of HR and increased or decreased performance with all participants during all CON and ICE trials. As a group, no significant differences were found between CON vs. baseline, regarding HR (p = 1.00) and RPE (p = 0.39), despite significant differences (p = 0.04) occurring in mean finishing time. However, mean finishing times for ICE were similar, yet significant differences were found between ICE vs. baseline for both HR (p = 0.01) and RPE (p = 0.03). Hence, there does not appear to be a consistent pattern between performance times and HR and/or RPE.
It can be assumed that a lower HR may be associated with slower times, since HR and intensity levels tend to be linearly related. However, only participants 1, 5, and 6 consistently ran slower during both CON and ICE second day performances with lower HR during both trials. During the ICE trials, only participants 1, 5, 6, and 9 ran slower and had a lower HR. During the CON trials, only 1, 3, 5, 6, ran slower and had a lower HR. Also, soreness and fatigue scores (Table 4) on the pre and post warm-up fatigue or soreness visual analog scales were not significantly different between CON and baseline versus ICE and baseline. These results indicate that all runners tended to feel the same prior to each second day 5 km trial. Therefore, since inconsistencies exist between HR and performance trials and no significant differences were found regarding RPE and fatigue or soreness visual analog scales, it is assumed that each participant completed each trial with similar effort.
Conclusion
The current findings of this study suggest that cold water immersion does not sufficiently enhance recovery (specifically regarding the group of runners as a whole). However, three runners benefited from cold water immersion. Hence, what works for one person may not work for another person. Thus, it may be beneficial for runners to undergo this protocol in order to see which type of recovery method improves their recovery process. Secondly, the results of the study may give credence to some runners’ perception of feeling better due to cold water immersion after a hard running effort. However, one should remember that individual variability existed in response to treatment (ice immersion) within the current study. Future research is needed to see if a greater length of time or slightly lower water temperature in cold water immersion will decrease the rate of decline more or if the effects of cold water immersion are even more predominant on second day performance of distances greater than 5 km.
References
Arnheim, D. D., & Prentice, W. E. (1999). Essentials of athletic training (4th ed.). Boston, MA: McGraw-Hill.
Barnett, A. (2006). Using recovery modalities between training sessions in elite athletes: Does it help? Sports Medicine, 36 (9), 781-796.
Bosak, A., Bishop, P., & Green, M. (2008). Comparison of 5km racing performance after 24 and 72 hours of passive recovery. International Journal of Coaching Science (In Review).
Brown, R. L., & Henderson, J. (2002). Fitness Running (2nd ed.). Champaign, IL: Human Kinetics.
Brozek, J., & Hanschel, A. (1961). Techniques for measuring body composition. Washington, DC: National Academy of Sciences.
Cheung, K., Hume, P., & Maxwell, L. (2003). Delayed onset muscle soreness: treatment strategies and performance factors. Sports Medicine, 33 (2), 145-164.
Crowe, M. J., O’Connor, D., & Rudd, D. (2007). Cold water recovery reduces anaerobic performance. International Journal of Sports Medicine, 28 (12), 994-998.
Dolan, M. G., Thorton, R. M., Fish, D. R., & Mendel, F. C. (1997). Effects of cold water immersion on edema formation after blunt injury to the hind limbs of rats. The Journal of Athletic Training, 32, 233-238.
Fitzgerald, M. (2007). Brain Training for Runners. New York, NY: Penguin Group.
Foss, M. L., & Keteyian, S. J. (1998). Fox’s Physiological Basis for Exercise and Sport. Ann Arbor, MI: McGraw-Hill.
Galloway, J. (1984). Galloway’s Book on Running. Bolinas, CA: Shelter Publications.
Gomez, A. L., Radzwich, R. J., Denegar, C. R., Volek, J. S., Rubin, M. R., Bush, J.A., Doan, B.K., et.al. (2002). The effects of a 10-kilometer run on muscle strength and power. Journal of Strength and Conditioning Research, 16, 184-191.
Higdon, H. (1998). Smart Running. Emmaus, PA: Rodale Press Inc.
Howatson, G., Gaze, D., & Van Someren, K. A. (2005). The efficacy of ice massage in the treatment of exercise-induced muscle damage. The Scandinavian Journal of Medicine and Science in Sports, 2005, 15 (6), 416-422.
