According to their length, races in ice speed skating are classified in sprint (500 and 1,000 m), middle (1,500 m), and long-distance events (3,000, 5,000, and 10,000 m). Covering a given distance in the shortest possible time is the objective in any speed skating competition. To reach this goal, it is necessary to distribute energy expenditure and the associated skating velocity during the respective races in an optimal manner (10,11).
For long-distance competitions (i.e., races lasting longer than 80-120 seconds), it is recommended to distribute skating velocity equally (4,13) and to make use of a “negative pacing” or “negative pacing strategy” (6,7,14,21). That is, when there is an increase in speed observed over the duration of the event (2,14). From a physiological point of view, the advantage of a “negative pacing strategy” compared with other pacing patterns is thought to improve performance by avoiding excessive early energy contribution from the anaerobic system (21), avoiding costly oxygen debt early in the race (14) or limiting accumulation of fatigue-related metabolites or all (1,19,22).
To the authors' knowledge, there are no studies available whether or not a “negative pacing” is adopted during competitive long-distance events in ice speed skating. According to a previous study in cycling, where road cyclists performed time trials over different distances on a bicycle ergometer, an increase in velocity in the second half of the race was only shown for the 3,000 m (8). On the contrary, no increase in velocity could be determined while examining shorter race distances over 500, 1,000, and 1,500 m. Another study by Mattern et al. (19) showed that compared with self-paced trials, a “negative pacing strategy” resulted in a significantly lower blood lactate concentration during the initial 9 minutes of a 20-km cycling time trial. Furthermore, the relatively low enforced starting power output resulted in significant improvements in overall performance. In addition, studies on pacing strategies adopted in several long-distance running events indicate that athletes reduce their speed after the start to an extent, which can be performed constantly followed by an attempt to increase speed during latter race sections (20,25).
The purpose of the present study was to describe the pacing patterns adopted by female and male elite skaters during long-distance World Cup races and to discuss possible factors influencing the self-selection of such patterns. More specifically, the analysis of a series of competitions performed during a complete World Cup season offered the unique opportunity to characterize the effects of athletes' performance level, gender, and rink location on the way of distributing race time during several long-distance events. The results from previous research suggest that velocity in long-distance events should be distributed equally along with an increase in speed during latter race parts. Therefore, we assume that performance achieved during prolonged speed skating races would benefit from a constant velocity distribution followed by an increase in pace toward the end of the race.
Understanding the behavior of elite athletes during official competitions is important for a number of applied and theoretical reasons. From a practical perspective, detecting differences in the pacing pattern between skaters of different calibers or in races held on different altitudes may provide some guidance for the design of training programs or both. From a theoretical point of view, knowledge concerning the optimal distribution of skating velocity during long-distance events may be helpful to specify and extend existing pacing strategies.
Experimental Approach to the Problem
To test our hypothesis, final and lap times performed during Essent World Cup Speed Skating Competitions 2008/2009 season organized by the International Skating Union were analyzed. The use of official split times has the advantage that they are derived from real competition scenarios, which are free of experimental manipulation (e.g., fixed exercise intensity or trial duration or both). The analysis of data from a whole season offers the unique opportunity to compare races of different lengths and different locations. During 2008/2009 season, a total of 6 competitions were held for the long distances. Two of these included women's 5,000 m and men's 10,000 m races, and 4 of these included women's 3,000 m and men's 5,000 m races. Except one World Cup (Salt Lake City), all competitions were held at low altitude. All races took place at an indoor ice rink (400 m oval).
Between 38 and 44 female and 37-45 male skaters took part in each race depending on race distance. All data were downloaded from publicly accessible official skating websites (www.isu.org); therefore, informed consent was not obtained from athletes for use of this information. Approval of the study was obtained by the Ethics Committee Beider Basel, Switzerland.
