Key performance indicators of Olympic windsurfers during a World Cup: RS:X class®

ABSTRACT

The aim of this study was to analyse differences in velocity, distance travelled and manoeuvres performed by Olympic sailors of the RS:X class using a GPS device. Fifty-three Olympic sailors of the RS:X class (28 males and 25 females) who competed in a World Championship were selected. The sample was divided into tertiles (T1, T2 and T3) according to their classification in the competition. Using a GPS device during the competition, mean velocity (VM), velocity made good (VMG), manoeuvres and distances in three different courses (upwind, reaching and downwind) were assessed during a regatta. Significant differences were found based on performance level in upwind (p < 0.001; ${\eta }_p^2$ = 0.288), sailors of T1 covering a shorter distance compared to those of T2 (p < 0.009) and T3 (p < 0.001). Regarding VMG, an effect was observed for performance level in upwind (p < 0.001; ${\eta }_p^2$ = 0.718), reaching (p < 0.001; ${\eta }_p^2$ = 0.469) and downwind (p < 0.001; ${\eta }_p^2$ = 0.575). Females covered a shorter distance compared to the males in upwind (p < 0.001; ${\eta }_p^2$ = 0.639) and downwind (p < 0.001; ${\eta }_p^2$ = 0.903). Distance and VMG are significant variables for establishing differences in performance level among Olympic sailors of the RS:X class when the wind speed is in a range of 8–21 knots.

KEYWORDS:

• Tactic
• Olympic
• sport performance
• windsurfing
• GPS

1. Introduction

Windsurfing is a sport that combines components of surfing and sailing (Andrianopoulos & Vogiatzis, 2017). The first appearance of this sport in the Olympic Games was in Los Angeles in 1984, and it was consolidated as an Olympic discipline in the games of Seoul (1988) and Barcelona (1992) (Chamari et al., 2003). Among the different windsurfing classes, the Neilpryde RS:X© class is the only one included in the Olympic Games. This class was proposed by the International Sailing Federation (ISAF) and by the International Olympic Committee as an Olympic board in 2006, replacing the Mistral One Design© class.

In the Olympic competition format for the RS:X class, the main objective is to cover a course established by buoys in the shortest possible time, in which the sailors have to navigate in upwind and downwind courses. Athletes in this class are characterised by high levels of isometric strength and resistance to fatigue in the flexor muscles of the upper and lower limbs and trunk extensors (Guevel et al., 2000; Jablecki & Garner, 2000; Wu et al., 2016). In addition, since 2016 the Olympic class has changed from the Mistral One Design® to Neilpryde RS:X®. With this change, pumping actions are allowed by the ISAF, which increases the physiological demands, including heart rate (from 110 to 165 beat · min⁻¹) and VO₂ (from 19.2 to 48.4 ml · kg · min⁻¹) (De Vito et al., 1997; Vogiatzis et al., 2002).

In the regatta, the physiological demands have been quantified in values of 70.0–80.0% of VO_2max at low and moderate wind velocity (Castagna et al., 2008; Guevel et al., 2000; Jaszczur-Nowicki, 2004). However, this can be affected by different factors, such as wind velocity, leg, gender and performance level. Thus, at high wind velocity (9–11 m · s⁻¹), higher VO₂ (assessed directly using a portable gas analyser) has been found in the downwind compared to the upwind (84.6 ± 4.5% vs. 61.9 ± 8.7 VO_2max), whereas, at low wind velocity (2–4 m · s⁻¹), the VO₂ is higher in the upwind (82.6 ± 3.2 vs. 61.9 ± 8.7% VO_2max) (Castagna et al., 2008). Finally, it should be noted that intensity (measured by heart rate) is directly related to performance; thus, there is a strong correlation (r = −0.70) between % heart rate max (HR_max) and the position obtained in a race at low wind velocities (Chamari et al., 2003).

Technical and tactical performances in a windsurfing regatta are determined by mean velocity (VM) and velocity made good (VMG), as well as by distance travelled and the number of manoeuvres performed throughout the course (Anastasiou et al., 2019; Hagiwara & Ishii, 2016). Board velocity depends on the technical mastery of the sailor in each of the courses covered during the regatta (Guevel et al., 2000). In a regatta, since the routes are simple and well defined, VMG on the upwind and downwind (reaching and downwind) courses are the most interesting to evaluate performance (Day, 2017). VMG is the term used to determine the performance of the board in relation to the maximum velocity (in knots above the ground) that it can reach during the trip, considering the vertical component of the velocity with respect to the wind direction (Pluijms et al., 2016). When analysing the velocity according to the leg, it is observed that the velocity reached in reaching and downwind (downwind) is higher compared to that reached in upwind (Caimmi & Semprini, 2017). Thereby, in downwind courses, the board exceeds the wind velocity and creates its own apparent wind, resulting in a greater velocity vector (Castagna et al., 2008). Studies have shown that the best-ranked sailors had a higher average velocity and VMG and, based on gender, men have higher velocity values compared to females (Anastasiou et al., 2019; Hagiwara & Ishii, 2016). In terms of, the highest-level sailors covered a shorter distance to complete the regatta, and males covered a greater distance than females in upwind (Anastasiou et al., 2019; Hagiwara & Ishii, 2016).