Howatson, G. & Van Someren, K. A. (2003). Ice massage: Effects on exercise-induced muscle damage. The Journal of Sports Medicine and Physical Fitness, 43 (4), 500-505.
Isabell, W. K., Durrant, E., Myrer, W., & Anderson, S. (1992). The effects of ice massage, ice massage with exercise, and exercise on the prevention and treatment of Delayed Onset Muscle Soreness. The Journal of Athletic Training, 27 (3), 208-217.
Kaufmann, D. A. & Ware, W. B. (1977). Effect of warm-up and recovery techniques on repeated running endurance. The Research Quarterly, 2, 328-332.
Merrick, M. A., Jutte, L. S., & Smith, M. E. (2003). Cold modalities with different thermodynamic properties produce different surface and intramuscular temperatures. Journal of Athletic Training, 38, 28-35.
Nicholas, C. W., Green, P. A., Hawkins, R. D., & Williams, C. (1997). Carbohydrate intake and recovery of intermittent running capacity. International Journal of Sport Nutrition, 7, 251-260.
Noakes, T. (2003). Lore of Running (4th ed.). Champaign, IL: Human Kinetics.
O’Conner, F. G., & Wilder, R. P. (2001). Textbook of Running Medicine. New York, NY: McGraw-Hill.
Paddon-Jones, D. J., & Quigley, B. M. (1997). Effects of cryotherapy on muscle soreness and strength following eccentric exercise. The International Journal of Sports Medicine, 18 (8), 588-593.
Pollock, M. L., Schmidt, D. H., & Jackson, A. S. (1980). Measurement of cardiorespiratory fitness and body composition in the clinical setting. Comprehensive Therapy, 6, 12-27.
Schniepp, J., Campbell, T. S., Powell, K. L., & Pincivero, D. M. (2002). The effects of cold water immersion on power output and heart rate in elite cyclists. Journal of Strength and Conditioning Research, 16 (4), 561-566.
Sellwood, K. L., Bruker, P., Williams, D., Nicol, A., & Hinman, R. (2007). Ice-water immersion and delayed-onset muscle soreness: a randomized controlled trial. British Journal of Sports Medicine, 41 (6), 392-397.
Sinclair, J., Olgesby, K., & Piepenburg, C. (2003). Training to Achieve Peak Running Performance. Boulder, CO: Road Runner Sports Inc.
Stamford, B., Giving injuries the cold treatment. (1996). The Physician and Sports Medicine, 23, 1-4.
Vaile, J., Gill, N. D., & Blazevich, A. J. (2007). The effect of contrast water therapy on symptoms of delayed onset of muscle soreness. Journal of Strength and Conditioning Research, 21 (3), 697-702.
Vaile, J., Halson, S., Gill, N., & Dawson, B. (2008). Effect of hydrotherapy on recovery from fatigue. International Journal of Sports Medicine, 29 (7), 5:39-544.
Vaile, J., Halson, S., Gill, N., & Dawson, B. (2008). Effect of hydrotherapy on the signs and symptoms of delayed onset muscle soreness. European Journal of Applied Physiology, 102(4), 447-455.
Verducci, F. M. (2000). Interval cryotherapy decreases fatigue during repeated weight lifting. The Journal of Athletic Training, 35, 422-426.
Yackzan, L., Adams, C., & Francis, K. T. (1984). The effects of ice massage on delayed muscle soreness. The American Journal of Sports Medicine, 12 (2), 159-165.
Submitted by: Linda Garza, MS – Purdue University and Sally J. Ford, PhD – Texas Woman’s University
Abstract
An intervention strategy was developed, implemented, and evaluated that aimed at minimizing performance anxiety. The goal was to guide NCAA Division I softball athletes in using a breathing technique that, by contributing to the management of performance anxiety, would help each athlete reach full potential on the softball field. The strategy focused on the effects of the breathing technique on the participants’ heart rates, in relation to daily anxiety events; a heart rate monitor and anxiety logs were used to obtain data. All 4 of the athletes studied indicated improvement at various stages in the program. (more…)
Submitted by: Jefrey L. Frost – United States Sports Academy
Abstract
Identifying particular characteristics (qualities and abilities) of successful sports coaches could offer other coaches help in improving their performance. Toward this end, 15 high school coaches completed a survey on 17 possible such characteristics, ranking 5 of them above the rest (≥ 90th percentile): quality of practice, communicating with athletes, motivating athletes, developing athletes’ sports skills, and possessing knowledge of the sport. Coaches seeking to enhance their success might focus on these characteristics.