For each race, absolute lap times to distance plots were generated by comparing athletes' performance level, sex, and rinks' location. For athletes' performance level, skaters were divided in a top-ranked (include the top 10 skaters in the respective race) group and a bottom-ranked (include the other skaters in the respective race) group. For rinks' location, pacing patterns adopted at low-altitude races (<1,000 m) were compared with those performed at high altitude (>1,000 m), for women's 3,000 m and men's 5,000 m separately. Because women and men performed races over 5,000 m, direct gender comparisons were made additionally.
Separate analyses of variance, with repeated measures on lap using lap time as dependent variable were used to investigate differences in the distribution of skating time for athletes' performance level (top-ranked vs. bottom-ranked athletes), gender (women's vs. men's 5,000 m), and rinks' location (low vs. high altitude). Post hoc tests were conducted to identify statistically significant differences between populations and between the laps, with alpha adjusted using Bonferroni correction. The opener was excluded from all analyses. All analyses were performed using Statistical Package for Social Sciences version 16.0. The alpha level for significance was set at p ≤ 0.05. Values are shown as mean ± SD.
The mean total times were 4:06.18 ± 3.08 seconds and 4:19.15 ± 7.48 seconds in women's 3,000 m, 7:04.85 ± 5.23 seconds and 7:24.73 ± 8.53 seconds in women's 5,000 m, 6:29.31 ± 49.08 seconds and 6:48.39 ± 38.75 seconds in men's 5,000 m, and 13:14.84 ± 8.17 seconds and 13:53.45 ± 17.53 seconds in men's 10,000 m for top- and bottom-ranked skaters, respectively. After a fast opener, female and male speed skaters slowed down throughout the 3,000 m or 5,000 m or both races (Figures 1A-C). For women's 3,000 m, differences between successive lap times were significant in each instance (p < 0.001). For women's 5,000 m, analysis indicated times shortest on laps 1 and 2 (≈ 33.41 seconds) with laps 3 and 4 not different from each other but faster than laps 5-12 (all p < 0.001), where laps 11 and 12 did not differ from each other. For men's 5,000 m, analysis detected times also shortest on laps 1 and 2 (≈ 30.74 seconds) with laps 3-5 and 6 and 7 not different from each other but faster than laps 8-12 (all p < 0.001), where laps 10-12 did not differ from each other. In men's 10,000 m, skaters showed a slow opener before they adopted a relatively even pace (Figure 1D). Here, male skaters performed the first lap fastest (31.76 ± 0.90 seconds; p < 0.001) with laps 2-7 not different from each other but faster than laps 19-24 (all p < 0.05), which did not differ from each other. Irrespective of race distance, comparison of pacing patterns for athletes' performance level showed that top-ranked female and male skaters performed faster lap times at each lap compared with bottom-ranked skaters (p < 0.001 for all races). In general, the pacing pattern of the recent world record holders was not different from that of the other skaters. A difference was only observed in men's 10,000 m race, where the recent world record holder was able to increase his pace for the last 5 laps (Figure 1D).
Lap times to distance plots comparing women's and men's 5,000 m races are shown in Figure 2A. Male skaters completed the races significantly faster than female skaters (6:43.53 ± 42.29 vs. 7:18.97 ± 11.90 seconds; Δ = 35.44 seconds; p < 0.001). As before, differences between successive lap times were significant with shortest times performed on laps 1 and 2 and with laps 3-5 not different from each other but faster than laps 8-12 (all p < 0.001), where laps 11 and 12 did not differ from each other. Comparison of pacing patterns for athletes' sex showed that male skaters performed each lap faster than female skaters (p < 0.001).
Lap times to distance plots comparing races performed at low vs. high altitudes for women's 3,000 m and men's 5,000 m events are shown in Figure 2B. Total race time at high altitude was significantly shorter compared with that at low altitude, in women's 3,000 m (4:05.35 ± 4.16 vs. 4:17.10 ± 8.23 seconds; Δ = 11.75 seconds; p < 0.001) and in men's 5,000 m races (6:22.21 ± 8.16 vs. 6:46.65 ± 44.34 seconds; Δ = 24.44 seconds; p < 0.015). As before, the analysis detected that shorter lap times during early but longer lap times during later race sections were obtained. In women, differences between successive lap times were significant in each instance (p < 0.001). In men, shortest time was obtained for lap 1 (30.08 ± 0.83 seconds) with laps 2-5 not different from each other but faster than laps 6-12 (all p < 0.001), where laps 10-12 did not differ from each other. For both sexes, the analysis indicated that at high altitude, skaters performed shorter lap times at each lap compared with that at low altitude (p < 0.001).
