Secondary Logo

Journal Logo

Original Research

Warm-up Practices in Elite Boxing Athletes: Impact on Power Output

Cunniffe, Brian1,2; Ellison, Mark3; Loosemore, Mike1,2,3; Cardinale, Marco4

Author Information
Journal of Strength and Conditioning Research: January 2017 - Volume 31 - Issue 1 - p 95-105
doi: 10.1519/JSC.0000000000001484
  • Free



Warm-ups are used before competitive sporting activity as a means to activate the body, reduce injury risk and increase performance in subsequent tasks (14). This may be achieved through a variety of physiological and biochemical responses, including increased muscle and tendon compliance, elevated blood flow to the muscles, acceleration of metabolic reactions, and faster nerve conduction (5,18,33). Other psychological effects have been hypothesized to occur as a result of competition warm-up such as increased preparedness (5), whereas warm-up strategies have also been shown to accelerate the rate of oxygen consumption (V̇o2 kinetics) in addition to blunting the lactate response during subsequent exercise (6,17,35). Increased muscle temperature is one of the main outcomes of warm-up strategy, and it has been shown that this has a positive influence on muscle function. In particular, greater anaerobic energy (ATP) turnover has been shown to occur during the initial 2 minutes of intense exercise after passive (warm water immersion) muscle heating (15). Previous work on passive heating of large muscles has been shown to acutely shift the force-velocity (F/V) and power-velocity (P/V) relationship of the lower limbs muscles to the right (2). Furthermore, improvements (∼9%) in sprint cycling performance have been observed by reducing the drop in muscle temperature after warm-up through use of heating garments (13). Taken together, findings have shown that implementation of a correct warm-up strategy is essential to athletic performance and that maintenance or increase in muscle temperature after warm-up seems to be beneficial for power-generating capacity.

The Olympic sport of amateur male boxing is a dynamic, physically demanding sport that requires a high level of physical fitness, complex skills, speed, and tactical excellence for success over 3 × 3 minutes (male) or 4 × 2 minutes rounds (female). In previous studies, athletes have been shown to exercise between 85–90% heart rate maximum (HRmax) during simulated competition, whereas blood lactate values of ∼9–12 mmol have been reported after 3 rounds of sparring (1,11). In amateur boxing, upper-body power output (PO) is important as fast punches are more likely to score points, in particular during the early stages of a fight (8). Therefore, optimization of warm-up strategies seems pertinent for this sport. Although warm-up routines are often individually varied in both time and intensity, it is generally observed that at amateur-level, a typical athlete warm-up may last between 10 and 20 minutes in duration. In turn, the time gap between cessation of warm-up and start of amateur boxing competition can also vary considerably (usually 5–15 minutes). This is normally dependant on tournament structure, injuries in the ring, and importance of tournament (dignitary visits, prefight protocols, etc.). Previous work has shown an improvement of 1.5 ± 1.1% in 200 m free-style swimming performance when performed 20 minutes post–warm-up (vs. 45 minutes) (34). In amateur boxing, anecdotal observations would suggest that it is conceivable that an athlete may have to wait up to 20 minutes from scheduled fight-time during a World championship or an Olympic tournament. Therefore, the timing and type of warm-up activity, length of warm-up, and practices adopted by the athlete during the period between finishing normal warm-up and actual start of competition would be obvious areas to consider in ensuring competition day performance is optimized by practitioners.

Despite the widespread acceptance that warm-up confers a psycho-physiological benefit; to the author's knowledge, no studies have investigated the implications of such in combat sports. Furthermore, there is a paucity of information describing how long the proposed benefits of athletic warm-up (physiological and/or performance related) actually last and what the consequences of gap times are. Recent data in basketball have shown that fast declines in key performance indicators (lower-body PO, speed) occur when players remain inactive after a warm-up, also known as GAP period (14). Decreases were shown to be linear in fashion and were paralleled by a decrease in body temperature. These data, together with recent experimental findings showing the benefits of elevating and maintaining muscle temperature after warm-up during the early stages of exercise (13,16) have obvious implications to the sport of amateur boxing.

The aim of this study was to assess standard practice of elite amateur boxers and quantify the effects of warm-up and GAP to provide some advice to optimize performance. For this scope, athletes were assessed before, during, and after a typical warm-up protocol. Specific focus was to characterize changes in body temperature with performance variables after warm-up (GAP period) and commencing amateur boxing activity.


Experimental Approach to the Problem

An observational study was used to capture routine warm-up practices performed by the athletes. Collection of information (warm-up specifics, temperature measurements, physical performance tests) took place over 2 weeks at the squad's national indoor training base 3 months before the London 2012 Olympic Games.