Submitted by: David J. Laliberte, MSS, MA – Minnesota Hockey Coaches Association
Abstract
Cutting-edge technologies and space-age synthetics are dramatically recreating ice hockey sticks today. But how does current scholarship view these high-priced innovations, particularly during performance of the slap shot, hockey’s most explosive maneuver? This literature review on both slap shot biomechanics and technological developments in ice hockey sticks suggests that player technique and strength exert much greater influence on slap shot puck velocity than does stick composition. Moreover, this study illuminates how stick flexibility, rather than composition, should be the key mechanical consideration in stick selection, since highly flexible sticks can enhance both stick deflection and strain energy storage, two important variables in the velocity of slap shots.
Biomechanics of Ice Hockey Slap Shots: Which Stick Is Best?
At its historical core, hockey is a game rooted in the natural environment. First played on the frozen lakes and rivers of upper North America, ice hockey—begun as the Native American game of shinny—featured carved wooden poles as sticks and hand-sewn fabrics as balls (Oxendine, 1988). As Europeans took up the game, they applied their technologies to this traditional equipment, gradually yet substantially changing the hockey stick by constructing it out of multiple pieces of wood, curving the stick blade, and wrapping the stick in fiberglass and laminate plastics to increase its durability and performance (Pearsall, Montgomery, Rothsching, & Turcotte, 1999).
Now, however, burgeoning technologies are virtually recreating hockey sticks with each passing day. Wood sticks, once the paragon of the sport, have largely been replaced by high-tech—and high-priced—graphite and composite models. Because of the seeming popularity of these “one-piece” composite sticks amongst professional players, hordes of youth and high-school-age hockey participants are now outfitting themselves with these technological marvels, much to the delight of proliferating hockey equipment companies. Certainly, the need for scholarly research on hockey technology has never been greater: Thousands of participants in the sport stand to benefit from a deeper understanding of the new developments in hockey stick technology.
This paper, then, provides a scholarly education on hockey sticks, both by analyzing the biomechanics of ice hockey shooting and by investigating the extant literature on hockey stick research. In particular, this essay explores the implications of stick technologies and biomechanics for the hockey slap shot, presenting the stick selections and key bodily mechanics that stand to enhance performance of this complex and critical hockey skill.
Slap Shot Mechanics
The Slap Shot’s Six Phases
A variety of scholars have explored the biomechanical aspects of ice hockey, with studies centering primarily around skating (Bracko, 2004; De Koning & Van Ingen Schenau, 2000) and shooting (Doré & Roy, 1978; Hache, 2002; Pearsall, Turcotte, & Murphy, 2000; Roy & Doré, 1976). Of these, several studies have analyzed the mechanics involved in various types of hockey shots, including the wrist, snap, slap, and backhand shots, performed both while stationary and when skating (Carr, 2004, p. 42; Doré & Roy, 1976, 1978; Hache, 2002, p. 84; Alexander, 1964, cited in Pearsall et al., 2000, p. 689; Cotton, 1966, cited in Pearsall et al., 2000, p. 689; Furlong, 1968, cited in Pearsall et al., 2000, p. 689). The slap shot in particular has garnered much scholarly attention, with researchers dividing the shot into six distinct phases: backswing, downswing, preloading, loading, release, and follow-through (Pearsall et al., 1999; Villasenor, Turcotte, & Pearsall, 2006). Three of the six—the preloading, loading, and release phases—concern the mechanical behaviors exhibited by the stick after its contact with the ice surface. This blade-ice contact time has been the intense focus of the majority of researchers investigating the hockey slap shot.