The advances in technology devices such as GPS have been applied in various sport studies, demonstrating a good validity and reliability for measuring velocity, distances and manoeuvres (Anastasiou et al., 2019; Aughey, 2011; Caimmi & Semprini, 2017; Jennings et al., 2010).

Currently, most of the research carried out has focussed on quantifying the physiological requirements in the RS:X class and their differences in performance, as well as on describing their association with the intensity of technical and tactical variables such as velocity, distance travelled and manoeuvres performed in a regatta. However, the differences between technical and tactical aspects with performance in competitive-level sailors have not been studied to date. Therefore, the aim of our study was to identify, in Olympic sailors of the RS:X class, the effects of level of performance and sex on the velocity, distance travelled and manoeuvre variables performed during an actual regatta.

2. Methods

2.1. Participants

The study sample consisted of 53 Olympic sailors (28 males and 25 females) of the RS:X windsurfing class, with an age range between 19 and 38 years, who competed in the final of the RS:X class World Cup held in 2019 in Marseille (France). The ethical approval and the written/informed consent from all participants were not necessary because the sailors’ data were obtained from a publicly accessible website. The data of the sailors were collected from World-Sailing® (Caraballo et al., 2021). According to the final results of the regatta, the sample was divided into three performance groups, per sex (Table 1), classified based on the position occupied by the athletes in the competition, differentiating the best athletes as those who were within the first tertile (T1), a second group of performance constituted by the athletes within the second tertile (T2) and a third group with those in the third tertile (T3).

Table 1. Data of mean velocity, distance, manoeuvres and velocity made good (VMG) in upwind, reaching and downwind in athletes of different performance level and sex.

		Upwind				Reaching	Downwind
Variable	Sex	T1	T2	T3	T1	T2	T3	T1	T2	T3
Distance (m)	Females	36,380 ± 610 *, ^A	38,183 ± 1105 *	38,573 ± 1624 *	9136 ± 78	9972 ± 222	9180 ± 105	22,860 ± 597 *	22,709 ± 249 *	23,322 ± 1087 *
	Males	39,750 ± 782 *, ^B	40,410 ± 1064 *, ^C	42,029 ± 1551 *	9814 ± 120	9915 ± 73	9768 ± 91	26,801 ± 597 *	27,470 ± 615 *	27,017 ± 628 *
	Total	38,164 ± 1863^A	39362 ± 1554^C	40,392 ± 2349	9495 ± 363	9942 ± 471	9489 ± 316	24,947 ± 2144	25,229 ± 2493	25,267 ± 2077
Manoeuvres	Females	55.4 ± 3.3	55.9 ± 8.4	61.4 ± 7.6	4.8 ± 2.3	7.6 ± 3.7 *	6.0 ± 1.6 *	62.0 ± 16.3	62.8 ± 10.7 *	71.0 ± 16.8 *
	Males	48.3 ± 5.5	56.0 ± 8.3	58.9 ± 9.8	4.3 ± 2.0	3.3 ± 1.5 *	3.3 ± 0.9 *	43.8 ± 5.0	47.8 ± 5.2 *	42.8 ± 6.8 *
	Total	51.6 ± 5.8^B	55.9 ± 8.1	60.1 ± 8.7	4.5 ± 2.1	5.4 ± 3.5	4.6 ± 1.9	52.4 ± 14.7	54.8 ± 11.1	56.1 ± 19.0
Mean velocity (knots)	Females	9.3 ± 0.3 *	9.3 ± 0.2 * ^C	9.0 ± 0.4 *	14.3 ± 0.3 *, ^A	13.9 ± 0.3 *	13.7 ± 0.2 *	13.0 ± 0.3 *, ^B	12.8 ± 0.4 *	12.4 ± 0.4 *
	Males	10.6 ± 0.4 *, ^B	10.3 ± 0.3 *	10.1 ± 0.3 *	16.1 ± 0.4 *, ^A	15.6 ± 0.4 *	15.2 ± 0.5 *	15.8 ± 0.3 *, ^A	15.1 ± 0.4 *, ^C	14.7 ± 0.5 *
	Total	10.0 ± 0.7^B	9.8 ± 0.6^C	9.6 ± 0.7	15.3 ± 0.1^A	14.8 ± 0.9	14.5 ± 0.9	14.5 ± 1.5^A	14.1 ± 1.2^C	13.6 ± 1.3
VMG (knots)	Females	4.4 ± 0.1 *, ^A	4.2 ± 0.9 *, ^C	4.0 ± 1.0 *	14.1 ± 0.3 *, ^A	13.5 ± 0.2 *	13.5 ± 0.3 *	9.2 ± 0.2 *, ^B	9.1 ± 0.3 *, ^C	8.6 ± 0.4 *
	Males	5.3 ± 0.1 *, ^A	5.0 ± 0.1 *, ^C	4.7 ± 0.2 *	15.8 ± 0.4 *, ^A	14.9 ± 0.5 *	14.9 ± 0.5 *	11.1 ± 0.3 *, ^A	10.4 ± 0.2 *	10.2 ± 0.2 *
	Total	4.9 ± 0.4^A	4.6 ± 0.4^C	4.4 ± 0.4	15.0 ± 0.9^A	14.4 ± 0.9	14.2 ± 0.8	10.2 ± 1.0^A	9.8 ± 0.7^C	9.4 ± 0.9

