Lack of Neuromuscular Fatigue Due to Recreational Doubles Pickleball

Abstract

Background: The lack of knowledge about physical responses to pickleball creates a clear gap about performance in this sport. The purpose of this study was to investigate neuromuscular fatigue caused by playing doubles pickleball. Methods: Recreational pickleball players (n = 32, mean age = 60.0 years) were recruited to perform sets of four countermovement jumps (CMJs) on a force plate before and after doubles pickleball matches. Results: For players who had not played a match prior to testing, there was a significant learning effect across trials within the baseline set of jumps for five outcomes from the CMJ test, including propulsive peak force (p = 0.005); however, there was no significant learning effect for jump height. There were significant improvements in the large effect size for all except one dependent variable (propulsive phase time) between the first and second set of jumps (i.e., after one match). Neither further increases nor decreases were seen after the second set of jumps. Conclusions: Participants saw significant increases in CMJ performance across trials after one pickleball match, indicating learning and potentiation effects. After three matches of doubles pickleball, no fatigue effect was detected.

1. Introduction

Pickleball is a sport with steadily increasing numbers of people participating in the last few years; USA Pickleball membership has increased up to almost 79,000 people, an increase of 15% from 2023 [1]. According to the Sports & Fitness Industry Association (SFIA), in 2024, pickleball was the fastest growing sport for the third year in a row [2,3]. The SFIA stated that in the last 3 years, the number of people playing pickleball casually (which they defined as one to seven times per year) increased by 208%; the number of core players (defined as people playing eight times or more a year, but who on average, play 40 times per year) increased by 255% [2,3]. Pickleball has become a professionalized sport, with professionals reporting an average payout of USD 96,000 in 2022 [4]. For recreational players, amateur and open tournaments are increasing in popularity [1]. One interesting aspect of pickleball players is how many are older—the 65+ age category is tied as the second largest, only 10% lower than the 25–34-year-old category, making the average age of those playing eight times or more a year about 42 years old. Among both casual and core players, the sex breakdown seems to be about 60% males and 40% females [3]. The overall increase in pickleball play has contributed to an increase in the absolute number of injuries resulting from pickleball play [5,6]. The key to both injury prevention and sport performance is physical training [7,8]. In order to know what and how to train for a sport, one must first discern what measurable qualities are important for success in that sport. Despite the rising popularity of pickleball, there is almost no research; as of 3 April 2024, only 23 studies were listed in PubMed when we searched for “pickleball”, and almost all of them either collect injury statistics and reports or examine the psychosocial benefits of the sport, with very few examining any physical qualities. Those that did focus on physical qualities primarily looked at acute cardiometabolic response to pickleball with the purpose of trying to determine if pickleball counts as aerobic exercise [9,10,11]. There is a significant research opportunity to increase our understanding of the physical qualities related to success in pickleball.

In many sports, lower body force production is a key physiological characteristic related to sport performance and other fitness characteristics like sprint speed [12,13,14,15,16,17,18,19]. A common test of lower body force production is the counter movement jump (CMJ) test [20]. It is safe to perform, even in elderly individuals with sarcopenia or potentially at risk for falls [21,22,23]. Claudino et al. [20] performed a meta-analysis of 151 research articles to determine the sensitivity of the CMJ test to monitor neuromuscular status. Their review found that jump height was sensitive enough to detect fatigue of a moderate effect size when the average of trials performed was used, but not the single highest trial. They also determined that several variables, including jump height, peak power, and peak force, were sensitive enough to detect supercompensation after training, with large effect sizes seen when the average across trials was used. The primary conclusions and recommendations of the meta-analysis were that CMJ testing is an appropriate way to monitor neuromuscular status, though they cautioned that more work is needed on detecting fatigue, and that researchers should perform multiple trials and use the average of all trials to represent performance. Recently, Watkins and colleagues [24,25] have investigated fatigue as measured by CMJ performance and determined that a strength-phase resistance training session performed for the purposes of inducing fatigue led to significant decreases of about 12% in vertical jump height and 5% in peak power; these numbers give researchers and coaches easily determinable comparative criteria to determine if the activity in question causes fatigue (or in the words of Watkins et al. [24], if the activity is “extremely fatiguing” (p. 197)), though these should not be interpreted as cut-offs to first detect fatigue, as the onset of fatigue is likely much lower.

There are, however, problems in generalizing the conclusions of the meta-analysis [20] or the recent work by Watkins et al. [24,25]. The first problem is that the protocol for conducting the CMJ test, especially in terms of warm up, familiarization, and measured trials completed, has not been standardized. For example, as the meta-analysis points out, most researchers and coaches follow the rule of thumb in testing: giving the athlete three trials and taking the best one. Even following the recommendation of Claudino et al.’s [20] meta-analysis of using the average of the three trials, it is possible that the participant has not achieved their peak performance. Few studies have devoted time to the details of their research, like Gathercole and colleagues [26], to provide and report a robust familiarization protocol and test for reliability. Gathercole et al. [26] reported having participants complete a familiarization session at least 7 days prior to the baseline data collection session, which consisted of completing 10 ± 4 CMJ trials until the difference between four consecutive jumps was within 10% for peak and minimum displacement and velocity and peak power. Next, they performed three testing sessions, each separated by one day, to assess the reliability of the CMJ; they found inter-day differences to be trivial and concluded that the CMJ test (at least when a whole session for familiarizing is done) is near perfectly reliable.

However, the subjects that Gathercole et al. [26] tested were male college-level team-sport athletes, which brings up the second limitation highlighted in the meta-analysis by Claudino et al. [20]: of the pooled 4834 participants captured in the meta-analysis, the mean age was 23 ± 12 years, and 80% of subjects were male. Furthermore, 60% of participants were athletes, and of those the authors considered non-athletes, 69% were labeled as physically active individuals or physical education/sports science students. Only 9% were elderly, and 12% were sedentary. Therefore, the vast majority of the literature on CMJ testing represents young male athletes and may not be generalizable to older adults, especially older females. A large demographic segment of pickleball players are over 50 years old, and pickleball is played recreationally and competitively in mixed-sex pairs.