Blade Orientation
Past studies have uncovered several key differences between elite and novice performers of this critical blade-ice contact portion of the slap shot. For example, researchers have cited the orientation of the stick blade during its contact with the ice as an element differentiating elite from recreational performers. For instance, in their study of 15 college-age hockey players, Lomond, Turcotte, and Pearsall (2007) reported that experts tended to demonstrate a unique blade orientation whereby on contact with the ice, the stick blade was tilted forward (or cupped) more than recreational players’ sticks. In addition, Lomond et al. described a distinctive “rocker” component between the loading and release phases of the shot, during which the cupped stick blade almost instantaneously tilted perpendicular to the ice, infusing the puck with additional kinetic energy generated from the slight recoil of the stick blade itself. The authors noted this “rocker” component in the slap shot execution of all subjects in their study, both elite and recreational; blade “rocker” would seem, then, to be a component of slap shots in general. The Lomond et al. report does, however, emphasize the importance of the more tilted blade orientation demonstrated by expert players, a finding corroborated by greater puck velocities during their slap shots (Lomond et al., 2007).
Hand Position
In addition, researchers have cited player hand position as a distinguishing factor in expert slap shot performance. Wu and colleagues, studying male and female collegiate hockey players, noted that a lowered bottom hand, even past the midpoint of the shaft, generated additional stick bend and thus more strain energy, resulting in greater puck velocities (Wu et al., 2003); work of Canadian physicist and hockey enthusiast Alain Hache has seconded these mechanical benefits (Hache, 2002, p. 88). Thus, while it remains unquantified for now, some contribution to force generation in the hockey slap shot seems to result from a low bottom-hand grip on the stick, even past the shaft midpoint.
Impulse Duration
Beyond blade orientation and hand position, two additional factors likely play considerable roles in determining slap shot velocity. The first of these significant contributors is impulse duration, or the force applied to an object over time, the elongation of which increases the transfer of force to an object (Carr, 2004, p. 38). Carr cites the “whiplike” effect of a kinetic chain—a progressive increase in velocity from the most massive to the least massive body parts—as one key technique that allows for a lengthened application of impulse which imparts greater force to the struck object (2004, p. 39). Hockey players employ this “whiplike” technique in a slap shot by rotating the torso, the shoulders, the biceps, and the forearms in sequence, elongating the duration of stick blade contact with the puck. This extended impulse duration has been noted as a primary factor in heightened velocities of hockey slap, wrist, and backhand shots (Roy & Doré, 1976).
Further, Villasenor, Turcotte, and Pearsall (2006) found that among 20- to 30-year-old male slap shot performers, both expert and recreational, the longer the blade contacted the puck, the greater the final puck velocity. Moreover, all elite players in the study demonstrated longer blade-puck contact time than their nonelite counterparts (an average 38 ms for elite players vs. an average 27 ms for nonelite players), corresponding to substantially greater slap shot velocities for experts than for novices (averaging 120 km/h for elite players vs. 80.3 km/h for nonelite players) (Villasenor et al., 2006). Clearly, extending the blade’s contact time with the puck provides an advantage for players seeking greater slap shot velocity.
Stick Bending
A final (and perhaps most important) area contributing to the speed of slap shots is the bending of the stick’s shaft, which begins when the stick blade contacts the ice and lasts through the recoil of the stick just before a player’s follow-through. Hockey scientists David Pearsall, Rene Turcotte, and Stephen Murphy have gone so far as to attribute 40% to 50% of final slap shot velocity to the amount of deflection, or bending, in the stick shaft (Pearsall et al., 2000, p. 690), and photographs in Alain Hache’s Physics of Hockey attest to the considerable stick bend generated by contemporary National Hockey League players (Hache, 2002).
In exploring the stick-bending phenomenon, Villasenor et al. (2006) determined that several crucial relationships exist between stick bending and increased slap shot velocities. First, they noted that elite hockey performers initiated stick bending at the instant of, or shortly before, first contact with the puck, whereas recreational players commenced stick bending after contacting the puck and fully halfway through their stick blade’s contact time with the ice. Expert players also spent a greater percentage (28.8%) of the ice-stick blade contact window bending the stick, in comparison to their nonexpert counterparts (18.2%). Finally, elite performers employed a lower “kick point”—or area of maximum deflection—along the stick shaft than less skilled players did, which has spurred current hockey stick companies to engineer composite sticks designed to lower this spot of maximum bend (Hache, 2002, p. 95). Overall, Villasenor et al. describe a “strong relationship” between final puck velocity and maximum angle of stick deflection, underlining the importance to hockey athletes of initiating considerable stick bend during their slap shots (Villasenor et al., 2006). Alongside blade orientation, hand position, and impulse duration, stick bending contributes to the multiplicity of mechanical factors generated by the player during the performance of this most forceful of hockey skills.