The purpose of the present study was to analyze the pacing pattern adopted by women and men during a series of official long-distance World Cup ice speed skating races. In this context, the distribution of skating time was compared for athletes' performance level, gender, and rinks' altitude. The main findings were (a) that during all long-distance races, elite speed skaters adopted a similar pacing strategy by starting fast before pace gradually declined throughout the duration of the race-“positive pacing”; (b) that top-ranked compared with bottom-ranked female and male skaters performed faster lap times; (c) that during 5,000 m races, male skaters showed shorter lap times than female skaters; and (d) that at high altitude, female and male skaters achieved shorter lap times than at low altitude.
Overall, the results of this study showed that irrespective of race aspect considered (performance level: world record holder, top-, and bottom-ranked skaters; gender: women, men; rinks' location: low and high altitudes), skaters adopted a similar “positive pacing strategy,” whereby after the opening section, pace progressively decreased throughout the duration of the events in a more (3,000 and 5,000 m races) or less (10,000 m races) pronounced way. This finding is consistent with observations reported in speed skating handbooks (7) and with the studies on pacing strategies (5,17), which maintain that during long-distance events, a constant speed along with a decrease in speed during latter race parts should be performed. In fact, the observed slowing throughout the races indicates the presence of fatiguing processes. For example, Kindermann and Keul (16) were able to show that after finishing long-distance races, female and male ice speed skaters averaged values between 11 and 15 mmol/L lactate in the arterial blood, see Ref. (3) for comparable findings).
On the other hand, a “negative pacing strategy” is recommended for prolonged races of more than 2 minutes (2,14,21), but the adoption of a “positive pacing strategy” was observed in the present study. The reason for this divergence is unclear but may be because of differences in the race format being performed. Studies that propagate that an increase in speed during late race sections is optimal for the achievement of a short total race time investigated in cycling (8,19), running (22), or swimming (24) events. These types of sport require that the time of the winning athlete be only marginally shorter than that of other competitors in the same race (head-to-head competitions). Thus, other athletes may influence the pacing strategy. In contrast, in the ice speed skating races described in the present study, the overall time of the winner needs to be faster in absolute terms (racing against the clock) because the best athletes are not directly competing against each other. According to this, energy that is left at the finish is useless. As a consequences, ice speed skaters may self-select a pacing strategy that does not allow to hold some energy back.
A slowing down in lap times was found for the top- and bottom-ranked skaters. However, top-ranked athletes were faster at each lap compared with the bottom-ranked athletes. Physiological or technical or both factors could be stated as one reason for these differences. Foster and Thompson (12) reported that there is in general, a step-down in maximal oxygen consumption from better skaters to less accomplished skaters. Also, van Ingen Schenau et al. (28) showed that elite ice speed skaters had a significantly higher oxygen consumption and external power output compared with well-trained skaters. Furthermore, their analysis of skating technique revealed that elite speed skaters were using a higher stroke frequency and a smaller preextension angle, offering a better skating position. Differences in years of training or experience in competitions are also mentioned as possible reasons (15,23). For example, faster skaters may have more competitive experience and may therefore show a pacing pattern closer to their optimum for the given distance.
That male compared with female skaters showed smaller lap times could also be because of physiological or technical or both differences. From a technical point of view, female skaters showed a larger preextension knee angle than male skaters resulting in a more inappropriate skating position. As a result, females achieve a lower amount of work per stroke during race (27). A further disadvantage associated with a larger preextension angle is the fact that at the same speed, the amount of loss of skating velocity to air friction is also higher. From a physiological perspective, female skaters showed a higher postexercise lactate level compared with male skaters (18), suggesting that female athletes have less resistance to fatigue. Furthermore, differences in variation of skating velocity and the associated rate of lactate accumulation (9,12) might also provide an explanation for the differences in lap times.