Vertical jump testing has been conducted in older populations of both males and females [21,22,23,27,28,29]. Even in sarcopenic individuals, vertical jump testing has been found to be safe [21]. Protocols vary widely between studies in both athletic and older adult populations, but despite the variability in methods between studies, CMJ testing is considered to be reliable in both populations [22,23,29,30,31]. However, basic inter-trial reliability does not guarantee that CMJ testing will be sensitive to neuromuscular fatigue; tests can have acceptable inter-trial reliability, especially as indicated by the intraclass correlation coefficient (ICC), while still having a large enough coefficient of variation, especially in untrained populations, to make detecting meaningful acute changes difficult [20,25,26,32]. In the context of this particular study, it is yet untested if CMJ testing will be sensitive to changes due to playing pickleball.

Only one study was found that tested vertical jump in relation to pickleball [10]; in this study, inactive adults 50 years and older were recruited to an intervention study, where pickleball was the intervention. The study tested the impact of 6 weeks of pickleball play (3 days per week, 1 h per session) on outcomes, including CMJ performance, and found that vertical jump height (as calculated by flight time measured on a jump mat) increased significantly (p < 0.001) by 11%. In testing CMJ performance, subjects were given three trials; no warmup or familiarization was included, and the authors did not describe if they used the best or average of the three trials in their statistical analysis. Aside from the feasibility of using CMJ testing in older, previously sedentary, pickleball players, the fact that there was in improvement in CMJ performance indicates that pickleball stimulates lower body power and therefore provides evidence that lower body power could be important in this sport, as with so many other sports. If pickleball can cause enough acute physiological stress to stimulate chronic increases in lower body power, it stands to reason that it may stimulate acute decreases in power, i.e., fatigue. Since there is no prior research on pickleball, we shall use tennis, table tennis, and padel as the closest surrogates. As expected, acute fatigue (both physical and mental) has been shown to reduce physical and technical performance in tennis [33,34,35,36,37], table tennis [38,39], and padel [40]. Specifically, tennis play can induce neuromuscular fatigue, as assessed by jumping power, after the completion of a match or even mid-match [36,37]. Padel may represent the total movement patterns of pickleball more closely than tennis due to more similar court dimensions. While no research has examined neuromuscular fatigue due to playing a true padel match yet, one study did use a simulated padel competition to examine fatigue response [41]; this study did not find neuromuscular fatigue, as assessed by a variety of jump tests, due to their simulated padel competition. What is unknown is if pickleball, which has a similar intensity to table tennis, doubles tennis, and padel (measured in METs) [9,11,42], will cause acute neuromuscular fatigue in lower body power. Determining if, and how much, fatigue would be caused during doubles pickleball play is an important first step in determining playing and training strategies.

Initially, our research question was the following: How much fatigue, if any, is caused by playing a match of doubles pickleball? We hypothesized that there would be some detectable decrease in CMJ performance after doubles pickleball. As is described in the methods and results, after preliminary analysis, we expanded our research question to the following: How many matches of doubles pickleball are needed to see a fatigue effect? We also had to add several key secondary questions to our research. In the end, we addressed the following objectives:

• Determine whether or not pickleball players exhibit a learning effect on the CMJ test.
• Determine whether or not playing doubles pickleball causes neuromuscular fatigue.
• Determine the reliability of testing CMJ performance across multiple doubles pickleball matches.
• Investigate factors that may influence CMJ performance in pickleball players.

2. Materials and Methods

2.1. Approach to the Problem

In our study, we tested CMJ performance on a force plate to quantify change in performance in response to an acute bout of exercise (a pickleball match). Initially, participants performed CMJs before and after a single pickleball match. Based on initial results, we realized we needed to expand beyond a single match to potentially observe a fatigue effect. Therefore, we expanded the protocol to capture CMJ performance before and after multiple pickleball matches.

2.2. Participants

Recruitment criteria and procedures were identical in the original and expanded methods of our research. We recruited adults who were actively playing pickleball at public courts. Potential participants were excluded if they had any lower-body injury in the past 6 weeks, a medical condition that would jeopardize their safety, or if they were under 18 years old. The University IRB committee approved the procedures (IRB study #270) in accordance with the latest version of the Declaration of Helsinki. All participants verbalized consent to participate as we did not collect any personally identifiable information in the study. Participants were made aware that they could withdraw at any time.

2.3. Protocol

The research team went to public pickleball courts in the area; we arrived and set up testing at or by the time the courts opened for play. Thus, we started recruiting participants each day at or near the start of their play. Some players arrived at the facility and played a game or two before they could be recruited to the study; to account for this, as described below, the number of matches played prior to enrollment in the research was recorded. Data collection lasted for about 3 h each session, capturing the whole time that the facilities visited were open for play.

Immediately after consent, participants filled out a short demographic questionnaire, additionally indicating if they had already played any matches of pickleball prior to participating in the research. Participants were instructed to complete eight CMJs with their hands on their hips to eliminate impact on force, take-off velocity, or power due to arm swing [43]. The first four jumps were performed as warm up and familiarization: jumps 1–2 were at 50% of their maximal effort, and jumps 3–4 at 75% of their maximal effort. Then, jumps 5–8 were performed at full effort and recorded as the performance data. All eight jumps were separated by 30–40 s of rest. Jumps 5–8 were completed on a Hawkins Dynamics Force Plate (Hawkins Dynamics, Westbrook, ME, USA) with each foot designated to one plate. After baseline testing, participants completed any self-chosen warm up among the people they played with (while not measured in the research, most people choose to hit the ball back and forth in a mostly static fashion for 2–5 min prior to starting a game), then played a standard match to 11 points (win by two points, per standard rules). Immediately post-match, participants completed four jumps at maximal effort, following prior protocol and maintaining an inter-trial rest period of 30–40 s.

Initially, participants were only measured before and after one game. After initial results were analyzed and presented at a conference [44], we expanded the protocol to ask participants to perform the CMJ test at baseline and after every match they played during their session. To accommodate the venue and players’ play schedule, no limitations were imposed on the number of games played and the amount of rest in between each game—we measured participants after as many games as they were able to play and willing to continue in the research.

To quantify different aspects of neuromuscular fatigue, a variety of measures were obtained from the CMJ either directly from the Hawkin Dynamics software (version 1.10) or with minor transformation via normalization to bodyweight (as captured by the force plate during the steady weighing period). Previous studies have pointed out the necessity to consider how fatigue may affect primary markers of CMJ performance (e.g., jump height, peak power), as well as phase (e.g., braking, propulsion) and time characteristics [20,25,26,32]. To quantify aspects of the entire CMJ, the following nine dependent variables were obtained:

• Jump height (calculated from takeoff velocity).
• Net impulse normalized to body mass (hereafter, net impulse).
• Time to takeoff.
• Reactive strength index modified (RSImod; jump height divided by time to takeoff).