Stick Composition
Beyond each hockey player’s individual slap shot technique, an additional facet of the shot remains variable: the stick. With the onslaught of new hockey technologies over the past decade, no shortage of stick options exists. Whereas hockey sticks were once constructed almost exclusively out of Rock elm, then in the 1990s from aluminum for the shaft and wood for the blade, 21st-century trends now incorporate space-age composite materials like graphite, Kevlar, and carbon in hockey stick design (Sports Materials, 2005; Hache, 2002; Marino, 1998; Pearsall et al., 1999; Wu et al., 2003). Technological advancement, however, has not come without cost, both in monetary terms (most composite sticks retail for at least $100, compared to $40 for a wood stick) and in reduced sensitivity for puckhandling (“feel”) attributed to composite sticks (Barpanda, 1998; Hache, 2002, p. 94; Hove, 2004; Marino, 1998). Nevertheless, today’s hockey players largely face three distinct stick options: an all-wood stick, a stick with a composite shaft and wood blade, or a fully composite stick. The remainder of this paper explores mechanical differences that can be discerned among these construction types during the performance of hockey slap shots.
Stick Construction Materials’ Role in Shot Velocity
Key to enhancing slap shot velocity is maximizing strain energy stored in and released from the hockey stick. Indeed, the current revolutions in hockey stick materials are efforts to capitalize on this mechanical principle. Several scholars have recently studied the effect of hockey stick composition on slap shot velocities, yielding intriguing and somewhat unexpected results. In a study of wood, graphite, and aluminum stick constructions and their role in slap shot velocity, for instance, Wu et al. found that puck velocity was influenced not by stick type but by player skill level and overall body strength. Although the authors reported stick bend to be a key factor in force generation during a slap shot, they attributed any significant differences in stick bend (and therefore puck speed) to the athlete’s bottom hand placement rather than to differences in stick composition (Wu et al., 2003).
Analyzing synthetic-shaft sticks in slap shots performed by varsity high school players, Rothsching found that, although relatively flexible sticks achieved the greatest puck velocities overall, “substantial variation between subjects occurred, emphasizing the greater importance of player technique and strength” (1997, cited in Pearsall et al., 2000, p. 691). Similarly, in an experiment with identical models of wood sticks with laminate shafts, Villasenor et al. (2006) found that stick deflection angles and subsequent puck velocities were significantly higher for elite versus recreational players, indicating that slap shot speeds generated by identically constructed sticks vary greatly from athlete to athlete. To date, then, and contrary to much conventional belief, scholars have not linked any particular stick material to increased slap shot velocity. Rather, what has surfaced from research reports is the clear primacy of the athlete’s variables—technique and strength—over any differences in stick composition.
Stick Stiffness and Flexibility
Beyond the individual athlete’s overriding influence on slap shot speeds, what has also emerged from recent scholarly investigations is the notion that stick flexibility, not stick composition, is of primary concern. In fact, several slap shot studies involving both wood and composite sticks demonstrate the influence of stick flexibility on shooting velocity. For instance, in a study of composite sticks exhibiting eight different stiffness levels (from “low” to “pro-stiff”), Worobets, Fairbairn, and Stefanyshyn (2006) found that in wrist shots, highly flexible sticks stored the most strain energy during the loading phase. Complicating matters, however, are the authors’ conclusions that the benefits of utilizing a flexible stick did not extend to slap shots, where “it is the athlete and not the equipment influencing shot speed” (p. 191). With this conclusion, Worobets et al. issue hockey players a strong reminder of the primacy of their own performance over any technological innovations in hockey sticks.