Data presented as M ± SD. *Statistical difference between males and females of a same tertile; ^Astatistical difference between T1 versus T2 and T3; ^Bstatistical difference between T1 and T3; ^Cstatistical difference between T2 versus T3; T1: first tertile; T2: second tertile; T3: third tertile. Statistical differences were set up p < 0.05.

2.2. Regatta

The data of the present investigation belong to the data collected through the SAP-Sailing® application, which were recorded in the final of the RS:X class World Cup held in 2019 in Marseille (France) (SAP-Sailing, 2019).

The SAP-Sailing® application evaluates the following technical and tactical variables in the upwind, reaching and downwind courses: velocity, distance, manoeuvres and VMG (Figure 1). This application uses GPS trackers with mobile connection, which are placed on the board before the start of the competition; these devices were removed when the sailors finished the competition (Figure 2).

Figure 1. Velocity made good (VMG) in the courses of upwind, downwind and reaching.

Figure 2. Regatta course by sex.

This device sent the data to the SAP-Sailing® application through the mobile network. The position obtained in the regatta was considered as the representative variable of the performance of the sailor, which allowed classifying the athletes in the different performance groups. The analysed regatta had a total of 10 races. The race course consisted of six legs: two upwind, two reaching and two downwind. The race course was the same for males and females. In each race, the females began to compete 5 min after the males. The average wind speed in the regatta was 13.6 knots, with a maximum value of 21.1 knots and a minimum of 8.1 knots (Table 2). All sailors were treated independently in the data analysis, taking into account that they all competed in the total of the 10 races that compose the regatta (both male and female).

Table 2. Data of wind speed in the regatta.

		Race 1	Race 2	Race 3	Race 4	Race 5	Race 6	Race 7	Race 8	Race 9	Race 10
Wind speed (knots)	Maximum	8.4	8.9	8.9	12.8	14.1	15.7	19	18	19.3	8.1
	Minimum	10	9.8	9.7	13.1	15.7	16.6	20.2	20.7	21	10

2.3. Statistical analysis

The data are presented as means (M) ± standard deviations (SD). The normality of distribution of the variables was verified using the Kolmogorov–Smirnov test and the homogeneity of variances was determined through the Levene’s test. A two-way ANOVA was performed to identify the possible differences, based on performance level (T1, T2 and T3), sex (males and females) and the interaction between performance level and sex. A post-hoc Bonferroni test was performed when the main effect was detected. The ANOVA effect size (ES) was calculated using partial eta-squared (${\eta }_p^2$) with <0.25, 0.25–0.63 and >0.63 as small, medium and large, respectively (Ferguson, 2016). Statistical significance was set at p < 0.05. All statistical tests were performed using the Statistical Package for Social Sciences (version 20.0 for Mac, SPSS™ Inc., Chicago, IL, USA). According to the sex, a Pearson correlation coefficient and a multiple linear regression analysis were carried out with the aim of examining the association between predictive variables and performance, considering the following as predictive variables: VM upwind; distance upwind; VMG upwind; manoeuvres upwind; VM reaching; distance reaching; VMG reaching; manoeuvres reaching; VM downwind; distance downwind; VMG downwind; manoeuvres downwind; total distance and total manoeuvres. The performance as dependent variable means rank in regatta. The forward stepwise regression method was used to perform the multiple linear regression analysis.

3. Results

No differences were observed between the sexes in terms of age (25.5 ± 1.1 years in females and 25.5 ± 1.0 years in males), performance level or performance level · sex. In the total distance travelled in upwind, it was reported a lower distance covered by female sailors (p < 0.001; ${\eta }_p^2$ = 0.639) in each one of the three tertiles (p < 0.001). The performance level also showed differences (p < 0.001; ${\eta }_p^2$ = 0.397), with a shorter distance travelled upwind by the T1 sailors compared to T2 (p = 0.014) and T3 (p < 0.001) and T2 versus T3 sailors (p = 0.048) (see Table 1). Based on sex, the male T3 sailors covered a greater distance than the T2 (p = 0.016) and T1 sailors (p < 0.001), whereas the female T1 sailors covered a shorter distance than the T2 (p = 0.013) and T3 sailors (p < 0.001). In reaching distance, no effect was observed for sex (p = 0.097; ${\eta }_p^2$ = 0.058), performance level (p = 0.194; ${\eta }_p^2$ = 0.067) or performance level · sex (p = 0.403; ${\eta }_p^2$ = 0.038); however, a statistically shorter distance at downwind was obtained by females versus males (p < 0.001; ${\eta }_p^2$ = 0.903), with this difference being detected in all three tertiles (p < 0.001), without differences based on performance level (p = 0.352; ${\eta }_p^2$ = 0.043).