Specific CMJ phase variables included:

• Braking phase mean power normalized to body mass (hereafter, braking mean power).
• Propulsive phase peak power normalized to body mass (hereafter, propulsive peak power).
• Propulsive phase peak force normalized to body mass (hereafter, propulsive peak force).
• Propulsive phase mean power normalized to body mass (hereafter, propulsive mean power).
• Propulsive phase time.

Additionally, system weight, as captured by the force plate, was used to quantify participant body mass.

2.4. Data Analysis Methods

The nine dependent variables were analyzed according to the following objectives. All statistical analyses were performed in R Version 4.3.0 [45]. The statistical methods to address each objective are as follows.

• Objective 1: Determine whether or not pickleball players exhibit a learning effect on the CMJ test.

To test whether or not a learning effect occurred, we tested each dependent variable under the mixed-effects model using the lme4 package in R [46]. The mixed-effects model estimated the expected difference of each dependent variable between the first and second jump set and between the second and third set (fixed-effect), and it accounted for repeated measures of each subject (random-effect). There were 26 subjects available for comparing the first and second set and 12 subjects for comparing the second and third set. We calculated Cohen’s d and η² to quantify the effect sizes using the effectsize package in R [47]. Per standard interpretation, Cohen’s d values of <0.2 are considered trivial, >0.2 but <0.5 is small, >0.5 but <0.8 is medium, and >0.8 is large; η² values were interpreted as <0.01 is trivial, >0.01 but <0.06 is small, >0.06 but <0.14 is medium, and >0.14 is large [48,49]. We also performed the Bland–Altman analysis with 95% limits of agreement (LOA) between the first and second set and between the second and third set [50]. For this analysis, four repeated measures per set were averaged. Because some participants (16 subjects) had played a pickleball match prior to enrolling in the research, we performed additional data analysis to test whether or not the learning effect of each dependent variable depended on the number of prior pickleball matches.

In addition, among those who had not played a pickleball match prior to the first jump set, we also tested for whether or not the learning effect of each dependent variable occurred within the first set (i.e., the expected difference of each dependent variable per repetition within the first set).

• Objective 2: Determine whether or not playing doubles pickleball causes neuromuscular fatigue.

We detected a learning effect between the first and second jump set, with no further increases thereafter. The inclusion of these initial “pre-familiarized” trials would systematically reduce baseline scores meant to represent an unfatigued, maximally performing state. Therefore, 12 subjects who performed at least the second set and the third set (i.e., played at least two games during the research) were included in the analysis. The linear mixed-effects model was used to test for the slope of each dependent variable with respect to the set.

• Objective 3: Determine the reliability of testing CMJ performance across multiple doubles pickleball matches.

The third objective was to quantify the reliability of each dependent variable between sets. The ICC was calculated under the mixed-effects model for each dependent variable after familiarization (removing the first set), and the standard error of measurement (SEM) and coefficient of variation percentage (CV%) were calculated as well. We also calculated these statistics for the first, second, third, and fourth jump set separately.

• Objective 4. Investigate factors related to CMJ performance in pickleball players.

The fourth objective was to investigate factors related to each dependent variable. The mixed-effects model accounted for the random effects due to the repeated measures of each subject, and sex (female or male), age (in years), aerobic exercise (no or yes), resistance training (no or yes), and the frequency of playing pickleball were used as fixed-effects. For this analysis, we included all data points including the first set, and we accounted for the learning effect and the number of matches played prior to the jump set in the model.

3. Results

A total of 32 subjects enrolled in the study and completed at least one set of CMJs, with ages ranging from 20 to 80 years old. Demographic and activity participation characteristics can be seen in

Table 1. Age, body mass, and activity participation.

	Total Sample (n = 32)	Males (n = 14)	Females (n = 18)	50+ Years Old (n = 26)	under 50 Years Old (n = 6)
	Mean (SD)
Age (years)	60.0 (15.9)	63.7 (13.6)	57.1 (17.3)	66.7 (7.3)	31.0 (7.7)
Body mass (kg)	77.2 (18.8)	90.5 (14.1)	66.8 (15.1)	76.5 (19.4)	79.8 (17.2)
Pickleball experience (months)	14.1 (15.5)	23.8 (18.7)	6.6 (5.7)	16.6 (16.2)	3.2 (2.4)
Weekly frequency of pickleball in last month (days/week)	2.4 (1.2)	2.9 (1.5)	2.0 (0.8)	2.6 (1.2)	1.5 (0.4)
	Number of participants (note: one participant did not answer these questions, hence the total of 31 instead of 32)
In the last 30 days, did you participate in 12+ hours of aerobic physical activity (including pickleball) most weeks?	Yes = 24 No = 7	Yes = 14 No = 0	Yes = 10 No = 7	Yes = 20 No = 5	Yes = 4 No = 2
In the last 30 days, did you usually perform at least 2 days of resistance training most weeks?	Yes = 16 No = 15	Yes = 7 No = 7	Yes = 9 No = 8	Yes = 11 No = 14	Yes = 5 No = 1

. About half the sample (n = 16) had played pickleball for at least a year, with a maximum experience of 5 years, and a minimum experience of having just started playing that month. Almost everyone (n = 29) played at least twice a week, with 12 people playing between 3 and 5 days per week. Most participants engaged in regular physical activity, including pickleball, and half engaged in regular resistance training.

As will be presented shortly, there were significant differences in many of the measured outcomes between pre and post first pickleball match. Therefore, for descriptive statistics, we present the average of all four trials for the pre and post first match sets of CMJ in

Table 2. Average CMJ performance before and after one match.