Data were collected from 6 international level Olympic male amateur boxing athletes [age 22.5 (2.5) yr., height 1.79 (0.11) m, weight 68.5 (19.8) kg, sum of 8 skinfolds 54 (2.3): mean (±SD)]. All athletes were ranked within the top 10 of the AIBA World Rankings at the time of testing and competed in separate weight categories, ranging from flyweight (<52 kg), bantamweight (<56 kg), lightweight (<60 kg), light-welterweight (<64 kg), welterweight (<69 kg), and super-heavy weight (>91 kg).

All athletes provided written informed consent before volunteering for the study and any athlete who wished to withdraw could do so at any time. All experimental procedures were approved by the Research Ethics committee of University College London (3165/003).


Before data collection, the athletes were familiarized with the nature of the exercise testing and the procedures. The GAP time was determined as the time between the end of structured warm-up and getting into the boxing ring for sparring activity. To investigate the period of change in body heat, the GAP was set to 25 minutes fixed duration. This represented the theoretical maximum waiting time possible between an athlete warming up and beginning boxing activity in an amateur competition at the time of investigation and was primarily based from anecdotal observations in competition. All athletes and coaches were briefed as to the intended goals of the investigation and asked to select a warm-up protocol that was routinely used by each boxer before a “typical” competition. Athletes were asked to do what they would “normally do” should a time GAP exist between finishing their warm-up and getting into the boxing ring but to refrain from food or drink. During amateur boxing activity, athletes completed 4 × 2 minutes rounds of sparring (1 minute standing rest between rounds) against an opponent from the same weight class within a standard competition ring (4.9 × 4.9 m). Each athlete wore hand wraps, 0.45 kg (16 oz) gloves, and customary head/jock protection during all boxing activity (including warm-up). After sparring, each athlete then undertook 4 × 2 minutes rounds of “pads and bags” work, whereby they either freely hit a punching bag (weight range 34–45 kg) and pads under guidance of nearby coach. A 1-minute standing rest was taken between rounds.

Body Temperature Measurements

To assess changes in body core temperature (Tcore), athletes ingested a thermistor capsule (precision 0.01° C; Vital Sense-Mini Mitter, Inc., Bend, OR, USA) 3 hours before warming up along with 250 ml of water. In addition, integrated wireless dermal patches (Vital Sense, Inc., Bend, OR, USA) were affixed to 3 body locations (upper thigh, lower back, upper arm) for measurement of body skin temperature (Tskin). For comparison between timepoints, the average of these 3 locations was then calculated. After a telemetric check to confirm that the capsule and dermal patches were transmitting temperature signals to the VitalSense telemetric monitor, the athletes then had the monitor securely fitted to a neoprene waist pouch and belt for continuous measurement. During sparring activity, the monitor belts were removed but the unit was kept within close proximity of the athlete by research/coaching staff to avoid undue loss of signal data. No signal loss in temperature or HR data occurred. Athletes were asked to avoid consuming large volumes of water so as to prevent interference with Tcore measurements.

At fixed intervals, high-resolution thermal imaging (ThermaCAM SC640, 640 × 480 pixel, 7.5–13 μm spectral range, 30 Hz sampling rate; FLIR Systems, Kent, UK) was used to visualize changes in anterior/posterior reflected cutaneous temperature (Tc) variations and possible temperature reduction after warm-up during the GAP period. The camera was set for human skin emissivity (e 0.98). Athletes observed standard preparatory procedures for thermal imaging measurements as outlined previously (27). More specifically, the athletes were asked to remove their tracksuit and stand on an identified ground marker placed 4 m from camera lens (tripod mounted), thus permitting simultaneous acquisition of specific regions of interest (ROIs). To standardize images and reduce signal noise, the period from removal of tracksuit to image capture was fixed at 1 minute. Both anterior and posterior images were captured at 5 set time intervals. These were as follows: (a) before commencing warm-up, (b) immediately after warm-up, (c) +15 minutes, (d) +25 minutes after warm-up, and finally, (e) after sparring activity (Figure 1). With respect to the fourth image capture, this was taken 5 minutes before entering the ring for sparring activity to allow athletes sufficient time to “glove up”. This concept refers to the application of wrist protection, strapping, and gloves before entering the ring. Given the nature of amateur boxing activity and demands placed on the upper body, ROIs were defined as (a) trunk + upper body and (b) limbs. Tc temperature data for trunk/upper body were analyzed from average data taken from chest, torso, back (upper and lower), and shoulders. Limb data included arms (below shoulder) and upper thigh (Figure 2). Postevent digitizing of ROIs was performed using associated software (ThermaCAM Researcher Pro 2.8 SR-3; FLIR Systems, Kent, UK). Within subject and between user coefficient of variation for select Tc ROIs were both <1%. To standardize conditions, all images were captured beside the boxing ring in a temperature controlled room in which the athletes were asked to enter ∼30 minutes before the first image capture (pre–warm-up). All measurements were performed in the late morning and environmental conditions were assessed during data collection using a portable thermometer (Higbo; Oregon Scientific, Berkshire, UK) at 15 minutes intervals. To assess intensity during warm-up and sparring activity, athlete HR data were collected wirelessly at 1-second intervals (Polar Team System; Polar Electro, Kempele, Finland). Heart-rate readings were classified into 5 intensity zones, ranging from 50–59%, 60–69%, 70–79%, 80–89%, and 90–100% of individual HR maximum as used in similar previous investigations (9,28). Athlete HRmax was determined using the Karvonen method (7).

