Factors influencing pacing in triathlon.

Triathlon is a multisport event consisting of sequential swim, cycle, and run disciplines performed over a variety of distances. This complex and unique sport requires athletes to appropriately distribute their speed or energy expenditure (ie, pacing) within each discipline as well as over the entire event. As with most physical activity, the regulation of pacing in triathlon may be influenced by a multitude of intrinsic and extrinsic factors. The majority of current research focuses mainly on the Olympic distance, whilst much less literature is available on other triathlon distances such as the sprint, half-Ironman, and Ironman distances. Furthermore, little is understood regarding the specific physiological, environmental, and interdisciplinary effects on pacing. Therefore, this article discusses the pacing strategies observed in triathlon across different distances, and elucidates the possible factors influencing pacing within the three specific disciplines of a triathlon.

Introduction

Triathlon is a unique sport that consists of consecutive swim, cycle, and run disciplines completed over a variety of distances. The origin of triathlon is unclear; however, the first officially organized Ironman triathlon was conducted in Hawaii in 1978 with only 12 participants.1 Over the last 30 years, the popularity of this sport has increased tremendously, driving the emergence of other shorter (ie, sprint, Olympic, half-Ironman) and longer (ie, double to 20× Ironman) distances as well as other formats (ie, off-road) of triathlons.2 Of these, the most popular standard triathlon distances include the sprint (swim: 0.75 km, cycle: 20 km, run: 5 km, ∼1 hour), Olympic (swim: 1.5 km, cycle: 40 km, run: 10 km, ∼2 hours), half-Ironman (swim: 1.9 km, cycle: 90 km, run: 21.1 km, ∼4–5 hours), and Ironman (swim: 3.8 km, cycle: 180 km, run: 42.2 km, 8–17 hours). Due to the large variations in distances, and thus exercise duration, the metabolic demands and physiological responses during such races could vary greatly.3,4

It is well accepted that the distribution of speed, work, or energy expenditure throughout an exercise task is extremely important in optimizing performance.5–7 This pattern of energy expenditure or distribution of speed is known as “pacing” and, although often used interchangeably, differs slightly to the term “pacing strategy”, which refers to a conscious strategy or plan to manipulate effort. Indeed, it has been proposed that the distribution of speed throughout an exercise task may be partially regulated on a subconscious level8 and is, therefore, presumably disjointed somewhat from the athletes’ pre-race strategy or plan. Within this context, energy expenditure is constantly regulated in response to complex interactions between peripheral feedback and central drive to ensure physiological systems are maintained within homeostatic or manageable limits, whilst delaying the negative effects of fatigue and, thus, maximizing performance.1,5,9–13 More recently, Edwards and Polman14 introduced the concept of a “pacing awareness” model, which provides an alternative theory by proposing that pacing regulation occurs via relative states of awareness, ranging from minimum (sleep) to maximum (fully aware), rather than exclusive subconscious or conscious states. For instance, minor corrections in homeostasis require little or no conscious awareness to accomplish; however, strenuous activities involving large metabolic disturbances such as glycogen depletion necessitate a more conscious approach requiring significant behavioral alterations.14 It is plausible that metabolic demands will differ among the several disciplines and various competition distances within triathlon and, thus, the levels of subconscious and conscious awareness.

To date, studies examining the mechanisms that influence pacing have focused on single sport events such as running,15,16 cycling,6,7,17–19 swimming,20,21 and rowing.22 These single sport studies have identified possible factors regulating pace, including the availability of fuel substrates,4,23,24 thermoregulation,9,24 previous experience,17,25,26 knowledge of exercise duration,25,27 physical fitness,23 cognitive capacity,28 mood,29,30 peripheral feedback,25 and central regulation.4,5 However, the majority of these studies has been performed under laboratory conditions and may not precisely replicate the demands during actual triathlon competition. Understanding pacing during such events is complex since athletes are required to not only distribute their effort over the entire event but also over each independent discipline. Indeed, recent studies have demonstrated that the self-selected pacing patterns differ greatly during the swim,31–34 cycle,10,32–35 and run11,12,32–34 portions of triathlon events.

The sport of triathlon provides a unique model for pacing analysis, due to the involvement of sequential swim, cycle, and run disciplines in continuum, and the ability to examine the influence of race distance on pacing. To date, a comprehensive review examining factors that may be responsible for differences in pacing over various triathlon distances and disciplines is currently lacking. As such, the strategy or strategies that assist in optimizing performance during the various standard triathlon events is presently unclear. Therefore, the purpose of this review is to: 1) identify factors regulating and influencing pacing in triathlon; 2) discuss the effectiveness of and provide recommendations for the currently adopted pacing strategies in triathlon; and 3) examine the influence and relationships of performance during individual disciplines on/with subsequent and overall triathlon performance.