		Total Sample (n = 32)	Males (n = 14)	Females (n = 18)	50+ Years Old (n = 26)	under 50 Years Old (n = 6)
		Mean (SD)
No prior matches	Jump height (m)	0.133 (0.0465)	0.105 (0.0287)	0.159 (0.0445)	0.116 (0.0404)	0.176 (0.0308)
	Net impulse (N.s/kg)	2.29 (0.391)	2.07 (0.311)	2.50 (0.345)	2.15 (0.343)	2.66 (0.225)
	Time to takeoff (s)	0.815 (0.1655)	0.881 (0.0977)	0.754 (0.1919)	0.777 (0.161)	0.909 (0.1420)
	RSImod (no units)	0.1743 (0.08326)	0.1215 (0.03878)	0.2224 (0.08422)	0.1654 (0.09528)	0.1965 (0.03393)
	Braking mean power (W/kg)	−6.18 (1.875)	−5.32 (1.688)	−6.97 (1.713)	−5.82 (1.927)	−7.10 (1.429)
	Propulsive peak power (W/kg)	28.80 (5.957)	25.14 (3.296)	32.12 (5.927)	27.49 (6.34)	32.07 (3.153)
	Propulsive peak force (N/kg)	21.13 (2.95)	20.16 (1.878)	22.02 (3.468)	21.96 (3.040)	19.07 (1.213)
	Propulsive mean power (W/kg)	16.32 (3.748)	13.96 (2.033)	18.47 (3.680)	15.87 (4.216)	17.45 (1.885)
	Propulsive phase time (s)	0.243 (0.0463)	0.252 (0.0348)	0.235 (0.0542)	0.222 (0.0319)	0.297 (0.0308)
One prior match	Jump height (m)	0.152 (0.0725)	0.168 (0.0809)	0.137 (0.0603)	0.134 (0.0575)	0.230 (0.0816)
	Net impulse (N.s/kg)	2.44 (0.560)	2.55 (0.602)	2.33 (0.499)	2.30 (0.467)	3.04 (0.535)
	Time to takeoff (s)	0.793 (0.1824)	0.817 (0.1996)	0.769 (0.1626)	0.788 (0.1903)	0.815 (0.1441)
	RSImod (no units)	0.1977 (0.09886)	0.2096 (0.09909)	0.1862 (0.09816)	0.1772 (0.08472)	0.2877 (0.10816)
	Braking mean power (W/kg)	−6.78 (2.182)	−6.89 (2.44)	−6.68 (1.920)	−6.30 (1.841)	−8.92 (2.321)
	Propulsive peak power (W/kg)	30.55 (7.816)	32.14 (7.379)	29.02 (7.982)	28.75 (6.761)	38.46 (7.34)
	Propulsive peak force (N/kg)	21.52 (2.776)	21.59 (2.508)	21.46 (3.034)	21.50 (2.846)	21.62 (2.514)
	Propulsive mean power (W/kg)	17.36 (4.524)	18.18 (4.392)	29.02 (7.982)	16.47 (3.981)	21.26 (4.789)
	Propulsive phase time (s)	0.244 (0.0524)	0.251 (0.0464)	0.236 (0.0572)	0.237 (0.0513)	0.272 (0.0492)

.

Objective 1

There were significant differences between the first and second set for all dependent variables except for propulsive phase time, but there were not significant differences between the second and third set for any of the dependent variables (

Table 3. Estimated mean change (95% CI), p-value, effect sizes (Cohen’s d and η²), and 95% LOA for comparing each dependent variable between the first and second set and between the second and third set.

The First and Second Set	Estimate (95% CI)	p-Value	d	η2	95% LOA
Jump height (m)	0.014 (0.011, 0.017)	<0.001	1.129 ***	0.28 ***	(−0.011, 0.038)
Net impulse (N.s/kg)	0.124 (0.087, 0.161)	<0.001	0.877 ***	0.199 ***	(−0.163, 0.404)
Time to takeoff (s)	−0.042 (−0.072, −0.012)	0.007	−0.436 *	0.042 *	(−0.285, 0.185)
RSImod (no units)	0.030 (0.022, 0.038)	<0.001	1.013 ***	0.233 ***	(−0.031, 0.091)
Braking mean power (W/kg)	−0.621 (−0.923, −0.318)	<0.001	−0.628 **	0.086 **	(−2.719, 1.448)
Propulsive peak power (W/kg)	2.005 (1.671, 2.340)	<0.001	1.743 ***	0.446 ***	(−0.363, 4.362)
Propulsive peak force (N/kg)	0.663 (0.317, 1.009)	<0.001	0.481 *	0.076 **	(−2.298, 3.700)
Propulsive mean power (W/kg)	1.191 (0.888, 1.493)	<0.001	1.131 ***	0.258 ***	(−1.028, 3.534)
Propulsive phase time (s)	−0.006 (−0.013, 0.002)	0.156	−0.25 *	0.012 *	(−0.083, 0.065)
The Second and Third Set	Estimate (95% CI)	p-Value	d	η2	95% LOA
Jump height (m)	−0.004 (−0.007, 0.000)	0.058	−0.376	0.044	(−0.028, 0.020)
Net impulse (N.s/kg)	−0.020 (−0.060, 0.021)	0.341	−0.228	0.011	(−0.263, 0.214)
Time to takeoff (s)	−0.017 (−0.049, 0.014)	0.277	−0.244	0.015	(−0.163, 0.131)
RSImod (no units)	−0.001 (−0.010, 0.007)	0.785	−0.09	0.001	(−0.064, 0.059)
Braking mean power (W/kg)	−0.041 (−0.337, 0.256)	0.787	−0.016	0.001	(−1.682, 1.658)
Propulsive peak power (W/kg)	−0.194 (−0.605, 0.216)	0.354	−0.225	0.011	(−2.725, 2.220)
Propulsive peak force (N/kg)	−0.041 (−0.414, 0.331)	0.828	−0.073	0.001	(−1.828, 1.710)
Propulsive mean power (W/kg)	−0.056 (−0.499, 0.387)	0.804	−0.085	0.001	(−3.006, 2.783)
Propulsive phase time (s)	−0.006 (−0.015, 0.003)	0.170	−0.298	0.023	(−0.050, 0.038)

Notes: bolded p values indicate significance; * = small effect size, ** = medium effect size, *** = large effect size.

). When interpreting both the Cohen’s d and η², there were large effect sizes for jump height, net impulse, propulsive mean power, propulsive peak power, and RSImod, as well as medium effect sizes for braking mean power. The other dependent variables, including propulsive phase time, exhibited small to medium effect sizes.

The expected difference in jump height between the first and second set was not significantly different based on whether or not subjects had pickleball matches prior to the first set (p = 0.728), and it was not significantly different for the propulsive phase time (p = 0.232), time to takeoff (p = 0.251), net impulse (p = 0.789), braking mean power (p = 0.336), propulsive mean power (p = 0.240), and RSImod (p = 0.460). When compared to the first set, the expected propulsive peak force increased by 1.2 N/kg among those who did not have prior matches, but only by 0.3 N/kg among those who had prior matches (p = 0.006), and the expected propulsive peak power increased by 2.43 W/kg among those who did not have prior matches and by 1.72 W/kg among those who did have prior matches (p = 0.019).