Figure 2.
Figure 2.:
A) Sample image representing Tc from anterior and posterior thermal views (without markers) before and after standard warm-up, during period of inactivity (GAP) and after boxing activity. B) Changes in cutaneous temperature (Tc) values for (A) trunk/upper body and (B) limbs for all subjects. *Significantly different from +25 minutes GAP values, p ≤ 0.05. #Indicates a tendency from +25 minutes GAP values, 0.05 > p ≤ 0.1.

Performance Assessment

Dynamic upper-body PO was assessed from load-power curves generated from concentric-only bench press actions. To begin the test, the bar was positioned on the athlete's chest and was required to remain there for ∼1 second before beginning of movement. The athlete was then instructed to perform a concentric-only action from this starting position, as quickly as possible, until full extension of the elbows occurred. A trial was discounted if an initial countermovement of the bar appeared, if the athlete's lower back or buttocks were elevated off the bench, or if the athlete failed to achieve full-elbow extension (23). Bar displacement, average velocity (m·s−1), average force (F), and average power (W) for the concentric phase were recorded by attaching a linear encoder to the end of the bar (4020e; Muscle lab, Langesund, Norway). Customized software (MuscleLab Software V8.20; Ergotest, Langesund, Norway) was then used to generate individual F/V and load-power curves for each bench press repetition performed throughout the whole range of motion at known resistances. Given the large heterogeneity in strength between the athletes at different weight classes, resistances corresponding to 20, 40, 60, and 80% of previously established individual 1 repetition maximum (1RM) were chosen. Establishment of individual 1RM was performed on the bench press movement within 3 weeks of commencing the investigation. Three repetitions of bench presses were performed at each load and 1 minute of passive rest was observed between sets.

Lower-body PO was assessed indirectly from countermovement jump (CMJ) and 30 cm drop jump (DJ-30 cm) protocols set up on a touch mat interfaced with a laptop computer (KMS Innervations, Australia). After recording of body weight and a gentle warm-up (3–4 sub-max jumps), athletes were instructed to jump as high as possible, feet shoulder with apart and hands on hips. In all instances, each athlete performed 3 maximal CMJs with 30 seconds between each repetition. After instructions, athletes were then asked to perform 3 DJs from the edge of a 30-cm box onto the contact mat. Athletes were encouraged to react immediately on landing and jump as high as possible with upright trunk and small knee bend. Before the current investigation, all athletes were already familiarized with jump testing procedures. Timings of both upper- and lower-body tests with reference to beginning and ending exercise activity (warm-up) and boxing were kept consistent for each athlete. Given the variation in time taken by athletes for individual warm-up and amateur boxing activity, temperature (Tskin and Tcore) data were averaged for the first and last 5 minutes of each exercise category.

Figure 1.
Figure 1.:
Testing schematic.

Statistical Analyses

Students paired t-tests were then used to locate differences in Tskin and Tcore data at the start and end of warm-up and boxing activity. After normality checks, all remaining measures were analyzed using a repeated-measures analysis of variance (ANOVA). If significant main effects were found, the Bonferroni post hoc test was used to locate differences. Differences were considered significant at p ≤ 0.05, and a tendency is reported when 0.05 > p ≤ 0.1. (4,10,20,25) Where significance is indicated, mean ±90% confidence limits are presented for estimate of the population mean difference. Effect size (ES) were calculated using Hedges g for small sample sizes with the ranges of 0.2–0.6, 0.61–1.19, and >1.20 considered small, medium, and large effects, respectively (19). For ANOVA calculations, main ESs are presented as partial eta-squared. Data are reported as mean ± SE. All statistical analyses were performed using SPSS v 22.


No difference in room environmental temperature (20.1 ± 0.8° C) or relative humidity (30 ± 2%) values were noted throughout testing. In all cases, athletes were left to self-select their normal warm-up routine. After taping and application of hand wraps and gloves, all boxers were observed to perform general calisthenics, skipping, and shadow boxing involving high-frequency punches. This was then followed by short high-intensity efforts (∼1 minute per effort) on pads with their respective coaches. Feedback on technical aspects relating to punch angle, frequency, and shot section was provided by each athletes coach in between high-intensity efforts. Although content of warm-up was relatively similar between athletes, the mean duration for warm-up activity was quite varied across the group (12.1 ± 1.4 minutes; range: 7.4–18.5 minutes).