Factors influencing pacing in triathlon

Of the multitude of factors influencing pacing, some important factors associated with triathlon performance include: exercise distance/duration;13 race dynamics (drafting conditions,36 influence of other competitors;37,38) environmental factors (sea currents, wind velocities,39,40 topography3,41,42); transitions (swim to cycle, cycle to run); age;32,43 and sex.43–45 Of these factors, exercise duration appears to have the most significant influence on both the self-selected and the optimal pacing strategies selected by athletes during competition. Certainly, ultra-endurance events (>6 hours)46 induce greater neuromuscular fatigue due to greater demands on metabolic substrates and psychological factors.5 Despite the important role that exercise duration plays on pacing, there is currently no research that has extensively examined the influence of race distance on pacing in triathlon. Indeed, the majority of studies that have examined the distribution of pace within the triathlon have focused on the Olympic distance triathlon,10,11,32–34 and only one study has investigated the cycle discipline of the Ironman.35 Clearly, the abovementioned factors may influence the regulation of pace during triathlon to different extents. These factors are reviewed below in the context of understanding the effectiveness of such strategies and providing recommendations for optimal pacing.

Distance

Exercise duration appears to be one of the most important factors influencing both optimal and self-selected pacing.22,31,61,63–66 Indeed, it is plausible that differences in race distance will have a significant influence on the mechanisms responsible for fatigue and, thus, athletes’ pacing.9 While a gradual reduction in speed (ie, positive pacing) was observed in the Olympic32,33,35 and Ironman distance triathlons,35 the mechanisms responsible for such changes in pace are likely to vary. For instance, during shorter duration triathlon events (ie, the sprint distance), the progressive reduction in pace may be associated with metabolite accumulation and accompanying neuromuscular fatigue. Indeed, the progressive reduction in speed during relatively short duration 100 m, 200 m, 400 m, and 800 m15 running events has been attributed to the accumulation of anaerobic metabolites, which in turn increases muscular acidity, impairing glycolysis48 and muscular contractions.49 However, during longer duration triathlons (ie, half-Ironman or Ironman), fatigue is likely to be associated with reductions in muscle glycogen content50–55 and neuromuscular activity.9,56–58 For instance, reduced time to fatigue during cycling and increased performance time during time trials were observed during cycling exercise without carbohydrate ingestion, as compared to carbohydrate supplementation.54 Furthermore, a reduction in carbohydrate availability has also been shown to reduce cycling pace after ∼2 hours of prolonged exercise.55 Supporting the reduction in neuromuscular activity during endurance exercise, St Clair Gibson et al58 observed a progressive decline in electromyogram (EMG) activity and mean power output of intermittent 4 km high-intensity intervals during a 100 km cycling time trial.

Despite the different energy demands and physiological responses between triathlons of various distances, athletes of various caliber typically adopt a fast-start strategy, regardless of race distance.33,35 This fast start is likely to be caused by the high initial swim intensity required to achieve a good drafting position during the early portions of the swim discipline33,34 (see “Race dynamics and drafting in mass-start events”). Furthermore, evidence suggests that a fast-start pacing strategy may enhance oxygen kinetics and improve performance during short- to middle-distance (3- to 7-minute) exercise tasks.18,22,59 For instance, Bailey et al59 compared the effect of a fast-start, even-start, and slow-start pacing strategy on 3-minute cycling performance. The authors found superior performance with the fast-start (7% greater power output), as compared with the slow- and even-start, and attributed this to a faster oxygen consumption (VO2) response. As such, it is possible that the adoption of a fast-start pacing strategy, especially during the swim discipline of shorter sprint triathlons, may enhance oxygen kinetics60 and improve overall swim performance.61 While the adoption of a fast-start pacing strategy minimizes the time required to reach peak velocity and may improve performance during short explosive (≤2-minute) sporting events,5,40 the acceleration phase could have limited influence on performance during longer endurance events. Instead, due to the prolonged nature of exercise in a triathlon (∼1 hour or more), the adoption of a fast-start pacing strategy could lead to suboptimal distribution of energy resources and cause premature fatigue.62,63 Under such circumstances, a more conservative pacing strategy that allows for the conservation of glycogenic resources may be more optimal for overall performance.