Most subjects showed positive differences for jump height, propulsive mean power, net impulse, propulsive peak power, and RSImod between the first and second set due to the learning effect, but it was common to observe both negative and positive differences between the second and third set after the learning effect disappeared (Bland–Altman plots are provided in Supplemental Figure S1).

Among those who did not have prior matches, we found that jump height essentially did not change at all from jump to jump within the first set. However, several other dependent variables exhibited a significant improvement with a medium to large effect size within the first set of jumps, indicating a learning effect on most variables (see

Table 4. Estimated slope (mean change per repetition in the first set; 95% CI), p-value, and effect size (η²) among those who did not have a match prior to the first set (n = 11) and among those who had a prior match (n = 21).

Among Those Who Did Not Have a Prior Match (n = 11)	Estimate (95% CI)	p-Value	η2
Jump height (m)	0.000 (−0.003, 0.003)	0.976	<0.001
Net impulse (N.s/kg)	0.014 (−0.019, 0.046)	0.407	0.023 *
Time to takeoff (s)	−0.036 (−0.060, −0.012)	0.006	0.227 ***
RSImod (no units)	0.008 (0.001, 0.015)	0.033	0.143 ***
Braking mean power (W/kg)	−0.289 (−0.547, −0.028)	0.035	0.139 **
Propulsive peak power (W/kg)	0.221 (−0.108, 0.550)	0.192	0.056 **
Propulsive peak force (N/kg)	0.372 (0.129, 0.613)	0.005	0.236 ***
Propulsive mean power (W/kg)	0.243 (−0.024, 0.510)	0.080	0.098 **
Propulsive phase time (s)	−0.005 (−0.009, −0.001)	0.012	0.192 ***
Among Those Who Had a Prior Match (n = 21)	Estimate (95% CI)	p-Value	η2
Jump height (m)	0.001 (−0.001, 0.004)	0.327	0.017 *
Net impulse (N.s/kg)	0.022 (−0.005, 0.049)	0.115	0.044 *
Time to takeoff (s)	−0.000 (−0.020, 0.019)	0.989	<0.001
RSImod (no units)	0.004 (−0.002, 0.009)	0.174	0.033 *
Braking mean power (W/kg)	−0.201 (−0.420, 0.015)	0.073	0.057 **
Propulsive peak power (W/kg)	0.174 (−0.111, 0.460)	0.233	0.026 *
Propulsive peak force (N/kg)	0.131 (−0.120, 0.385)	0.308	0.018 *
Propulsive mean power (W/kg)	0.166 (−0.027, 0.360)	0.096	0.049 *
Propulsive phase time (s)	−0.001 (−0.006, 0.003)	0.531	0.007

Notes: bolded p values indicate significance; * = small effect size, ** = medium effect size, *** = large effect size.

). Note that for net impulse, propulsive mean power, and propulsive peak power, even though the p values were not significant, the effect sizes were small to medium. In general, for each respective variable, the statistical significance were weaker and estimated effect sizes were smaller among those who had prior matches.

Objective 2

After the first jump set, there is neither evidence of a statistically significant further learning effect nor neuromuscular fatigue in any of the dependent variables (

Table 5. Estimated slope (mean change per set after the first set; 95% CI), p-value, and effect size (η²).

	Estimate (95% CI)	p-Value	η2
Jump height (m)	−0.001 (−0.002, 0.000)	0.246	0.009
Net impulse (N.s/kg)	0.004 (−0.010, 0.017)	0.617	0.002
Time to takeoff (s)	0.003 (−0.007, 0.013)	0.586	0.002
RSImod (no units)	−0.001 (−0.003, 0.002)	0.692	0.001
Braking mean power (W/kg)	−0.093 (−0.192, 0.007)	0.068	0.023 *
Propulsive peak power (W/kg)	−0.073 (−0.215, 0.069)	0.316	0.007
Propulsive peak force (N/kg)	0.027 (−0.097, 0.151)	0.669	0.001
Propulsive mean power (W/kg)	−0.046 (−0.180, 0.089)	0.507	0.003
Propulsive phase time (s)	0.001 (−0.002, 0.004)	0.609	0.002

Notes: * = small effect size.

).

Objective 3

After the familiarization (i.e., after the first set), the ICC values were 0.9 or above for jump height, net impulse, propulsive mean power, propulsive peak power, and RSImod, which indicates excellent reliability (ICC > 0.9; [51]), and these variables were shown to be highly reliable within each of the first four jump sets. The ICC values were about 0.75 or above for propulsive peak force, propulsive phase time, and braking mean power, which indicate good reliability according to Koo and Li [51]. The most unreliable variable was time to takeoff (

Table 6. Estimated ICC (95% CI), SEM (95% CI), and CV% (95% CI) after the first set and for the first, second, third, and fourth set, separately.