The observed overall pacing pattern adopted during low- and high-altitude events was the same (i.e., reduction in lap times after the initial acceleration phase); however, female and male skaters performed shorter lap times at high compared with low-altitude rinks, resulting in significantly shorter total race times (−4.6% for women and −6.0% for men), which is in good agreement with that reported by van Ingen Schenau et al. (29). A possible explanation could be that at high altitude, the air pressure and therefore the air resistance decreases (26), so that faster times may be expected than at low altitude. Because only one World Cup meeting was performed at high altitude (Salt Lake City) during season 2008/2009, we were only able to compare the pacing pattern performed here with that adopted during the other 3 World Cups, which were conducted at low altitude. From our view, this is a preliminary finding and further research should reveal whether rinks' altitude influences the pacing pattern and the performance outcome during ice speed skating long-distance races.
This study involves the use of official lap times. They have the advantage of being derived from real competition scenarios, which are free of experimental manipulation, like fixed exercise intensity, fixed duration trial, or prescribed pacing behavior or all. On the other hand, they only enable a rough characterization of an athlete's overall pacing pattern and do not provide detailed insight into the distribution of velocity throughout the event. This may be less of a problem in a discipline like ice speed skating, whose competitions are mostly carried out indoors. Thus, there are only few changes in the physical characteristics of the racing environment (e.g., changes in temperature, humidity, and so on). Consequently, lap times within track races may be used as relatively accurate indicators of pacing patterns, see also Refs. (20,25).
In conclusion, skaters observed during a series of long-distance World Cup races adopted a pacing pattern, where after initial acceleration, a progressive slowing down occurred-“positive pacing strategy”. The similarity of this pattern between both sexes, athletes' performance level, and rinks' altitude suggests that any improvements in total race time should be made without disrupting the overall distribution of time. Moreover, the faster lap times for top-ranked compared with bottom-ranked skaters and for male compared with female skaters indicate that technical or physiological or both aspects rather than pacing pattern need to be developed in those. The shorter lap times at high compared with low-altitude races suggest an important role of rinks' location for performance outcome, at elite level.
1. Abbiss, CR and Laursen, PB. Models to explain fatigue during prolonged endurance cycling. Sports Med
35: 865-898, 2005.
2. Abbiss, CR and Laursen, PB. Describing and understanding pacing strategies during athletic competition. Sports Med
38: 239-252, 2008.
3. Astrand, PO, Rodahl, K, Dahl, HA, and Stromme, AB. Textbook of Work Physiology: Physiological Bases of Exercise
(4th ed.). Champaign, IL: Human Kinetics, 2003.
4. Chaffin, ME, Berg, K, Zuniga, J, and Hanumanthu, VS. Pacing pattern in a 30-minute maximal cycling test. J Strength Cond Res
22: 2011-2017, 2008.
5. de Koning, JJ, Foster, C, Lampen, J, Hettinga, F, and Bobbert, MF. Experimental evaluation of the power balance model of speed skating. J Appl Physiol
98: 227-233, 2005.
6. de Koning, JJ and van Ingen Schenau, GJ. Performance-determining factors in speed skating. In: Biomechanics in Sport: Performance Enhancement and Injury Prevention. Olympic Encyclopaedia of Sports Medicine
. Vol 9. Zatsiorsky, VM, ed. Oxford, United Kingdom: Wiley-Blackwell Science, 2000. pp. 232-246.
7. Foster, C and de Koning, JJ. Physiological perspectives in speed skating. In: Handbook of Competitive Speed Skating
. Gemser, H, de Koning, JJ, and van Ingen Schenau, GJ, eds. Leeuwarden, the Netherlands: Eisma Publishers, 1999. pp. 117-137.
8. Foster, C, de Koning, JJ, Hettinga, F, Lampen, J, Dodge, C, Bobbert, M, and Porcari, JP. Effect of competitive distance on energy expenditure during simulated competition. Int J Sports Med
25: 198-204, 2004.