While a fast-start pacing strategy is often observed during the beginning of triathlon (ie, swim discipline), there is evidence to indicate that an even pacing strategy, achieved by maintaining a constant velocity despite varying external conditions (ie, wind and altitude), may be ideal during endurance events such as the triathlon.5–7,19,64 Within this context, endurance performance is compromised if an athlete’s speed decreases below the mean speed at any time throughout the event, even if they attempt to increase power output to make up for lost time during the final stages of the race.65 This is because an increase in velocity requires a dramatic increase in energy expenditure to overcome the nonlinear increase in resistive forces experienced, and could lead to premature fatigue.66,67 Consequently, an even pacing strategy results in the best possible balance of both propulsive and resistive forces, and thereby maximizes overall performance during endurance events of varying distance. For instance, Le Meur et al11 observed a more even running pace in the best runners during the 10 km run of an Olympic-distance triathlon, despite changes in gradient, and these runners demonstrated superior ability in limiting decrements in running speed during the later stages. Likewise, Lambert et al16 investigated the pacing strategies adopted by 67 runners during a 100 km ultra-marathon running race and observed a tendency for the better (fastest ten out of 67) performing runners to adopt a relatively even pacing strategy during the first 50 km of a 100 km race. These runners also experienced the lowest reduction in running velocity during the second 50 km. It is possible that these runners have, to the best of their ability, attempted to adopt an even pacing strategy, but involuntarily slowed down due to significant metabolic4,5,9,23,68 and/or psychological disturbances.5,25,37,47 It is important to note that, although an even pacing strategy may theoretically be optimal for performance during endurance exercise due to the balance of propulsive and resistive forces, this strategy may not be physiologically optimal. Indeed, large variations in power output to counteract external perturbations such as topography and wind (discussed in “Environmental factors” below) are required to maintain an even pacing strategy. Such changes in power output have been shown to increase physiological strain and reduce performance.69 Under such circumstances, variations in speed to maintain a relatively even distribution of power output/energetic resources may likely be a more optimal strategy for athletes.

Race dynamics and drafting in mass-start events

Numerous factors associated with race dynamics may also influence optimal and self-selected pacing in triathlon. The ability to conserve and maximize efficiency of one’s energy is crucial for triathlon success. During mass-start events such as triathlon, athletes are often able to draft within a sheltered position behind another competitor,70 which provides for an opportunity to conserve energy, and plays a considerable role in determining both optimal and self-selected pacing during triathlon.47 For example, drafting during the swim portion of a triathlon allows an athlete to achieve a faster swim time by drafting off a faster swimmer, which conserves energy for the latter cycle and run disciplines.70 Certainly, drafting has been shown to improve swimming economy by reducing the drag force on the drafter (10%–26%),71–73 blood lactate concentration (31%), rating of perceived exertion (21%),74 and oxygen consumption (5%–10%).41,73–79 The relatively fast speeds often observed at the beginning of triathlon events (especially for wave-starts or group starts) are therefore likely to be influenced by race dynamics. Indeed, starting at a swimming velocity faster than their mean speeds will allow athletes to achieve a strategic position behind the fastest swimmer so as to conserve energy throughout the remainder of the swim, reduce the delaying effects of previous waves of swimmers, and allow them to be better positioned for the beginning of the cycle discipline.32–34 For instance, during a seven-lap 40.2 km cycle draft-legal World Cup triathlon, Vleck et al33 observed that male swim finishers who were not able to maintain a position within the lead swim group (within 13.6±8.5 seconds of the lead swimmer’s finishing time) were unsuccessful in staying within the first two cycle packs by the end of lap one. Subsequently, these athletes had to cycle faster in laps two and three in order to “bridge the gap” by the end of lap three.