In a related investigation, Pearsall et al. (1999) explored slap shot velocities generated by four different “flexes” of carbon-fiber composite shafts with wood blades. The authors reported that, for each of the 6 college- and professional-level hockey player subjects, puck velocities were highest with the least stiff stick (“medium flex”); conversely, puck velocities were lowest when the subjects used the “extra stiff flex” stick. A “significant advantage” for puck velocity during slap shots was attributed to those hockey sticks with less shaft stiffness (p. 9). Qualifying such positive language, however, the authors also noted that variability in shooting velocity across subjects was greater than variability across shaft stiffness, concluding that “the subjects themselves are perhaps more important in determining slap shot velocity than the stick characteristics” (p. 10).
Finally, exploring slap shot velocities produced by 11-year-olds utilizing wood sticks of two different stiffness levels, Roy and Doré (1976) found that using the more flexible stick produced slightly higher slap shot speeds (56.8 km/h) than did using the stiffer model (54.4 km/h). The results prompted the authors to advise flexible sticks for use by younger players, since with flexible sticks, “lower forces are required to achieve the same puck velocity” recorded with stiffer shafts (Roy and Doré, 1976, cited in Pearsall et al., 2000, p. 690). Overall, then, the findings of Worobets et al., Pearsall et al. (1999), and Roy and Doré strongly suggest that the use of flexible hockey sticks contributes substantially to final puck velocity during the slap shot, especially when used by younger players. If any characteristic of a stick deserves to be considered for its effect on the slap shot, then, it appears to be stick flexibility, not stick composition.
Improved Slap Shot Performance
This review suggests that both player techniques and stick characteristics are important to slap shot success. Technical aspects of hockey shooting that may, if performed correctly, heighten ensuing puck velocities include intentionally tilting the stick blade forward to cup the puck and gripping the stick shaft low, even beyond the stick’s mid-point, to generate increased strain energy throughout the stick. In addition, expert shooters contacted the ice roughly 1 foot behind the puck to initiate stick bending at or before first contact with the puck—a crucial factor in maximizing shot velocity. Finally, accelerating the downswing phase first with the torso, then with the shoulders and arms, allows a hockey player to create a “whiplike” kinetic chain, lengthening the duration of impulse application to the stick, thereby increasing final puck velocity. Clearly, hockey coaches and players stand to adjust a variety of technical details to hone their technique and positively influence their level of success in the slap shot.
Recommendations for Hockey Stick Selection
Equally clear as the need for these technical adjustments is the extant literature’s recurring theme that player technique and strength are the most important variables influencing slap shot velocity. Across studies of players from youths to professionals and of sticks from wood to composite, stiff to flexible, the preeminence of player influence on achieved slap shot speeds rings consistently true and thus deserves to be the primary focus of performance-driven hockey coaches and players alike.
That said, this review has uncovered several findings relating to hockey sticks themselves. First, current research does not clearly demonstrate any advantage for one particular stick composition (wood, aluminum, or composite) over others. Instead, scholarly findings point to stick flexibility as the key mechanical consideration in stick selection. Several investigations attest to the mechanical benefits—most notably in stick deflection and strain energy storage—achieved with highly flexible sticks. It would seem sensible for coaches to advise hockey players to use the most flexible sticks possible (without incurring constant breakage) to maximize shooting velocity. This recommendation seems particularly apt for younger, less powerful players who may generate more stick bending with less applied force. Research suggests, then, that attention to hockey stick flexibility over any particular stick material may best aid players in heightening slap shot speeds.
While shooting remains only one of a multitude of hockey stick tasks—including the precision skills of stickhandling, passing, and receiving—players nevertheless stand to positively affect slap shot performance by supplementing the principal concerns of player technique and bodily strength with the use of flexible hockey sticks. In this regard, improvement in various aspects of ice hockey slap shots contributes toward every player and coach’s ultimate goal: enhancing athletic performance.
References
Barpanda, D. (1998). Dynamic performance characterization of hockey sticks and golf clubs using a combined vibrational energy level and modal analysis approach. Unpublished doctoral dissertation, University of Mississippi.
Bracko, M. R. (2004, September). Biomechanics powers ice hockey performance. BioMechanics 11(9), 1–7. Retrieved March 29, 2008, from http://www.biomech.com/full_article/?ArticleID=827&month=09&year=2004
Carr, G. (2004). Sport mechanics for coaches (2nd ed.). Champaign, IL: Human Kinetics.