Regarding upwind manoeuvres, differences were detected based on performance level (p = 0.008; ${\eta }_p^2$ = 0.188) with a lower number of manoeuvres in T1 versus T3 sailors (p = 0.006). No differences were reported by sex (p = 0.137; ${\eta }_p^2$ = 0.047). The total number of manoeuvres in reaching showed no differences based on performance level (p = 0.375; ${\eta }_p^2$ = 0.041), although a higher number of manoeuvres were observed in females versus males (p < 0.001; ${\eta }_p^2$ = 0.275), reaching statistical differences in T2 (p < 0.001) and T3 sailors (p = 0.008). The number of manoeuvres in downwind showed significant difference for sex (p < 0.001; ${\eta }_p^2$ = 0.491) with statistical differences in all three tertiles (p < 0.01), but no differences based on performance level (p = 0.565; ${\eta }_p^2$ = 0.024) or performance level · sex (p = 0.178; ${\eta }_p^2$ = 0.071) were detected.

VM in upwind was greater for males than females (p < 0.001; ${\eta }_p^2$ = 0.790) in all three tertiles (p < 0.001). Moreover, differences were detected in upwind VM based on performance level (p < 0.001; ${\eta }_p^2$ = 0.282), with lower VM in T3 versus T2 (p = 0.024) and T1 (p < 0.001). In males, a higher VM in upwind was only detected in T1 versus T3 (p = 0.003), whereas the female sailors showed a significantly higher upwind VM in T2 with respect to T3 (p = 0.043) and a trend towards statistical significance in female sailors of T1 versus T3 (p = 0.053). In VM at reaching, greater velocity was obtained by male versus female sailors (p < 0.001; ${\eta }_p^2$ = 0.870), with these differences in all three tertiles (p < 0.001). Regarding the performance level, a significant effect of performance level was observed (p < 0.001; ${\eta }_p^2$ = 0.500), with greater velocity obtained by sailors of T1 compared to sailors of T2 (p < 0.001) and T3 (p < 0.001), as well as a trend towards statistical significance between sailors of T2 and T3 (p = 0.063). Regarding VM in downwind, greater velocity was obtained by male sailors (p < 0.001; ${\eta }_p^2$ = 0.920), with differences between all three tertiles (p < 0.001). Based on the performance level, a significant effect was detected (p < 0.001; ${\eta }_p^2$ = 0.511), with a lower VM in sailors of T3 versus T2 (p = 0.003) and T1 (p < 0.001); in addition, the T2 sailors were slower than the T1 sailors (p = 0.004). In the male sailors´chase, a higher VM was detected in T1 versus T2 (p = 0.001) and T3 (p < 0.001) and T2 with respect to T3 (p = 0.042), whereas, in female sailors, a higher VM was detected in T1 compared to T3 (p = 0.004), as well as a trend towards statistical significance in sailors of T2 versus T3 (p = 0.055).

VMG in upwind, a higher velocity was observed in males versus females (p < 0.001; ${\eta }_p^2$ = 0.914), with differences in all tertiles (p < 0.001). Based on the performance level, statistical differences were also detected (p < 0.001; ${\eta }_p^2$ = 0.718), with a higher VMG in T1 with respect to T2 (p < 0.001) and T3 (p < 0.001), as well as between T2 and T3 (p < 0.001). Regarding VMG in reaching, differences were detected based on sex (p < 0.001; ${\eta }_p^2$ = 0.837) and performance level (p < 0.001; ${\eta }_p^2$ = 0.469). Thus, a higher VMG was obtained by male versus female sailors in each of the three tertiles (p < 0.001). Based on the performance level, a higher VMG was detected in sailors of T1 versus sailors of T2 (p < 0.001) and T3 (p < 0.001), without differences between T2 and T3 sailors (p = 0.629). With respect to VMG in downwind, a statistical effect was detected for sex (p < 0.001; ${\eta }_p^2$ = 0.899), performance level (p < 0.001; ${\eta }_p^2$ = 0.575) and performance level · sex (p = 0.015; ${\eta }_p^2$ = 0.163). Thus, a higher VMG was obtained by males versus females (p < 0.001), with differences in all three tertiles (p < 0.001). Regarding the performance level, a higher VMG was shown by the sailors of T1 versus T2 (p = 0.002) and T3 (p < 0.001). In addition, differences were observed in favour of T2 versus T3 (p < 0.001); however, in the case of males, a higher VMG was detected in sailors of T1 versus T2 and T3 (p < 0.001), whereas no differences were observed between T2 and T3 (p = 0.0137). In females, a lower VMG was only detected in sailors of T3 compared to T1 (p < 0.001) and T2 (p = 0.001).