Heart Rates

Lower peak HR was observed during warm-up (158 ± 3.6 b·min–1) in comparison to both sparring (189 ± 3.2 b·min–1; −8.2 ± 3.3%; p = 0.004; ES, 3.7) and pads (189 ± 2.5 b·min–1; −9.4 ± 2.8%; p = 0.001; ES, 4.1). Average HR achieved by the boxers during warm-up (123 ± 5.8 b·min–1; 63.5 ± 3.0% HRmax) was also lower than both sparring (164 ± 4.4 b·min–1 or 84.3 ± 2.8% HRmax; −7.6 ± 2.6%; p = 0.001; ES, 2.2) and pad-work (172 ± 1.7 b·min–1 or 88.5 ± 1.5% HRmax; −9.9 ± 2.8%; p = 0.002; ES, 3.3). After warm-up, decreases in both peak (−3.8 ± 2.0%; p = 0.01; ES, 1.3) and average HR (−4.9 ± 2.9%; p = 0.004; ES, 1.3) were observed during GAP. During the GAP period, most of the time (82.2 ± 12.7%) was spent in lower HR zone (50–59% HRmax); values lower than those observed during warm-up (−4.9 ± 2.1%; p = 0.005; ES, 1.2) or boxing activity (−5.9 ± 2.0%; p = 0.002; ES, 1.9). During boxing activity, a large percentage of time (46.1 ± 6.5%) was spent exercising in high HR zones (>90% HRmax), values higher than both the warm-up activity (2.7 ± 0.9% HRmax; 4.0 ± 2.0%; p = 0.01; ES, 1.5) and GAP (<1% HRmax; 4.0 ± 2.0%; p = 0.03; ES, 1.6) (Figure 3).

Figure 3.
Figure 3.:
Percentage of time spent at different interval percentages of athlete's maximal heart rate during warm-up, GAP, and boxing activity. Values are mean ± SE. *Significantly different from warm-up values, p ≤ 0.05. #Significantly different from GAP, p ≤ 0.05.

Temperature Data

Increases in Tskin (4.6 ± 2.2%; p = 0.01; ES, 0.4) and Tcore (3.7 ± 2.0%; p = 0.02; ES, 1.2) were observed as a result of individual athlete warm-up (Figure 4). Skin temperature during the last 5 minutes of boxing exercise decreased (−2.7 ± 2.0%; p = 0.05; ES, 0.6) below Tskin values observed at the start of exercise. No difference in Tcore was observed between the start (37.6 ± 0.13) and end (37.8 ± 0.20) of boxing exercise. A significant effect of time was observed for Tskin during the GAP period with mean group values decreasing (F (6, 20) = 9.82; p = 0.001; η2 = 0.710). Pairwise comparisons showed a tendency for decreases in Tskin from the first (31.53 ± 0.94° C) and last 5 minutes of the GAP period (31.35 ± 0.75° C; p = 0.06; ES, 0.5). No significant effect of time was observed for Tcore during GAP and no other differences between mean 5 minutes sampling points were observed for Tskin and Tcore across the 25 minutes GAP period.

Figure 4.
Figure 4.:
Average skin and core temperatures readings taken during the first (clear marker) and last (filled marker) 5 minutes of both warm-up (triangle) and boxing (square) activity. Average 5-minute interval temperature data also taken during GAP (open circles). Values are mean ± SE. *Significantly different from pre–warm-up values, p ≤ 0.05. #Significantly different from first 5 minutes of exercise, p ≤ 0.05. $Indicates a tendency from first 5 minutes of GAP, 0.05 > p ≤ 0.1.

Reflected cutaneous temperature (Tc) readings taken from thermal images showed that limb Tc (p = 0.09; ES, 1.1) and trunk/upper body Tc (p = 0.08, ES, 1.2) tended to increase after 25 minutes GAP (0.05 > p ≤ 0.1); Figure 4. A reduction in trunk/upper body Tc was then observed immediately postboxing activity (sparring; 30.0 ± 0.38° C) when compared with +25 minutes GAP values (31.9 ± 0.27° C); (p = 0.01; ES, 2.3).

Performance Data

Lower Body

In comparison to pre–warm-up values, a tendency for an increase (4.8 ± 1.8%) in CMJ height was observed after warm-up (p = 0.06; ES, 0.4). Similar findings were observed for increase (5.8 ± 2.6%) in DJ height (p = 0.07; ES, 0.55). After warm-up, a decline in both CMJ and DJ height was then observed during the GAP period, with decreases (−2.6 ± 1.2%) in CMJ height reaching significance +15 minutes into the GAP period (−2.6 ± 1.9%; p = 0.05; ES, 0.3); Figure 5. After boxing activity, CMJ height returned to values observed post–warm-up; with values higher than pre–warm-up values (2.9 ± 1.9%; p = 0.03; ES, 0.4).