9. Foster, C, de Koning, JJ, Rundell, KW, and Snyder, AC. Physiology of speed skating. In: Exercise and Sport Science
. Garrett, WE and Kirkendall, DT, eds. Philadelphia, PA: Lippincott Williams & Wilkins, 2000. pp. 885-893.
10. Foster, C, Schrager, M, Snyder, AC, and Thompson, NN. Pacing strategy and athletic performance. Sports Med
17: 77-85, 1994.
11. Foster, C, Snyder, AC, Thompson, NN, Green, MA, Foley, M, and Schrager, M. Effect of pacing strategy on cycle time trial performance. Med Sci Sports Exerc
25: 383-388, 1993.
12. Foster, C and Thompson, NN. The physiology of speed skating. In: Winter Sports Medicine
. Casey, MJ, Foster, C, and Hixson, EG, eds. Philadelphia, PA: F. A. Davis Company, 1990. pp. 221-240.
13. Ham, DJ and Knez, WL.An evalution of 30-km cycling time trial (TT30) pacing strategy through time-to-exhaustion at average TT30 pace. J Strength Cond Res
23: 1016-1021, 2009.
14. Holum, D. The Complete Handbook of Speed Skating
. Hillside, NJ: Enslow Publishers, 1984.
15. Hopkins, WG and Hewson, DJ. Variability of competitive performance of distance runners. Med Sci Sports Exerc
33: 1588-1592, 2001.
16. Kindermann, W and Keul, J. Anaerobic supply of energy in high-speed skating. Dtsch Z Sportmed
31: 142-147, 1980.
17. Kuhlow, A. Running economy in long-distance speed skating. In: Biomechanics V-B
. Komi, PV, ed. Baltimore, MD: University Park Press, 1976. pp. 291-298.
18. Maksud, MG, Wiley, RL, Hamilton, LH, and Lockhart, B. Maximal VO2
, ventilation, and heart rate of Olympic speed skating candidates. J Appl Physiol
29: 186-190, 1970.
19. Mattern, CO, Kenefick, RW, Kertzer, R, and Quinn, TJ. Impact of starting strategy on cycling performance. Int J Sports Med
22: 350-355, 2001.
20. Noakes, TD, Lambert, M, and Human, R. Which lap is the slowest? An analysis of 32 world record performances. Br J Sports Med
43: 760-764, 2009.
21. Publow, B. Speed on Skates
. Champaign, IL: Human Kinetics, 1999.
22. Robinson, S, Robinson, D, Mountjoy, RJ, and Bullard, RW. Influence of fatigue on the efficiency of men during exhausting runs. J Appl Physiol
12: 197-201, 1958.
23. Stewart, AM and Hopkins, WG. Consistency of swimming performance within and between competitions. Med Sci Sports Exerc
32: 997-1001, 2000.
24. Thompson, KG, MacLaren, DP, Lees, A, and Atkinson, G. The effect of even, positive and negative pacing on metabolic, kinematic and temporal variables during breaststroke swimming. Eur J Appl Physiol
88: 438-443, 2003.
25. Tucker, R, Lambert, MI, and Noakes, TD. An analysis of pacing strategies during men's world-record performances in track athletics. Int J Sports Physiol Perform
1: 233-245, 2006.
26. van Ingen Schenau GJ, de Boer RW, and de Groot G. Biomechanics of speed skating. In: Biomechanics of Sport
. Vaughan, CL, ed. Boca Raton, FL: CRC Press, 1989. pp. 121-167.
27. van Ingen Schenau, GJ and de Groot, G. On the origin of differences in performance level between elite male and female speed skaters. Hum Mov Sci
2: 151-159, 1983.
28. van Ingen Schenau, GJ, de Groot, G, and Hollander, AP. Some technical, physiological and anthropometrical aspects of speed skating. Eur J Appl Physiol
50: 343-354, 1983.
29. van Ingen Schenau, GJ, de Koning, JJ, and de Groot, G. A simulation of speed skating performances based on a power equation. Med Sci Sports Exerc
22: 718-728, 1990.