Table 3 shows the descriptive analysis of mean velocity, distance and VMG per leg for performance level and sex. It was observed that, in the male group, the sailors of T1 presented a better mean velocity and VMG compared to the sailors of group T3 in all legs. However, in the female group, multiple comparisons between groups T1, T2 and T3 indicate that the sailors of group T1 presented a higher VMG compared to the sailors of group T3 in upwind 1, downwind 1 and downwind 2. Regarding distance, the male sailors of group T1 travelled a shorter distance compared to the sailors of group T3 in upwind 1, although the analysis of the mean distance values did not show differences in the other legs.

Table 3. Data of mean velocity, distance and velocity made good (VMG) per leg in athletes of different performance-level groups and sex.

			T1	T2	T3
Upwind 1	Mean velocity (knots)	Females	9.3 ± 0.5	9.3 ± 0.1	8.8 ± 0.4
		Males	10.6 ± 0.3	10.4 ± 0.3	10.2 ± 0.4
	Distance (m)	Females	2315.6 ± 129.6 *, ^D	2480.5 ± 91.7	2445.5 ± 140.8
		Males	2241.1 ± 75.1**, ^B	2283.8 ± 53.5	2373.8 ± 101.4
	VMG (knots)	Females	4.4 ± 0.2 **, ^B	4.2 ± 0.1 *, ^C	3.9 ± 0.2
		Males	5.3 ± 0.1**, ^B	5.1 ± 0.1	4.8 ± 0.3
Reaching 1	Mean velocity (knots)	Females	12.9 ± 0.7	12.9 ± 0.3	12.3 ± 0.7
		Males	16.7 ± 0.4 **, ^B	15.9 ± 0.2	15.2 ± 1.2
	Distance (m)	Females	1005.6 ± 60.4	1109.1 ± 245.1	1020.1 ± 46.7
		Males	1048.7 ± 12.1	1050.1 ± 15.6	1014.9 ± 96.5
	VMG (knots)	Females	10.6 ± 0.4	10.4 ± 0.5	9.9 ± 0.8
		Males	14.5 ± 0.4 **, ^A	13.7 ± 0.2	13.2 ± 0.7
Downwind 1	Mean velocity (knots)	Females	10.7 ± 0.7	10.7 ± 0.6	10.1 ± 0.9
		Males	13.8 ± 0.3 **, ^B	13.2 ± 0.3	12.5 ± 0.8
	Distance (m)	Females	1388.6 ± 106.9	1435.1 ± 80.6	1450.3 ± 103.8
		Males	1865.1 ± 41.5	1920.1 ± 60.9	1853.1 ± 135.7
	VMG (knots)	Females	6.6 ± 0.2 **, ^B	6.4 ± 0.3 *, ^C	5.9 ± 0.5
		Males	8.2 ± 0.2 **, ^A	7.6 ± 0.1	7.3 ± 0.4
Upwind 2	Mean velocity (knots)	Females	12.4 ± 0.6	12.3 ± 0.5	12.1 ± 0.4
		Males	13.4 ± 0.3 **, ^B	13.1 ± 0.2	12.2 ± 1.1
	Distance (m)	Females	1181.4 ± 44.1	1181.1 ± 82.2	1239.7 ± 71.5
		Males	1587.9 ± 38.9	1603.1 ± 48.6	1646.5 ± 69.4
	VMG (knots)	Females	10.1 ± 0.5	10.1 ± 0.2	9.8 ± 0.2
		Males	10.2 ± 0.2 **, ^B	9.9 ± 0.2	9.1 ± 1.1
Downwind 2	Mean velocity (knots)	Females	12.9 ± 0.7	12.6 ± 0.9	12.1 ± 0.7
		Males	15,2 ± 0.3 **, ^B	14.7 ± 0.4 **, ^C	13.6 ± 1.1
	Distance (m)	Females	1307.5 ± 95.5	1311.7 ± 71.3	1321.8 ± 93.8
		Males	1402.1 ± 52.5	1440 ± 22.5 **, ^C	1355.7 ± 99.8
	VMG (knots)	Females	9.4 ± 0.3 *, ^B	9.2 ± 0.5	8.6 ± 0.6
		Males	10.8 ± 0.3 **, ^B	10.2 ± 0.3 **, ^C	9.5 ± 0.7
Reaching 2	Mean velocity (knots)	Females	13.3 ± 0.5	12.7 ± 0.3	12.6 ± 0.7
		Males	15.1 ± 0.3 **, ^B	14.5 ± 0.5 **, ^C	13.5 ± 1.1
	Distance (m)	Females	270.1 ± 15.5	271.7 ± 11.6	266.2 ± 14.7
		Males	274.5 ± 2.7	275.8 ± 2.1	263.1 ± 17.7
	VMG (knots)	Females	13.1 ± 0.5	12.5 ± 0.3	12.4 ± 0.7
		Males	14.9 ± 0.3 **, ^B	14.3 ± 0.5 **, ^C	13.3 ± 1.1

Data presented as M ± SD. *Statistical difference between different sailors of a different tertile (p < 0.05); **statistical difference between different sailors of a different tertile (p < 0.01); ^Astatistical difference between T1 versus T2 and T3; ^Bstatistical difference between T1 and T3; ^Cstatistical difference between T2 versus T3; ^Dstatistical difference between T1 and T2; T1: first tertile; T2: second tertile; T3: third tertile; VMG, velocity made good.