		ICC	SEM	CV%
Jump height (m)	After first set (n = 26)	0.986 (0.973, 0.992)	0.009 (0.008, 0.010)	6.687 (4.877, 8.496)
	First set only (n = 32)	0.969 (0.946, 0.979)	0.012 (0.011, 0.014)	7.822 (5.890, 9.753)
	Second set only (n = 26)	0.985 (0.957, 0.992)	0.009 (0.008, 0.011)	6.018 (4.314, 7.723)
	Third set only (n = 12)	0.986 (0.961, 0.994)	0.006 (0.005, 0.008)	4.982 (2.735, 7.229)
	Fourth set only (n = 7)	0.980 (0.914, 0.994)	0.008 (0.006, 0.012)	6.343 (2.093, 10.59)
Net impulse (N.s/kg)	After first set	0.971 (0.939, 0.984)	0.102 (0.092, 0.113)	4.295 (3.360, 5.231)
	First set only	0.945 (0.895, 0.965)	0.127 (0.111, 0.150)	4.592 (3.635, 5.549)
	Second set only	0.968 (0.937, 0.983)	0.105 (0.091, 0.125)	4.024 (3.006, 5.041)
	Third set only	0.971 (0.912, 0.988)	0.081 (0.066, 0.107)	3.254 (1.943, 4.565)
	Fourth set only	0.972 (0.914, 0.991)	0.082 (0.062, 0.118)	3.591 (1.273, 5.908)
Time to takeoff (s)	After first set	0.637 (0.462, 0.745)	0.085 (0.077, 0.095)	11.85 (9.555, 14.14)
	First set only	0.733 (0.595, 0.835)	0.095 (0.083, 0.111)	10.23 (8.115, 12.35)
	Second set only	0.548 (0.321, 0.691)	0.104 (0.090, 0.124)	11.39 (9.006, 13.78)
	Third set only	0.198 (0.000, 0.433)	0.076 (0.062, 0.100)	8.372 (3.757, 12.99)
	Fourth set only	0.499 (0.112, 0.799)	0.050 (0.038, 0.072)	5.970 (2.864, 9.076)
RSImod (no units)	After first set	0.955 (0.923, 0.970)	0.023 (0.021, 0.026)	13.52 (10.91, 16.13)
	First set only	0.922 (0.864, 0.950)	0.026 (0.022, 0.030)	13.50 (10.99, 16.01)
	Second set only	0.940 (0.873, 0.966)	0.027 (0.023, 0.032)	12.66 (9.725, 15.60)
	Third set only	0.954 (0.876, 0.975)	0.016 (0.013, 0.021)	8.222 (4.291, 12.15)
	Fourth set only	0.964 (0.867, 0.989)	0.015 (0.011, 0.021)	8.347 (4.527, 12.17)
Braking mean power (W/kg)	After first set	0.884 (0.801, 0.930)	0.807 (0.732, 0.898)	−11.98 (−14.90, −9.057)
	First set only	0.773 (0.657, 0.865)	1.047 (0.912, 1.230)	−13.61 (−16.22, −11.00)
	Second set only	0.863 (0.721, 0.921)	0.903 (0.779, 1.073)	−11.55 (−14.77, −8.330)
	Third set only	0.858 (0.648, 0.952)	0.647 (0.523, 0.848)	−8.787 (−12.15, −5.422)
	Fourth set only	0.930 (0.662, 0.972)	0.484 (0.370, 0.699)	−8.373 (−13.10, −3.643)
Propulsive peak power (W/kg)	After first set	0.984 (0.968, 0.990)	1.054 (0.957, 1.174)	3.411 (2.588, 4.234)
	First set only	0.967 (0.941, 0.981)	1.324 (1.153, 1.555)	4.036 (3.030, 5.043)
	Second set only	0.984 (0.971, 0.991)	1.018 (0.879, 1.210)	2.872 (1.980, 3.764)
	Third set only	0.975 (0.930, 0.991)	0.969 (0.784, 1.270)	2.906 (1.953, 3.858)
	Fourth set only	0.982 (0.923, 0.993)	0.843 (0.645, 1.217)	2.728 (1.531, 3.926)
Propulsive peak force (N/kg)	After first set	0.873 (0.741, 0.922)	0.961 (0.872, 1.069)	4.515 (3.517, 5.512)
	First set only	0.782 (0.636, 0.874)	1.150 (1.001, 1.350)	4.868 (3.934, 5.801)
	Second set only	0.846 (0.737, 0.900)	1.054 (0.910, 1.253)	4.495 (3.457, 5.533)
	Third set only	0.741 (0.503, 0.870)	0.841 (0.680, 1.101)	3.498 (2.310, 4.685)
	Fourth set only	0.836 (0.432, 0.937)	0.760 (0.582, 1.098)	2.953 (0.961, 4.945)
Propulsive mean power (W/kg)	After first set	0.959 (0.929, 0.976)	0.983 (0.892, 1.094)	5.175 (3.828, 6.523)
	First set only	0.948 (0.911, 0.969)	0.966 (0.841, 1.134)	5.351 (4.175, 6.527)
	Second set only	0.950 (0.905, 0.971)	1.083 (0.934, 1.287)	4.988 (3.260, 6.716)
	Third set only	0.969 (0.899, 0.988)	0.629 (0.509, 0.825)	3.339 (2.106, 4.571)
	Fourth set only	0.971 (0.906, 0.989)	0.657 (0.503, 0.949)	3.805 (2.270, 5.339)
Propulsive phase time (s)	After first set	0.743 (0.579, 0.829)	0.023 (0.021, 0.026)	8.493 (5.775, 11.21)
	First set only	0.870 (0.802, 0.920)	0.019 (0.017, 0.023)	7.367 (6.029, 8.705)
	Second set only	0.635 (0.463, 0.773)	0.029 (0.025, 0.035)	8.718 (5.714, 11.72)
	Third set only	0.859 (0.670, 0.937)	0.013 (0.010, 0.017)	4.963 (3.534, 6.392)
	Fourth set only	0.849 (0.472, 0.952)	0.013 (0.010, 0.019)	4.806 (2.628, 6.983)

).

For SEM and CV%, values of jump height, propulsive peak force, net impulse, braking mean power, and propulsive peak power were consistently lower in the second set when compared to the first set. The SEM and CV% values tend to be the lowest in the third or fourth set for all of the dependent variables of interest. The propulsive peak force, net impulse, propulsive mean power, and propulsive peak power remained at about a 5% or lower CV% for the first four sets (Table 6).

Objective 4

Under the mixed-effects model for each dependent variable, we found the following significant relationships. As would be expected, participant age was significantly related to jump height (p = 0.004), propulsive phase time (p = 0.003), time to takeoff, (p = 0.021), net impulse (p = 0.017), propulsive mean power (p = 0.040), and propulsive peak power (p = 0.005). The model estimated that the average jump height, propulsive phase time, time to takeoff, net impulse, propulsive average power, and propulsive peak power decreased by 0.025 m (about one inch), 0.014 s, 0.037 s, 0.162 N.s/kg, 1.112 W/kg, and 2.557 W/kg, respectively, per 10 years, given the other explanatory variables held constant in the models. The model estimated that men jump 0.042 m higher (p = 0.141) and generate greater peak power by 4.041 W/kg (p = 0.186) when compared to women, holding the other variables constant, but statistical significance was insufficient. The other explanatory variables in the model were not significantly related to the dependent variables except for the learning effect between the first and second set (

Table 7. Estimated regression parameters (95% CI), p-value, and effect size (η²) for the relationship between each explanatory variable and dependent variable.