Similarly, drafting is extremely important to performance within the cycle discipline of triathlon. Drafting during cycling drastically reduces fluid resistance (ie, aerodynamic drag)36,70,80,81 and, thus, can conserve a considerable amount of energy. For example, in a simulated sprint distance triathlon, Hausswirth et al80 observed a reduction in oxygen uptake (−14%), heart rate (−7.5%), and pulmonary ventilation (−30.8%) when athletes were drafting 0.2–0.5 m behind a lead cyclist at an mean speed of 39.5 km ·  h−1, compared with the non-drafting situation. Furthermore, the number of athletes present in the drafting pack may also influence the speed and energy utilization, which has a significant influence on the subsequent run.80 As such, the size and configuration of cycling packs, as well as the tactical locations of athletes within a pack, may considerably influence the distribution of exercise intensity within draft-legal races. Rather than maintaining an even pacing, drafting in the cycle discipline likely induces a stochastic power profile due to acute tactical changes in pacing.3,36 Indeed, Le Meur et al32 observed that, during a six-lap 40 km cycle discipline of an elite Olympic-distance triathlon, male athletes adopted an initial positive pacing strategy during the first three laps, followed by a slight increase in speed during lap four, which was maintained until the end of the cycle discipline. In a separate 40 km elite Olympic-distance cycle discipline, Vleck et al33 observed an entirely different pacing strategy, where male athletes began by increasing their speed for up to 50% of the distance (20.1 km), followed by decrements in speed until the 33.5 km point, and then increasing their speed thereafter until the finish. Collectively, these studies indicate that pacing during triathlons could be highly variable and out of the individual’s control depending on drafting and race dynamics; the beneficial effects of drafting may limit an athlete to cycling in a pack, which is highly influenced by peer race tactics.82,83

Due to the importance of maintaining a position with the lead athletes during the cycle discipline, the pacing of top contenders during elite competitions is highly dependent on the pacing strategy of the leading athlete, who may attempt to maximize their lead, which may or may not be matched by other competitors. Therefore, pacing could be highly variable across the three disciplines. For instance, three studies32–34 that have examined the pacing of elite athletes in draft-legal races have collectively reported that these athletes typically adopt a positive pace during the swim, a variable pace during the cycle (characterized by fluctuations in speed), and a reverse J-shaped pace (explained below) during the run (Figure 1). A tactic commonly observed during elite races is the contribution of minimal effort during cycling in order to maximize run performance, which could partly explain the highly variable cycling pace.32–34 Race dynamics may be further influenced by the strengths of the individual, who may adopt different pacing strategies in order to maximize performance of a specific discipline. However, the influences of specific tactics and pacing strategies adopted by individual athletes on other competitors and race dynamics in triathlons are yet to be examined. Although elite athletes are permitted to draft during the cycle discipline during selected races, age-group triathletes are not.10,11,32–34 For instance, in non-draft-legal events, athletes are required to maintain a specified distance behind the next competitor, and are given a specified time to pass the front athlete when overtaking. Specifically, a cycling distance of 7–12 m (draft zone) is maintained in the Ironman, and a passing time of 20–25 seconds is enforced once entering the draft zone. The overtaken athlete then has the same specified time to drop back out of the draft zone. This difference in drafting ruling could significantly affect pacing during triathlon, as cycling during non-draft-legal events is more similar to that of an individual time trial, as reflected by the stability of power output during flat cycling in a triathlon race.35,84 Under such conditions, race dynamics are likely to have less of an influence on cycling performance and, thus, athletes are likely to have a greater reliance on intrinsic control over pacing. However, it is also important to note that, despite the lack of drafting during the cycle discipline of some triathlon events, the race dynamics associated with mass-start events is still likely to influence pacing. Indeed, non-drafting athletes are still required to be positioned near the leading riders to be in contention for a high finish place. Furthermore, regardless of drafting format, all athletes are permitted to draft during the swim and run disciplines of all triathlon events. Collectively, the actions and tactics of competitors are likely to dramatically influence pacing during triathlon competition.

Figure 1

Example of pacing adopted by elite athletes during a draft-legal Olympic-distance triathlon.

Notes: (A) Swim. (B) Cycle. (C) Run.