The results of the correlation coefficients are shown in Table 4. For ranking, the correlation was significant with upwind mean velocity (r = −0.66; p < 0.01), upwind distance (r = −0.64; p < 0.01), upwind VMG (r = −0.91; p < 0.01), reaching mean velocity (r = −0.78; p < 0.01), reaching VMG (r = −0.74; p < 0.01), downwind mean velocity (r = −0.86; p < 0.01), downwind VMG (r = −0.92; p < 0.01), total distance (r = −0.65; p < 0.01), manoeuvres upwind (r = −0.66; p < 0.01), manoeuvres reaching (r = −034; p < 0.05) and total manoeuvres (r = 0.31; p < 0.05) in male windsurfers. The results in the group of female windsurfers showed significant correlation with upwind mean velocity (r = −0.53; p < 0.01), upwind distance (r = −0.62; p < 0.01), upwind VMG (r = −0.93; p < 0.01), reaching VMG (r = −0.57; p < 0.01), downwind mean velocity (r = −0.68; p < 0.01), downwind VMG (r = −0.71; p < 0.01), total distance (r = 0.48; p < 0.01), manoeuvres upwind (r = 0.30; p < 0.05), manoeuvres downwind (r = 0.27; p < 0.05) and total manoeuvres (r = 0.35; p < 0.05).

Table 4. Coefficients of correlation (r) among ranking and predictive variables in male and female windsurfers.

Male
Variable	UMV	UD	UVMG	RMV	RD	RVMG	DMV	DD	DVMG	TD	MU	MR	MD	TM
Ranking	−0.66**	0.64**	−0.91**	−0.78**	−0.16	−0.74**	−0.86**	0.12	−0.92**	0.65**	0.48**	−0.34*	−0.09	0.31*
Upwind mean velocity (knots)		−0.05**	0.75**	0.64**	0.11	0.66**	0.64**	−0.08	0.65**	−0.07	−0.49**	0.38*	0.10	−0.31
Upwind distance (m)			−0.68**	−0.49**	−0.24	−0.38*	−0.56**	−0.08	−0.54**	0.9**	0.27*	−0.07	−0.02	0.21
Upwind velocity made good (knots)				0.77**	0.25	0.71**	0.83**	−0.02	0.84**	−0.63**	−0.57**	0.34*	0.08	−0.39*
Reaching mean velocity (knots)					0.14	0.97**	0.81**	0.03	0.78**	−0.45**	−0.39*	0.13	−0.01	−0.33*
Reaching distance (m)						0.04	0.11	0.34*	−0.01	−0.02	−0.24	0.18	0.16	−0.09
Reaching velocity made good (knots)							0.78**	−0.03	0.76**	−0.37*	−0.42*	0.17	−0.05	−0.37*
Downwind mean velocity (knots)								0.14	0.91**	−0.46**	−0.33*	0.34*	−0.16	−0.33*
Downwind distance (m)									−0.25*	0.35*	0.13	−0.07	−0.04	0.07
Downwind velocity made good (knots)										−0.62**	−0.40*	0.35*	−0.01	−0.31
Total distance (m)											0.29	−0.09	−0.03	0.22
Manoeuvres upwind (number)												−0.31*	−0.03	0.80**
Manoeuvres reaching (numbers)													0.07	−0.08
Manoeuvres downwind (numbers)														0.55**
Female
Variable	UMV	UD	UVMG	RMV	RD	RVMG	DMV	DD	DVMG	TD	MU	MR	MD	TM
Ranking	−0.53**	0.62**	−0.93**	−0.64**	0.01	−0.57**	−0.68**	0.32	−0.71**	0.48**	0.30*	0.11	0.27*	0.35*
Upwind mean velocity (knots)		0.26	0.46**	0.39*	0.12	0.3	0.71**	0.15	0.36*	0.27	−0.48**	0.06	−0.21	−0.34*
Upwind distance (m)			−0.68**	−0.27	0.31	−0.28	−0.06	0.36*	−0.28	0.87**	−0.08	0.19	0.01	0.00
Upwind velocity made good (knots)				0.55**	−0.13	0.5**	0.61**	−0.16	0.54**	−0.53**	−0.13	−0.08	−0.22	−0.24
Reaching mean velocity (knots)					0.29	0.88**	0.58**	−0.05	0.47**	−0.03	−0.23	0.01	0.02	−0.06
Reaching distance (m)						−0.07	0.11	−0.12	0.25	0.64**	−0.15	0.67**	0.28	0.27
Reaching velocity made good (knots)							0.47**	−0.01	0.33*	−0.21	−0.16	−0.32	−0.08	−0.18
Downwind mean velocity (knots)								0.13	0.66**	0.06	−0.32	0.07	−0.30	−0.35*
Downwind distance (m)									−0.62**	−0.48**	0.12	−0.12	−0.07	−0.03
Downwind velocity made good (knots)										−0.25	−0.27	0.19	−0.14	−0.18
Total distance (m)											−0.08	0.41*	0.12	0.12
Manoeuvres upwind (number)												0.06	0.15	0.51**
Manoeuvres reaching (numbers)													0.33*	0.44*
Manoeuvres downwind (numbers)														0.91**