		Estimate (95% CI)	p-Value	η2
Jump height (m)	Age (per 10 years)	−0.025 (−0.039, −0.011)	0.004	0.288 ***
	Sex (male)	0.042 (−0.008, 0.092)	0.141	0.085 **
	Aerobics (yes)	−0.009 (−0.066, 0.048)	0.778	0.003
	Resistance Training (yes)	−0.001 (−0.044, 0.043)	0.983	<0.001
	Frequency of PB (sessions per week)	0.013 (−0.006, 0.032)	0.234	0.056 *
	Number of Prior Matches	−0.001 (−0.002, 0.000)	0.133	0.008
	Learning Effect	0.014 (0.010, 0.017)	<0.001	0.155 ***
Net impulse (N.s/kg)	Age (per 10 years)	−0.162 (−0.277, −0.047)	0.017	0.208 ***
	Sex (male)	0.336 (−0.071, 0.744)	0.147	0.082 **
	Aerobics (yes)	−0.088 (−0.554, 0.378)	0.735	0.005
	Resistance Training (yes)	0.060 (−0.292, 0.413)	0.759	0.004
	Frequency of PB (sessions per week)	0.075 (−0.082, 0.231)	0.394	0.029 *
	Number of Prior Matches	0.002 (−0.013, 0.017)	0.786	<0.001
	Learning Effect	0.113 (0.073, 0.152)	<0.001	0.096 **
Time to takeoff (s)	Age (per 10 years)	−0.037 (−0.065, −0.010)	0.021	0.195 ***
	Sex (male)	0.072 (−0.024, 0.169)	0.187	0.069 **
	Aerobics (yes)	−0.041 (−0.152, 0.071)	0.515	0.017 *
	Resistance Training (yes)	−0.088 (−0.171, −0.004)	0.069	0.127 **
	Frequency of PB (sessions per week)	0.026 (−0.011, 0.063)	0.217	0.060 **
	Number of Prior Matches	0.003 (−0.009, 0.014)	0.674	0.001
	Learning Effect	−0.055 (−0.087, −0.024)	0.001	0.037 *
RSImod (no units)	Age (per 10 years)	−0.023 (−0.044, −0.002)	0.055	0.139 **
	Sex (male)	0.030 (−0.044, 0.104)	0.472	0.021 *
	Aerobics (yes)	0.006 (−0.079, 0.090)	0.907	0.001
	Resistance Training (yes)	0.026 (−0.038, 0.091)	0.460	0.022 *
	Frequency of PB (sessions per week)	0.010 (−0.018, 0.039)	0.522	0.017 *
	Number of Prior Matches	−0.002 (−0.005, 0.001)	0.208	0.005
	Learning Effect	0.032 (0.023, 0.040)	<0.001	0.148 ***
Braking mean power (W/kg)	Age (per 10 years)	0.351 (−0.117, 0.821)	0.186	0.069 **
	Sex (male)	−1.041 (−2.704, 0.622)	0.267	0.049 *
	Aerobics (yes)	0.446 (−1.457, 2.356)	0.675	0.007
	Resistance Training (yes)	−0.834 (−2.275, 0.603)	0.303	0.043 *
	Frequency of PB (sessions per week)	0.047 (−0.593, 0.684)	0.895	0.001
	Number of Prior Matches	−0.076 (−0.193, 0.046)	0.215	0.005
	Learning Effect	−0.526 (−0.844, −0.214)	0.001	0.035 *
Propulsive peak power (W/kg)	Age (per 10 years)	−2.557 (−4.078, −1.036)	0.005	0.271 ***
	Sex (male)	4.041 (−1.361, 9.443)	0.186	0.069 **
	Aerobics (yes)	−0.468 (−6.642, 5.702)	0.892	0.001
	Resistance Training (yes)	0.289 (−4.384, 4.963)	0.911	0.001
	Frequency of PB (sessions per week)	0.961 (−1.109, 3.033)	0.407	0.028 *
	Number of Prior Matches	−0.150 (−0.304, 0.005)	0.059	0.012 *
	Learning Effect	2.048 (1.643, 2.455)	<0.001	0.251 ***
Propulsive peak force (N/kg)	Age (per 10 years)	0.111 (−0.408, 0.630)	0.701	0.006
	Sex (male)	0.093 (−1.746, 1.933)	0.927	<0.001
	Aerobics (yes)	0.621 (−1.491, 2.730)	0.597	0.011 *
	Resistance Training (yes)	0.772 (−0.820, 2.363)	0.386	0.030 *
	Frequency of PB (sessions per week)	−0.500 (−1.206, 0.207)	0.210	0.061 **
	Number of Prior Matches	−0.061 (−0.206, 0.083)	0.407	0.002
	Learning Effect	0.712 (0.332, 1.094)	<0.001	0.043 *
Propulsive mean power (W/kg)	Age (per 10 years)	−1.112 (−2.045, −0.179)	0.040	0.157 ***
	Sex (male)	1.990 (−1.323, 5.304)	0.286	0.045 *
	Aerobics (yes)	−0.143 (−3.933, 3.644)	0.946	<0.001
	Resistance Training (yes)	0.913 (−1.953, 3.781)	0.568	0.013 *
	Frequency of PB (sessions per week)	0.472 (−0.798, 1.744)	0.506	0.018 *
	Number of Prior Matches	−0.101 (−0.235, 0.032)	0.140	0.007
	Learning Effect	1.240 (0.890, 1.593)	<0.001	0.140 ***
Propulsive phase time (s)	Age (per 10 years)	−0.014 (−0.021, −0.006)	0.003	0.288 ***
	Sex (male)	0.028 (0.001, 0.055)	0.076	0.118 **
	Aerobics (yes)	−0.018 (−0.050, 0.013)	0.301	0.041 *
	Resistance Training (yes)	−0.017 (−0.041, 0.007)	0.200	0.063 **
	Frequency of PB (sessions per week)	0.010 (−0.001, 0.020)	0.097	0.103 **
	Number of Prior Matches	0.001 (−0.002, 0.004)	0.463	0.002
	Learning Effect	−0.008 (−0.016, −0.000)	0.039	0.014 *

Notes: bolded p values indicate significance; * = small effect size, ** = medium effect size, *** = large effect size.

).

4. Discussion

Originally, our intention was to examine acute fatigue due to playing recreational doubles pickleball. Contrary to our expectation, we found no evidence of acute neuromuscular fatigue after a single match of doubles pickleball. Instead, players actually increased their force and power production after the first jump set and playing a match. In the following jump sets, subjects produced similar results, and no subsequent significant increase or decrease in CMJ performance was observed.