Environmental factors

While race dynamics are likely to increase the variability of exercise intensity within both draft-legal and non-draft-legal races, pacing could also be influenced by other external factors such as water currents, wind conditions,35,39,85 topography,10,42,85 environmental heat, and humidity.86 For example, a higher mean power output of 192±21 W was reported during women’s world cup flat cycling races, as compared with 169±17 W observed during hilly races.87 Conversely, power output during the Tour de France increased with increasing hill gradient. Specifically, mean power outputs during flat, semi-mountainous, and mountainous stages were 218±21 W, 228±22 W, and 234±13 W, respectively.88 It is important to note that, when examining the pacing of athletes, the relationship between power output and speed is non-linear and could be vastly different; therefore, power output and speed should not be used interchangeably, especially when external factors such as fluid resistance39 and gravity42 vary. Indeed, in an attempt to maintain an even pace, a variable power output is sometimes necessary to account for periods of high external resistance such as traveling uphill42,89 or into a headwind.39,89 For instance, Atkinson and Brunskill39 and Atkinson et al89 demonstrated in laboratory conditions that, when compared with a freely-paced trial, superior cycling performance was achieved when athletes increased power output uphill or into a headwind and decreased power output when external resistance was low (ie, downhill and with a tailwind). However, no previous studies have examined the extent to which athletes adopt a varied distribution of power output in order to account for varying external conditions in actual triathlon competitions. Interestingly, a relatively even power output has been observed in well-trained triathletes during the cycle discipline of the Ironman35 and half-Ironman events (Wu, unpublished data, 2014; Figure 2) despite fluctuating wind conditions. In this case, it is possible that maintaining physiological homeostasis, minimizing neuromuscular fatigue, and conserving metabolic reserves for the run may be more important during ultra-distance triathlons than the time saved from varying power output to counteract the debilitating effects of wind on performance. Furthermore, the swim, cycle, and run disciplines of triathlons are performed at different velocities and with dissimilar external resistances. Such differences in the resistance to locomotion will influence the degree of variation in energy expenditure necessary to adopt optimal pacing strategies.

Figure 2

Power output and speed of a well-trained triathlete during the cycle discipline of a half-Ironman event.

Note: Fluctuations in speed compared to a relatively even power profile are of note.

During triathlon racing, pacing may be further influenced by heat and humidity.90,91 For instance, Peiffer and Abbiss86 observed an earlier decline in power output and subsequently lower mean power output during a 40 km time trial at 32°C, as compared with 17°C, 22°C, and 27°C. Despite the difference in performance, pacing strategies observed were remarkably similar, characterized by a gradual reduction in power followed by an increase toward the final stages of cycling.86 However, optimal pacing during triathlon in the heat will also depend on the abovementioned factors such as wind, topography, physiological characteristics, fitness, and rate of athlete’s heat dissipation. Further modeling research into the relationships between ambient temperature, power output, energy expenditure, drafting, and external resistance is required to predict the optimal pacing strategies during different triathlon distances.

Transition

One of the challenges in triathlon is in successfully maneuvering the swim-to-cycle transition (T1) and cycle-to-run transition (T2). Indeed, the presence of a preceding swim before the cycle and the cycle before the run can negatively impact physiological stress and performance. For instance, Kreider et al92 observed a 16.8% reduction in power output during 75 minutes of cycling after 800 m of swimming, compared with a control 75-minute cycle without prior swimming. Similarly, Guezennec et al51 observed a significant increase in oxygen cost (51.2 versus 47.8 mL · kg−1 · min−1) and heart rate (162 versus 156 bpm) during the 10 km run of an Olympic-distance triathlon following cycling, as compared with a control run without prior cycling. Certainly, differences in muscle groups utilized,3 energy expenditure, and requirements between T131,51,92 and T293 may contribute to the complexity of pacing within triathlon. The sections below aim to illustrate the influence of T1 and T2 transitions on triathlon pacing in further detail.

Swim to cycle

T1 consists of changing from a sport that is upper-body dominant to one that is predominantly lower-body dominant, which may be difficult due to blood pooling in the upper extremities after swimming.3 The detrimental effect of a prior swimming bout on subsequent cycling performance has been well documented.31,51,76,92,94,95 Indeed, when preceded with a 1,500 m swim (equal distance with an Olympic triathlon swim), an increase in energy cost (13% lower gross efficiency, 56% higher blood lactate concentrations, 9% higher heart rate, 5% higher VO2) was observed during the first 5 minutes of a 30-minute cycle at 75% maximal aerobic power.76 Nevertheless, the swim discipline within triathlon events is likely performed at a relatively high intensity, due to the importance of exiting the water in the lead group to form part of the first pack of cyclists, especially during draft-legal races.34,94,96 However, there is research to suggest that decreasing swim intensity would decrease subsequent fatigue and improve overall triathlon performance.31 Indeed, Peeling et al31 investigated the effect of swimming at 80%–85%, 90%–95%, and 98%–102% of the mean speed achieved during a control swim trial on subsequent cycle and run performance. They reported a faster cycling time when the swim was performed at 80%–85% and 90%–95%, as compared with 98%–102%.31 Further, a faster overall triathlon time was observed when swimming at 80%–85%, as compared with 98%–102%. This indicates that the pacing adopted during the swim of a triathlon not only affects subsequent cycling performance, but may also have an influence on overall triathlon performance. As adopting a low swimming speed may be counter-productive to overall triathlon performance, especially during drafting events, improving swimming ability may be crucial for athletes to achieve a high swim finishing position without exceeding 90% of their maximal swim speed.31