Note: UMV, upwind mean velocity (knots); UD, upwind distance (m); UVMG, upwind velocity made good (knots); RMV, reaching mean velocity (knots); RD, reaching distance (m); RVMG, reaching velocity made good (knots); DMV, downwind mean velocity (knots); DD, downwind distance (m); DVMG, downwind velocity made good (knots); TD, total distance (m); MU, manoeuvres upwind (number); MR, manoeuvres reaching (numbers); MD, manoeuvres downwind (numbers); TM, total manoeuvres (number); *indicates a significant correlation (p < 0.05); **indicates a significant correlation (p < 0.01).

Table 5 shows the results of the multiple linear regression analysis using the forward stepwise method for male and female windsurfers. The model obtained showed a linear relationship of 91% and goodness-of-fit or R² = 0.91 in male windsurfers. This model includes the constant, upwind VMG and downwind VMG, excluding the rest of the variables. According to female windsurfers, the model obtained showed a linear relationship of 97% and a goodness-of-fit of R² = 0.96. The variables of this model were the constant, upwind VMG, downwind VMG, manoeuvres upwind, manoeuvres reaching and reaching distance.

Table 5. Association of the rank in regatta with upwind, reaching and downwind velocity made good, mean velocity, manoeuvres and distance and total distance and total manoevures for male windsurfers (n = 28) and female windsurfers (n = 25).

Male
Model	Variables	B	95% CI	β	p Value
1	(Constant)	187.49	165.36–209.63		0.0^c
	Upwind velocity made good (knots)	−9.39	−13.84 to −6.03	−0.56	0.0^c
	Downwind velocity made good (knots)	−13.61	−20.51 to −6.72	−0.43	0.0^c
Female
Model	Variables	B	95% CI	β	p value
1	(Constant)	180.26	162.2–198.32		0.0^c
	Upwind velocity made good (knots)	−30	−33.86 to −26.13	−0.77	0.0^c
	Downwind velocity made good (knots)	−4.68	−6.49 to −2.86	−0.27	0.0^c
	Manoeuvres upwind (numbers)	0.09	0.06–0.18	0.09	0.03^c
	Manoeuvres reaching (numbers)	0.52	0.22–0.82	0.19	0.0^c
	Reaching distance (m)	−0.001	−0.002–0.0	−0.13	0.02^c

Note: Method used: forward stepwise; ^cstatistically significant association; B = linear regression coefficient; βß = standardised partial regression coefficient; CI = confidence interval for B.

4. Discussion

The aim of this study was to explore the potential differences among the RS:X class Olympic sailors as a function of performance level and sex in velocity, distance travelled and manoeuvres performed. The results of our study show that the higher-level sailors (T1) obtained a higher mean velocity and VMG on the upwind, reaching and downwind courses, while performing less upwind manoeuvres and covering a shorter distance in this leg. Regarding sex, statistical differences were reported in mean velocity, VMG, distance and manoeuvres.

Among the three courses analysed in the entire sample, the VMG was higher in reaching (14.5 ± 0.9 knots) compared to downwind (9.7 ± 0.9 knots) and upwind (4.6 ± 0.4 knots). In addition, it was observed that the highest velocity in all the participants was in the downwind courses (reaching and downwind). This may be due to the fact that the board exceeds the wind velocity in such a way that it creates its own apparent wind in these courses, resulting in a vector that combines the real wind velocity and the velocity of the hull. The boards of the RS:X class have a design that allows them to reach velocities much higher than the actual wind velocity (Castagna & Brisswalter, 2007). In addition to this, the planing condition, i.e., a situation in which the board reduces the surface that is in contact with the water and therefore also reduces the hydrodynamic resistance, allows the sailor to reach a higher velocity (Gourlay & Martellotta, 2011). Therefore, and based on our results, we can assume that sailors with a higher level (T1) manage to make better use of the apparent wind and increase the gliding condition of the board, thus obtaining higher values in mean velocity and VMG compared to the rest of their competitors. In addition, our results have shown a small difference in the mean velocity of each performance level in upwind. Therefore, with similar board speeds in upwind, higher-level sailors (T1) have a better sailing angle in windward compared to medium-level sailors (T2). Similarly, we observed that males achieved a higher mean velocity and VMG compared to females in these three directions. Both results coincide with those obtained in previous studies of RS:X class sailors (Anastasiou et al., 2019; Hagiwara & Ishii, 2016). This higher velocity in males was expected, since their sail surface is greater than that of females (9.8 m² vs. 8.5 m²), thus they have a greater potential to develop greater velocity.