De Koning, J. J., & Van Ingen Schenau, G. J. (2000). Performance-determining factors in speed skating. In V. M.
Zatsiorsky (Ed.), Biomechanics in sport: Volume IX of the Encyclopaedia of Sports Medicine (pp. 232–246). London: International Olympic Committee.
Doré, R., & Roy, B. (1976). Dynamometric analysis of different hockey shots. In P. V. Kumo (Ed.), Proceedings of the Fourth International Congress on Biomechanics, V-B (pp. 277–285). Baltimore: University Park Press.
Doré, R., & Roy, B. (1978). The biomechanics of hockey shots. In Proceedings: 1978 national coaches certification program level 5 seminar (pp. 59–71). Ottawa, Ontario, Canada: Canadian Amateur Hockey Association.
Hache, A. (2002). The physics of hockey. Baltimore: Johns Hopkins University Press.
Hove, P. (2004). Haptic perception of affordances of a sport implement: Choosing hockey sticks for power versus precision actions on the basis of “feel.” Unpublished doctoral dissertation, University of Cincinnati.
Lomond, K. V., Turcotte, R. A., & Pearsall, D. J. (2007). Three-dimensional analysis of blade contact in an ice hockey slap shot, in relation to player skill. Sports Engineering, 10(2), 87–100.
Madill, H. W. (1980). An EMG analysis of the validity of using weighted hockey sticks for specific overload training. Unpublished master’s thesis, McGill University, Montreal, Quebec, Canada.
Marino, G. W. (1998). Biomechanical investigations of performance characteristics of various types of ice hockey sticks. In H. J. Riehle & M. M. Vieten (Eds.), ISBS conference proceedings, 16th international symposium (pp. 184-187). Konstanz, Germany: International Society of Biomechanics in Sports. Retrieved March 30, 2008, from http://w4.ub.uni-konstanz.de/cpa/article/viewFile/1633/1535
Oxendine, J. B. (1988). American Indian sports heritage. Champaign, IL: Human Kinetics.
Pearsall, D. J., Montgomery, D., Rothsching, N., & Turcotte, R. (1999). The influence of stick stiffness on the performance of ice hockey slap shots. Sports Engineering, 2(1), 3–11.
Pearsall, D. J., Turcotte, R. A., & Murphy, S. D. (2000). Biomechanics of ice hockey. In W. E. Garrett, Jr., & D. T. Kirkendall (Eds.), Exercise and sport science (pp. 675–692). Philadelphia: Lippincott, Williams, and Wilkins.
Roy, B., & Doré, R. (1976). Kinematics of the slap shot in ice hockey as executed by players of different age classifications. In P. V. Komi (Ed.), International society on biomechanics (pp. 286–290). Baltimore: University Park Press.
Sports materials: materials for sports equipment have advanced dramatically over the past several years. Here is a sampling of some of the materials that enable players to move faster, hit the ball farther, pedal longer, and be better protected. (2005, October). Advanced Materials and Processes 163(10), 22-25. Retrieved March 29, 2008, from http://findarticles.com/p/articles/mi_hb5260/is_/ai_n20378099?tag=artBody;col1
Villasenor, A., Turcotte, R. A., & Pearsall, D. J. (2006). Recoil effect of the ice hockey stick during a slap shot. Journal of Applied Biomechanics, 22(5), 202–211.
Worobets, J. T., Fairbairn, J. C., & Stefanyshyn, D. J. (2006). The influence of shaft stiffness on potential energy and puck speed during wrist and slap shots in ice hockey. Sports Engineering, 9(4), 191–200.
Wu, T.-C., Pearsall, D., Hodges, A., Turcotte, R., Lefebvre, R., Montgomery, D., et al. (2003). The performance of the ice hockey slap and wrist shots: The effects of stick construction and player skill. Sports Engineering, 6(1), 31–40.
Author Note
David J. Laliberte, MSS, MA, Minnesota Hockey Coaches Association.
The author thanks Dr. Douglas Goar of the United States Sports Academy for his encouragement and insight regarding this essay.
Correspondence concerning this article should be addressed to David J. Laliberte, Minnesota Hockey Coaches Association, 1108 N. Seventh Ave., St. Cloud, MN 56303. E-mail: dlaliberte@usa.net.