Cycle to run

The negative effects of a preceding cycle bout on running performance is well known (for a review see Millet and Vleck97). These effects have been attributed to an increase in oxygen cost,50–52,92 glycogen depletion,50–52 ventilatory muscle fatigue,50 dehydration,51,52,98 decreased pulmonary compliance, exercise induced hypoxemia,99 muscle fatigue,52,100 and redistribution of muscle blood flow. Novice athletes reportedly experienced loss of coordination,97 associated with changes in gait frequency (1.5–2.0 to 1–1.5 Hz in cycling and running, respectively) and/or transition from a predominantly non-weight-bearing activity (cycling) to one bearing two to three times of body mass (running),52,101 and a shift from primarily concentric to stretch-shortening-type eccentric contractions in running.102 Therefore, it appears that careful pacing manipulation during the cycle discipline could benefit subsequent run performance. Indeed, it has been suggested that athletes may be able to alter power output and cadence during cycling in order to maximize subsequent running performance.79,84,103,104 For instance, Bernard et al84 investigated the effect of pacing during a 20 km time-trial on subsequent 5 km running performance within the laboratory setting. The authors reported a significantly faster running performance (1,118±72 seconds) after a constant intensity cycle bout as compared with variable (1,168±73 seconds) and freely chosen intensity cycle bouts (1,134±64 seconds), suggesting that an even cycling pacing strategy is preferable during a sprint distance triathlon. In accordance with these findings, an increase in power output during the final stages of a 20 km cycling bout has been shown to be detrimental to 5 km running performance.84 As such, during a World Cup Olympic triathlon, Le Meur et al32 observed a significant decrease in power output in both elite males (17%, P<0.05) and females (19%, P<0.05) from the first (lap one) to the last lap (lap six); presumably this was an attempt to minimize fatigue and maximize subsequent running performance. Taken together, these studies suggest that, depending on the drafting nature and distance of the race, different combinations of pacing for specific disciplines of the triathlon are required for optimal performance.

It is plausible that the drafting nature of the race could affect pacing during triathlon transitions. During draft-legal races, an increase in speed is sometimes observed during T134 and T2,34,97 likely in an attempt to achieve a prime position for the subsequent discipline. Indeed, there is evidence to suggest a positive relationship between T2 and overall finishing position.97 However, there remains a paucity of research data that clarifies the effect of T1 and T2 on pacing during triathlon. There is, however, an abundance of research on the increase in speed during the final stages of a triathlon run. This sprint finish is widely referred to as an “end-spurt.”8,25,30,105 The end-spurt phenomenon is characterized by an increase in speed after 80%–90% completion of the race when the athletes are supposedly most fatigued, and may be greater during draft-legal races where race outcomes are more likely decided by a sprint finish. The end-spurt has been associated with an increased central drive, provided that there is sufficient metabolic reserve25 and that the end-spurt poses no catastrophic risk to physiological homeostasis.8,105 Yet, the maintenance of an overly large reserve capacity toward the end of the event could be evocative of a suboptimal pacing strategy, where a higher running velocity could have been maintained for a longer duration of the run discipline.15,38,45,85 Therefore, the ability to manipulate run pacing such that energy stores are optimally utilized could highly influence triathlon performance. There is evidence to suggest that the run in a non-draft-legal triathlon is performed without an end-spurt. Taylor et al106 compared the pacing strategy adopted during a sprint triathlon run, as compared with a control 5 km run, and found no differences in run pacing. Further, no increase in running speed was observed in the final kilometer. It is, however, important to note that laboratory simulated triathlons are usually performed as isolated trials and may not replicate the psychosocial factors and peer influence experienced during actual races. Further research is required to elucidate pacing strategies adopted during non-drafting triathlons.