Figure 5.
Figure 5.:
Change (percent of pre–warm-up values) in countermovement and drop jump height across sampling period. Values are mean ± SE. *Significantly different from pre–warm-up values, p ≤ 0.05. #Indicates a tendency from pre–warm-up values, 0.05 > p ≤ 0.1. $Significantly different from post–warm-up values, p ≤ 0.05.

Upper Body

Average velocity measured in the concentric phase of the bench press lift tended to increase as a result of warm-up at 20, 60, and 80% 1-RM (0.05 > p ≤ 0.1), whereas significant decreases in velocity were observed post 25 minutes GAP at 20% (−6.9 ± 2.0%; p = 0.001; ES, 0.4) and 60% (−4.8 ± 1.9%; p = 0.005; ES, 1.0) 1-RM loads. Also, average velocity values at 80% 1-RM tended to decrease in comparison to post–warm-up values (p = 0.07). In comparison to post–warm-up values, decreases in upper-body force with loads corresponding to 20% 1-RM (−5.1 ± 3.5%; p = 0.01; ES, 0.3) and 60% 1-RM (−3.9 ± 1.9%; p = 0.04; ES, 0.3) were observed 25 minutes into GAP period. Force tended to decrease (p = 0.07) with loads corresponding at 80% 1-RM (0.05 > p ≤ 0.1). No differences in force output were observed between pre–warm-up values and those observed after 25 minutes GAP. These, together with changes in velocity data show an upward and rightward shift in athlete F/V curves after warm-up which then regressed to pre–warm-up values after 25 minutes GAP. Despite an overall increase (7.1 ± 6.0%) in upper-body peak PO after warm-up, these grouped load data did not reach significance (1.9 ± 2.0%; p = 0.11; ES, 0.5); Figure 6. Compared with pre–warm-up values, PO tended to increase after warm-up at 60% (1.9 ± 1.7%; p = 0.07; ES, 0.9) and 80% (1.8 ± 1.7%; p = 0.08; ES, 0.75) 1RM loads resulting in an upward shift in P/V curves (Figure 7). After warm-up, PO then declined with decreases observed at 20% (−9.7 ± 3.2%; p = 0.001; ES, 0.50) and 60% (−4.9 ± 2.1%; p = 0.004; ES, 0.80) 1RM loads 25 minutes into the GAP period causing a downward shift in P/V curves. No differences in bench press PO or average velocity were observed between pre–warm-up values and those observed 25 minutes post GAP.

Figure 6.
Figure 6.:
Change (percent of pre–warm-up values) in upper-body power output at varying percentage loads of individual one repetition maximum across sampling period. Values are mean ± SE. $Significantly different from post–warm-up values, p ≤ 0.05.
Figure 7.
Figure 7.:
Change in upper-body F/V and P/V curves (mean of 3 efforts) at varying percentage loads of individual one repetition maximum. Figures 7A and 7B show changes in variables before and after boxing warm-up. Figures 7C and 7D show changes in variables following warm-up and after a 25-min time gap. Figures 7E and 7F show variables before exercise and after a 25-min time gap. Values are mean ± SE. See results for significance effects at individual 1 RM loads.


Data in this study show that despite similar warm-up content and mean HR (123 ± 14 bpm) achieved during the warm-up activity, considerable variation in warm-up duration (12.1 ± 1.4 minutes; range 7.4–18.5 minutes) occurred within the present cohort of athletes. Despite this, significant increases in both Tskin (0.71 ± 0.14° C) and Tcore (0.30 ± 0.07° C) were observed, findings, which were evident in all athletes. In conjunction with the elevation in body temperature, acute improvements in vertical jump ability were caused by the warm-up strategies used (CMJ height +4.8 ± 1.8%; p ≤ 0.05) and (DJ +5.8 ± 2.6%; 0.05 > p ≤ 0.1). These data support previous findings that warm-ups might improve PO acutely, possibly as a result of both increases in blood flow to target muscles tissues and elevated rate of ATP turnover as a consequence of the higher muscle temperature (15). It is suggested that punch power and frequency represent key aspects of physical performance that contribute toward overall bout success in amateur boxing (11) Although all of these movements are performed from the upper body, use of lower-body musculature in amateur boxing should not be discounted, particularly in terms of contribution to punch power, body balance, and evasion of opponent shots through appropriate ring position, as well as the ability to transfer energy more effectively from the lower limbs through the arm and fist (22). Given that these movements need to be performed within very short periods, any gain in lower-body PO after warm-up can be conceived to provide performance benefits.