According to the distance covered, our results show that upwind is the only course where significant difference can be seen between the sailors according to their level of performance. The higher-level sailors (T1) travel a shorter distance in the upwind section of the course´; therefore, we can highlight its importance in relation to the performance of the sailor, as it is in this course where it is possible to establish a greater difference in distance with the rest of the sailors (Castagna et al., 2008). Furthermore, this distance seems to be stable, since no significant differences are observed in the reaching and downwind courses. Other studies have also found this difference in the distance travelled in the upwind course between higher- and lower-level sailors (Hagiwara et al., 2017). When comparing male and female windsurfers, it was observed that females travelled a shorter distance to carry out the leg in upwind and downwind; however, no differences were found in reaching. It can be considered that this greater distance travelled by males is compensated by their greater velocity, which allows them to reach the destination in a shorter time (7500 s vs. 8268 s). During upwind and downwind, the windsurfer can modify the orientation of the board with respect to the wind to facilitate the planing of the board (planing condition) and achieve a higher velocity, although this increased angle could also increase the distance travelled. In the RS:X class, the planing condition is different for males and females. The planing condition is conditioned by wind speed. Females have a lower wind speed at which they can plane compared to males. This disadvantage could be compensated by males performing sail pumping, which provides the board with additional propulsion. Sail pumping is considered particularly effective at wind velocities up to 15 knots (Vogiatzis et al., 2002). In the regatta that has been analysed in our study, the average wind speed was 13.6 knots and the wind speed was less than 15 knots in 5 of the 10 races analysed. Therefore, it could be considered that the conditions of the regatta were more favourable for the males, who could perform the sail pumping manoeuvre and thus obtain a higher speed of the board and travel a shorter distance compared to the females. In addition, we must bear in mind that the higher the velocity of the board, the greater the physiological demand; thus, this difference between males and females could be even greater (Vogiatzis & De Vito, 2015). In order to reduce the distance travelled by the board in upwind, the angle between the wind and the bow must be approximately 40º (Castagna et al., 2008)

In the manoeuvres performed, significant differences were only detected in upwind for the entire sample, with the higher-level sailors (T1) being those who performed the smallest number of manoeuvres. Reducing the number of manoeuvres is very important for the performance of the sailor, since these actions reduce the velocity of the board, thus the smaller the number of manoeuvres, the lower the loss of velocity during the upwind route (Bojsen-Møller et al., 2015). Another aspect to take into account is that, after each manoeuvre, the sailor has to perform the pumping action of the sail (sail pumping) in order to regain the maximum velocity of the board in the shortest possible time (Anastasiou et al., 2019). This action consists of pushing and pulling the sail in a rhythmic way to increase the pressure of the wind on the sail, which can demand up to 80% −90% of the VO_2max from the sailor and requires great muscle activity in the upper and lower limbs (Castagna & Brisswalter, 2007). Therefore, performing more manoeuvres not only leads to a loss of velocity but can also lead to greater fatigue in the sailor, thus we can infer that the higher-level sailors are more efficient in the upwind course. The comparison between males and females indicates that there were no differences in upwind. However, the males in groups T2 and T3 performed fewer manoeuvres in reaching and downwind. These differences in the number of manoeuvres could also explain the lower average velocity and VMG of females compared to males on these three courses.

In our study, we can consider some limitations. Firstly, regarding the number of participants analysed, the study sample consisted of only 53 elite sailors. However, this represents 96% of the total number of sailors who have participated in the regatta. Secondly, the variables have not been analysed as a function of wind speeds and racing. The ranges in wind speed in each of the races would have an effect on performance. Moreover, future studies could be focused on the analysis of the technical and tactical variables in different wind conditions and specifically analyse each of the races that make up the regatta.

The Neilpryde RS:X© class will be replaced by the IQFoil class as the official Olympic equipment for male and female windsurfers for Paris 2024. This new class has similarities to the RS:X. Therefore, the current findings of our study can be of great interest to coaches and windsurfers, since they can help them to better target the objectives of their training. Our results suggest that the training carried out in the water should be mainly focused on the upwind and with the following objectives: to increase the average velocity and VMG, reduce the distance travelled and reduce the number of manoeuvres performed (turns). Similarly, in the reaching and downwind, the mean velocity and the VMG must also be improved.

5. Conclusions

This is the first study to analyse VMG, distance travelled and manoeuvres performed using a GPS device in RS:X class sailors and in a qualifying regatta for the Olympic Games. In the RS:X class, sailor performance is determined by upwind and downwind VMG.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon request.

Additional information

Funding

The author(s) reported that there is no funding associated with the work featured in this article.