Biological sex

The influence of internal and external factors on pacing may differ between males and females, due to internal (physiological and morphological107) and external (participation rates108) differences between sexes (for review see Lepers et al107). The physiological differences have been attributed to a 12%–15% lower maximal oxygen consumption (VO2max),109 5%–10% lower hemoglobin concentration, and ∼8% higher body fat percentage in elite females, as compared with elite males.107,110,111 Le Meur et al32 observed that, during a World Cup Olympic-distance triathlon, females spent a greater percentage of total time (45%) above maximal aerobic power when cycling up hills, as compared with males (32%). Females also tended not to bridge gaps that were formed during the cycle section.33 These results are in accordance with those previously reported in events based on a single mode of locomotion, such as a cross-country mountain bike World Championships.82 It is speculated that the lower maximal aerobic power-to-weight ratio in females results in greater time spent on uphill sections of the race.32 Since it has been previously shown that better cycling performance results from minimizing time spent traveling uphill89 and adopting a more even distribution of speed, it appears that females could benefit from improving uphill riding performance or maximal aerobic power/weight ratio.32,82 Collectively, research on the pacing differences in cycling/triathlon between sexes suggests that external factors such as topography and wind are likely to be more detrimental to performance in females than in males.32,33,45,82,112 As such, it is possible that drafting may benefit females more than males in triathlon and cycling performance. This could alter the pacing strategies that are adopted based on the drafting nature of the race. Indeed, there is evidence to suggest that males typically begin the swim33 and cycle disciplines32 of draft-legal Olympic races with a relatively more “aggressive” initial starting pace, as compared with females. However, despite the aforementioned physiological and morphological differences between sexes, and the weight bearing nature of running,45,107 male and female triathletes typically adopt similar positive pacing during draft-legal Olympic-distance triathlon runs.11,32 To date, the benefit of this positive pacing strategy remains unclear.11,32

It is possible that different pacing strategies could be observed with males and females during non-drafting triathlons, due to the various extents of influence of race dynamics107 and peer influence (participation rates and depth of field)108 between sexes. For example, the sex differences in participation are generally higher in non-drafting triathlons compared with draft-legal elite races. This difference has been shown to account for more than one-third of the differences in sex variations in marathon running.113 Therefore, further research is warranted to elucidate the specific differences in influence of physiological, sociological, and possibly psychological factors on pacing between sexes.

Age

The decline in endurance performance due to advancing age is well-known.102,107,110,114,115 The sex gap for the age-related declines in endurance performance remains at 10%–15% until 50 years of age, and progressively widens thereafter.107,110 The gap between sexes for overall triathlon performance ranges from 12% to 18.2%;107 however, there is a similar widening between sexes after 55 years in the Olympic and Ironman triathlons.110,116 Further, the duration of triathlon also influences the age-related declines in triathlon performance. Specifically, more pronounced declines have been observed during shorter-distance triathlons, as compared with the Ironman.107 The age-related decline in triathlon performance has been attributed to: a reduction in VO2max;107,114 a reduction in muscle strength and mass; attenuation of repair and hypertrophy of skeletal muscles;117 lower resting muscle glycogen content;13 and a reduction in training volume and intensity.107 As these factors play a critical role in fatigue development, aging can influence the distribution of self-selected pace by triathletes of different age-groups. However, there is no research to date on the influence of age on pacing in triathlon. Nevertheless, previous research has demonstrated the effect of age on pacing during prolonged endurance exercise.118 Specifically, March et al118 investigated the age-related changes in pacing of 319 finishers during a marathon, and reported that women and older athletes adopted a relatively more-even pacing as compared with men and younger athletes. Certainly, the complexity of triathlon and the increase in oxygen cost during running in a triathlon51 indicates that pacing during triathlon could alter with age.

Influence of performance throughout individual disciplines on subsequent triathlon performance

Due to the importance of discipline-specific and overall pacing in triathlon, it is important to note the relationships between each individual discipline and overall performance during pacing. Indeed, the performance during each discipline can influence overall triathlon performance to different extents.12,31–34,51,52 Therefore, the interdisciplinary associations are highlighted below. Performance during the swim discipline of a triathlon is likely to be of greater importance within shorter distance triathlons (sprint and Olympic), as compared with longer distance triathlons (half-Ironman and Ironman). Indeed, evidence indicates that the position achieved during the first 200–500 m in a 1,500 m Olympic-distance triathlon swim largely determines the final swim outcome.32–35 Supporting this, a significant correlation (r=0.99 and 0.97 for males and females, respectively) was observed between the position attained at 350 m into a swim of 1,500 m and final swim outcome during a World Cup Olympic-distance competition.32 In a separate World Cup race, the final swim position and velocity was reported to correlate with overall race position (r=0.44 and −0.52, respectively).34 Furthermore, there is evidence to suggest that decreasing swim intensity could benefit subsequent cycling and, therefore, overall sprint-triathlon performance.31 This indicates that the pacing strategy adopted during the swim of a triathlon not only affects subsequent cycling performance, but may also have a global influence on overall triathlon performance.