Despite this, changes in upper-body speed and power represent 2 obvious variables of interest to target within an amateur boxing warm-up. However, no study to date has determined the physiological impact of warm-up activity on such parameters. Results in this study indicate that a warm-up can produce upward and rightward shift in both F/V and P/V curves at varying loads of upper-body 1-RM. Specifically, data showed that upper-body PO tended to increase (13.5%; 0.05 > p ≤ 0.1) at 60 and 80% of individual 1RM loads, with similar values observed for average velocity (∼12%). These results, when taken in consideration with lower-body PO data, are suggestive that typical warm-up practices adopted by elite amateur boxing athletes can be successful in conferring acute performance benefits. However, it should be conceded that overall increases in mean grouped load PO (7.1 ± 6.0%) or force (4.9 ± 3.0%) data did not reach overall statistical significance. This was possibly due to lack of statistical power given the low number of subjects (6 elite athletes) in the current investigation and due to variation in chosen warm-up duration and content. Future studies should therefore attempt to investigate varying warm-up strategies on boxing specific performance tasks similar to those used in the current investigation using a larger cohort.

Although it is important to know the magnitude of physical benefits imparted from a typical warm-up, perhaps equally important from a coaching perspective is how long such conferred benefits actually last. In an athletic event, it is common for athletes to experience delays between completion of the warm-up and the start of competition or where there are delays between repeated efforts/rounds of competition (24,34). Such periods of time may be sufficient for body temperature, and in particular muscle temperature (Tm) to drop below an optimal level (34). In this study, a fixed time GAP of 25 minutes was used in a deliberate attempt to mimic a “worst case scenario” and monitor the impact of self-selected athletic activity during this period as viewed from changes in body temperature and PO. Despite athletes being instructed to do “what they would normally do” should a GAP (delay in competition time) arise, HR data showed that athletes remained relatively inactive during this period. Both average and peak HR data were lower than those achieved during warm-up, with significantly greater proportion of time spent exercising within the lowest HR zone. Although a decrease in HR would have been inevitable during a GAP, average HR values were still much lower than would have been expected. Despite this, decreases in Tcore were not observed during this 25 minutes window of relative inactivity. After an initial increase with structured warm-up, data suggest that Tcore was well maintained during a period of inactivity on a group basis. However, it should be noted that in 2 of the 6 athletes, Tcore was observed to decrease back to pre–warm-up values after ∼20 minutes GAP, indicating that some degree of body heat loss was evident, whereas the athletes waited to get into the boxing ring. In 3 of the athletes (lowest weight classes), Tskin dropped below values observed before warm-up ∼10 minutes into GAP indicating that surface heat loss had occurred. This occurred presumably as a result of inactivity and a combination of evaporative and convective heat transfer after warm-up. Given the small sample size, it was difficult to establish whether or not the magnitude of temperature changes were weight class dependent.

In comparison to post–warm-up values, CMJ height significantly decreased +15 minutes into GAP period (p ≤ 0.05). Furthermore, average force and PO (upper body) values at 20% 1RM and 60% 1RM loads were also significantly lower 25 minutes into the GAP period. Similar decreases were observed for bench press velocity indicating that physical benefits achieved after warm-up had abated after this 25 minutes GAP. This resulted in a downward shift in both F/V and P/V curves to pre–warm-up levels by this time point. It is unlikely that observed decreases were a result of physical fatigue given the HR values recorded during the GAP period and so perhaps observed changes were temperature dependent. It is acknowledged that measurement of muscle temperature (Tm) would have provided greater clarity on heat loss/gain as a function of warm-up and GAP period, in addition to Tm related effects on F/V and P/V curves. However, this was not feasible in our study due to the invasive nature of Tm measurements. Nevertheless, the relationship between Tm and skeletal muscle function has been previously well established (3,12,29–31). Studies have shown that with increasing Tm, force/velocity relationship shifts are evident in both fast- and slow-twitch muscle fibers (12,30) and that with every 1° C rise in Tm there is ∼4–10% improvement in peak PO (3,13,31). Indeed, improvements (∼9%) in sprint cycling PO have been observed by reducing the drop in Tm (during a similar GAP duration) after warm-up (13). Previous work has shown that thermal infrared imaging permits the quantitative evaluation of specific cutaneous whole-body thermal adaptations, which occur during and after graded physical activity (26). In an effort to visualize avenues of heat loss, thermal images were taken at fixed intervals throughout the GAP period. After an initial decrease in Tc after warm-up, most probably due to redistribution of blood flow to deeper muscles, a progressive increase in Tc was observed with inactivity in the upper body/trunk. Because during upper-body exercise, there is a greater reliance on torso dry heat loss for temperature regulation (32), this is perhaps not surprising and so insulating garments should perhaps focus on this body area in particular.