During longer half-Ironman and Ironman distances, however, the percentage contribution of swimming to total race time is considerably lower (∼10% and ∼20% during longer and shorter distances, respectively).4 Hence, it has been suggested that performance in the swim portion in longer triathlon events is not tantamount to overall performance.119 Since the majority of field studies have been based on the Olympic distance, further studies on the pacing strategies adopted during the swim on subsequent triathlon performance in other distances, in both draft-legal and non-draft-legal races, are warranted.

In draft-legal racing, athletes commonly complete the cycling discipline in packs (pelotons). The effect of pacing and performance during cycling appears to affect overall performance more than swimming due to the high correlation of overall winners belonging to the first-finishing cycling pack. Indeed, a higher correlation was observed between cycle and overall performance (r=0.52–0.74) as compared with swimming and overall performance (r=0.36–0.52)32–34 during draft-legal Olympic-distance triathlons. Presently, there is a lack of research on the effect of pacing during swim and cycle bouts on subsequent running performance.51,106 Further research is warranted to determine the pacing strategies required for optimal pacing in disciplines prior to the run in a triathlon.

The run discipline of a triathlon is completed last, following considerable energy expenditure in the swim and cycle disciplines. Additionally, the full weight-bearing nature of running89,95 is likely to induce more muscular fatigue and damage when compared with swimming and cycling. Certainly, a decrease in running efficiency after a previous cycling bout51,52 suggests higher metabolic costs, which are likely to induce fatigue and affect performance. This is especially critical in triathlons where energy efficiency could play a significant role in determining performance.120 Thus, it seems reasonable to postulate that performance during the run discipline would have the greatest influence on overall race position. Indeed, the highest correlations have been reported between running performance and final race position (r=0.71–0.99) as compared with other disciplines during the Olympic-distance triathlon.32–34 However, further research is needed to determine the relationships between the three individual disciplines and overall standings during other triathlon distances such as the sprint, half-Ironman, and Ironman.

A reverse J-shaped pacing strategy is commonly observed during the elite 10 km Olympic-distance triathlon run.11,32–34 The reverse J-shaped pacing is characterized by a fast start followed by a reduction in speed and a subsequent increase in speed toward the end of the run (for review see Abbiss and Laursen5). It is currently unclear whether a reverse J-shaped pacing is optimal for the triathlon run. However, due to the importance of run performance relative to overall performance, it is highly likely that optimizing run pacing could benefit triathlon performance. To date, only a single study has investigated the effect of manipulating run pace on overall triathlon performance. Hausswirth et al12 investigated the effect of altering the first run kilometer (by 5% slower, 10% slower, and 5% faster than the mean speed during a control run) during a simulated Olympic-distance triathlon in ten highly trained male triathletes. They reported significantly faster 10 km performance in the −5% condition, as compared with the +5% and −10% condition. A slower run performance observed in the +5% trial was attributed to the downregulation of exercise intensity in a feedforward manner controlled by the brain, to prevent catastrophic failure of the physiological systems.12 Further, a faster start requires higher metabolic demands,6,20,121,122 which could result in premature fatigue prior to the end of the race, as demonstrated by the +5% run. As such, a slightly slower to even start, which resultantly elicits a more even overall run pace, may be ideal for an Olympic-distance triathlon run. However, further research is required to determine the optimal run pacing strategies during other triathlon distances.

Although there is an extensive amount of triathlon research, no studies have systematically investigated the effect of distance, age, and sex on triathlon pacing. Further modeling research is required to establish the relationships between pacing during individual disciplines and overall performance across different triathlon distances.

Conclusion

The understanding of pacing in multisport events such as the triathlon is still poorly understood. The manipulation of pacing in triathlon is complex, due to the sequential swim, cycle, and run disciplines, and is attributed to the fatigue that accumulates between disciplines. Furthermore, triathlon pacing can be influenced by a multitude of intrinsic and extrinsic factors including wind velocity, topography, influence of other competitors, transition, age, drafting, biological sex, and duration of event. It appears that a reduced intensity in prior swimming and cycling could result in faster subsequent cycling and running performance, respectively. However, the optimal pacing strategies across the sprint, Olympic, half-Ironman, and Ironman triathlons are currently unclear. Further research is required to establish the best possible pacing strategies to adopt across various triathlon distances.