Practical Applications

Data in this study showed a clear beneficial shift in F/V and P/V profiles after a typical amateur boxing warm-up but that observed benefits had almost completely abated after a 25-minute GAP. This highlights the importance of correct timing of athletic warm-up in advance of competition, the impact of keeping athletes active to maintain body temperature during a GAP and perhaps more importantly, methods to prevent drop in Tm during this GAP period. This may be particular relevant to sports that are of short duration and are more heavily reliant on high levels of power production. With particular reference to the sport of amateur boxing, athletes are required to exert forces dynamically and statically over short periods, primarily using upper-body musculature. It is known that when performing exercise at a given PO, both the absolute and relative exercise intensity (% of peakV̇o2) is greater during upper-body exercise (32). Therefore, in an effort to maximize the benefits of a routine warm-up and avoid inducing local fatigue/glycogen depletion through continued exercise, it may be prudent for amateur boxing athletes to prevent heat loss should an unwanted GAP period present. Insulating garments, particularly in athletes with low body fat levels are one possible avenue to attenuate loss of body heat although recent data suggest that passive heat garments are more effective in preventing a drop in Tm (13). This should only be performed so with previous consideration of environmental conditions and avoidance of thermoregulatory strain. Although the strength of this study is represented by the assessment of the real scenarios in an elite athletic cohort, a greater number of subjects would have reduced the chances of a type II error. With additional subject numbers, it may have been possible to identify those subjects who displayed a greater loss in body heat and whether or not, this was dependent on anthropometric characteristics and/or weight class. Furthermore, although the focus of this work was to capture what athletes routinely do as part of warm-up, variation in warm-up intensity and duration may have affected physiological parameters studied during subsequent sampling points. Future studies should focus on the most appropriate strategies to warm-up in combat sports but also in identifying the best solutions to reduce the negative effects of a GAP.

Current data highlights the importance of athlete warm-up in advance of competition and the gradual decline in performance parameters during a typical GAP period.


The authors wish to thank the staff at British Boxing, in particular Ian Pyper, Robert Gibson and Rob McCracken for their kind assistance with coordination of this study. Their appreciation is also extended to Ashley Gray, Sports Technology Institute, Loughborough University and Carissa Fallan (BOA) for their assistance with Thermal imaging analysis. Finally, thanks to all the athletes who kindly agreed to participate in this study and the British Olympic Association for the funding support.


1. Arseneau E, Mekary S, Léger LA. VO₂ requirements of boxing exercises. J Strength Cond Res 25: 348–359, 2011.
2. Ball D, Brown P, Cardinale M. In vivo contractile characteristics of human muscle: The effect of acute temperature change. Med Sci Sport Exerc 35(Suppl 1): S389, 2003.
3. Bergh U, Ekblom B. Influence of muscle temperature on maximal muscle strength and power output in human skeletal muscles. Acta Physiol Scand 107: 33–37, 1979.
4. Biensø RS, Ringholm S, Kiilerich K, Aachmann-Andersen NJ, Krogh-Madsen R, Guerra B, Plomgaard P, van Hall G, Treebak JT, Saltin B, Lundby C, Calbet JA, Pilegaard H, Wojtaszewski JF. GLUT4 and glycogen synthase are key players in bed rest-induced insulin resistance. Diabetes 61: 1090–1099, 2012.
5. Bishop D. Warm up II: Performance changes following active warm up and how to structure the warm up. Sports Med 33: 483–498, 2003.
6. Burnley M, Doust JH, Jones AM. Effects of prior warm-up regime on severe-intensity cycling performance. Med Sci Sports Exerc 37: 838–845, 2005.
7. Camarda SR, Tebexreni AS, Páfaro CN, Sasai FB, Tambeiro VL, Juliano Y, Barros Neto TL. Comparison of maximal heart rate using the prediction equations proposed by Karvonen and Tanaka. Arq Bras Cardiol 91: 311–314, 2008.
8. Chaabène H, Tabben M, Mkaouer B, Franchini E, Negra Y, Hammami M, Amara S, Chaabène RB, Hachana Y. Amateur boxing: Physical and physiological attributes. Sports Med 45: 337–352, 2015.
9. Cunniffe B, Fallan C, Yau A, Evans GH, Cardinale M. Assessment of physical demands and fluid balance in elite female handball players during a 6-day competitive tournament. Int J Sport Nutr Exerc Metab 25: 78–88, 2015.
10. Curran-Everett D, Benos DJ. Guidelines for reporting statistics in journals published by the American Physiological Society. Am J Physiol Endocrinol Metab 287: 189–191, 2004.
11. Davis P, Wittekind A, Beneke R. Amateur boxing: Activity profile of winners and losers. Int J Sports Physiol Perform 8: 84–91, 2013.
12. De Ruiter CJ, De Haan A. Temperature effect on the force/velocity relationship of the fresh and fatigued human adductor pollicis muscle. Pflugers Arch 440: 163–170, 2000.
13. Faulkner SH, Ferguson RA, Gerrett N, Hupperets M, Hodder SG, Havenith G. Reducing muscle temperature drop post warm-up improves sprint cycling performance. Med Sci Sports Exerc 45: 359–365, 2013.
14. Galazoulas C, Tzimou A, Karamousalidis G, Mougios V. Gradual decline in performance and changes in biochemical parameters of basketball players while resting after warm-up. Eur J Appl Physiol 112: 3327–3334, 2012.
15. Gray SR, De Vito G, Nimmo MA, Farina D, Ferguson RA. Skeletal muscle ATP turnover and muscle fiber conduction velocity are elevated at higher muscle temperatures during maximal power output development in humans. Am J Physiol Regul Integr Comp Physiol 290: R376–R382, 2006.
16. Gray SR, Soderlund K, Watson M, Ferguson RA. Skeletal muscle ATP turnover and single fibre ATP and PCr content during intense exercise at different muscle temperatures in humans. Pflugers Arch 462: 885–893, 2011.
17. Hajoglou A, Foster C, de Koning JJ, Lucia A, Kernozek TW, Porcari JP. Effect of warm-up on cycle time trial performance. Med Sci Sports Exerc 37: 1608–1614, 2005.
18. Hedrick A. Physiological responses to warm-up. NSCA J 14: 25–27, 1992.
19. Hopkins W, Marshall S, Batterham A, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 41: 3–12.19, 2009.
20. Kiilerich K, Ringholm S, Biensø RS, Fisher JP, Iversen N, van Hall G, Wojtaszewski JF, Saltin B, Lundby C, Calbet JA, Pilegaard H. Exercise-induced pyruvate dehydrogenase activation is not affected by 7 days of bed rest. J Appl Physiol 111: 751–757, 2011.
21. Lakens D. Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Front Psychol 4: 863, 2013.
22. Loturco I, Nakamura FY, Artioli GG, Kobal R, Kitamura K, Cal Abad CC, Cruz IF, Romano F, Pereira LA, Franchini E. Strength and power qualities are highly associated with punching impact in elite amateur boxers. J Strength Cond Res 30: 109–116, 2016.
23. Marques MC, van den Tilaar R, Vescovi JD, Gonzalez-Badillo JJ. Relationship between throwing velocity, muscle power, and bar velocity during bench press in elite handball players. Int J Sports Physiol Perform 2: 414–422, 2007.
24. McGowan CJ, Pyne DB, Thompson KG, Rattray B. Warm-up strategies for sport and exercise: Mechanisms and applications. Sports Med 45: 1523–1546, 2015.
25. McGowan CJ, Thompson KG, Pyne DB, Raglin JS, Rattray B. Heated jackets and dryland-based activation exercises used as additional warm-ups during transition enhance sprint swimming performance. J Sci Med Sport 19: 354–358, 2016.
26. Merla A, Mattei PA, Di Donato L, Romani GL. Thermal imaging of cutaneous temperature modifications in runners during graded exercise. Ann Biomed Eng 38: 158–163, 2010.
27. Merla A, Romani GL. Functional infrared imaging in clinical applications. In: The Biomedical Engineering Handbook. Bronzino JD, eds. New York, NY: CRC Press, 2005. pp. 32.1–32.13.
28. Póvoas SC, Seabra AF, Ascensão AA, Magalhães J, Soares JM, Rebelo AN. Physical and physiological demands of elite team handball. J Strength Cond Res 26: 3365–3375, 2012.
29. Rall JA, Woledge RC. Influence of temperature on mechanics and energetics of muscle contraction. Am J Physiol 259: R197–R203, 1990.
30. Ranatunga KW. Force and power generating mechanism(s) in active muscle as revealed from temperature perturbation studies. J Physiol 588: 3657–3670, 2010.
31. Sargeant AJ. Effect of muscle temperature on leg extension force and short-term power output in humans. Eur J Appl Physiol Occup Physiol 56: 693–698, 1987.
32. Sawka MN, Pandolf KB. Upper Body Exercise: Physiology and Training Application for Human Presence in Space. Warrendale, PA: SAE, 1991. p. 1–19.
33. Shellock FG, Prentice WE. Warming-up and stretching for improved physical performance and prevention of sports-related injuries. Sports Med 2: 267–278, 1985.
34. West DJ, Dietzig BM, Bracken RM, Cunningham DJ, Crewther BT, Cook CJ, Kilduff LP. Influence of post-warm-up recovery time on swim performance in international swimmers. J Sci Med Sport 16: 172–176, 2013.
35. Wittekind A, Cooper CE, Elwell CE, Leung TS, Beneke R. Warm-up effects on muscle oxygenation, metabolism and sprint cycling performance. Eur J Appl Physiol 112: 3129–3139, 2012.

athlete preparation; body temperature; force; velocity

© 2016 National Strength and Conditioning Association