Our initial hypothesis was that one match of doubles pickleball would produce enough fatigue to see a significant decrease in CMJ performance. Upon examination of 14 subjects, we found that there was instead a significant increase after the match. This contradicted our hypothesis, and we changed the methodology to see how many matches it took for neuromuscular fatigue to occur. This lack of neuromuscular fatigue after a match was similar to a lack of change in CMJ performance seen after a simulated padel match [41]. The results from our expanded methods reinforced that subjects increased their power and force output after the first match. From the second set of jumps onward (i.e., after the first match), there was no significant increase or decrease in CMJ performance. Therefore, we are unable to quantify at this time how many matches it takes to see neuromuscular fatigue from pickleball; we can only conclude that we have no evidence of a fatigue effect after 1–6 matches of recreational doubles pickleball. Additionally, it is possible that other tests of performance besides the CMJ could show a fatigue effect within this range of games, whether that is a test of physiological performance (e.g., change of direction ability or upper body power) or skill (e.g., serve success or number of unforced errors).

Instead of determining fatigue effects from doubles pickleball, we instead uncovered details about learning effects and potentiation due to doubles pickleball. Among participants who had not played any matches prior to enrolling in the research, there were no differences between trials within the first set of jumps (i.e., baseline) for jump height (see Table 4), the ICC was extremely high (see Table 6), and the SEM and CV% were low (see Table 6). Thus, our research supports the conclusions of others—that jump height is highly reliable—even with only four trials for warm up and familiarization [22,23,29,30,31]. However, our study found significant and large improvements across trials within the first set of jumps for time to takeoff, RSImod, braking mean power, propulsive peak force, and propulsive phase time, indicating that participants were learning how to improve the performance of the CMJ, even if that performance was mostly in movement timing and technique (as may be indicated especially by the changes in time to takeoff, braking mean power, and propulsive phase time). The learning effect seen in these variables contradicts the prior research that concentrated only on jump height, which indicated there was no learning effect on the CMJ test.

The fact that jump height was virtually unchanged within the first jump set, while time to takeoff, RSImod, braking mean power, propulsive peak force, and propulsive phase time did change, may indicate changes in technique, even if the net result is the same. This result supports the recommendation of several research groups to examine more than just jump height when testing for vertical jump performance [20,25,26,32]. Future research should continue to investigate these detailed aspects of CMJ performance as related to pickleball, as changes in these factors (for example, braking power) may indicate worsening performance or even risk of injury [52,53,54]. Even if jump height does not exhibit a fatigue effect due to doubles pickleball because players are able to compensate, if some of these other factors do show a fatigue effect, that could help inform game management and training strategies for players.

There were significant and large increases in almost all outcomes, including jump height, between the first and second set of jumps, even when accounting for prior matches played. For individuals who had not played any matches prior to baseline testing, the increases in propulsive peak power and propulsive peak force were of significantly greater magnitudes than the increases for those who had played. Two possible hypotheses may explain what was observed. The first explanation is that the first match may have actually warmed up the subjects. The participants may have activated more motor units during their second set of jumps. Greater motor unit activation could have increased their ability to produce a higher, more forceful, and more powerful jump when compared to the first set of jumps. This explanation follows the well-known phenomenon of potentiation [55,56,57], with the muscle recruitment and load into the lower body caused by playing the pickleball match potentiating jump performance immediately post-match. The second explanation is that a learning effect may have occurred. Through repetition, participants may have learned how to jump more efficiently, even without specific coaching on aspects like trying to minimize the amortization phase. A combination of both factors may explain the total improvement seen from the first to second set of jumps. However, due to the lack of increases in CMJ performance seen in the prior literature in older adults just from repeated trials of jumps without any other stimulus (i.e., prior studies have shown no evidence of just learning effect in jump performance among older adults; [22,23,29,30,31]), it is likely that potentiation explains at least the majority of the inter-set increase seen in the present study. This is further supported by the results in the current study, including that participants who had already played a match prior to enrolling in the research did not exhibit significant changes in any outcome within their first set of jumps—the potentiation effect of their prior match likely overrode any inter-trial learning effect.

Overall, the continued significant improved performance over 12 jump repetitions (counting the four warm up trials provided and the whole first and second set of tested jumps) indicates that the question about if there is a learning effect, and if so, what is its magnitude, is still not answered. Determining the learning effect and when it ends, for sample averages and for a specific individual, is an important prerequisite for any research into fatigue; without a stable peak performance past learning, familiarization, or potentiation, it will be harder to detect fatigue. Future research, in any and all populations where CMJ testing may be valuable (e.g., athletes and/or elderly adults, as intersect in the current studied population), should have participants perform as many jumps as needed until clear learning and fatigue effects are established. Software and statistical methods, such as those created by our research group, could help in this endeavor [58].

This study has high ecological validity in terms of it representing the natural behavior of recreational pickleball players, but simultaneously, it has several limitations in determining the research question. First, rest periods between subsequent pickleball matches were not recorded or controlled; it is possible that the time people had to wait for a court to open up or chose to wait until they felt ready to play was sufficient for them to recover without demonstrating fatigue. Another potential explanation may be that, when people started to feel fatigue, they chose to stop playing and leave. Alternatively, the data may be able to be taken at face value, which indicates that partaking in up to six matches of recreational doubles pickleball does not cause fatigue, as indicated by the CMJ test. This potential explanation may be similar to a conclusion drawn in the similar sport of padel, where there seems to be sufficient rest time within a match to prevent neuromuscular fatigue, at least as measured by a CMJ [41]. To resolve this, future research could control match play timing and enforce a standard rest period between matches. Second, we were unable to quantify match load, which would be an important determinant of the stress and subsequent fatigue of playing pickleball. Basic match information like match length, number of serves, and number of rallies, as well as tracking the heart rate, training impulse (TRIMP) scores, accelerometery data, or step counts of players could serve to provide a basis of match load. Furthermore, in conjunction with addressing the first limitation, setting a standard duration of match time, coupled with standardizing the rest period between matches, could help in terms of research design. In practical terms, this may mean having matches last for 15 min, regardless of score, instead of the standard rules of playing to 11 points, then resting for 3–5 min before the next match. While these confines may be less representative of normal recreational play, they could create a more controlled setting for a valid investigation of fatigue due to doubles pickleball. Lastly, while the sample here was relatively heterogenous in terms of age, sex, and pickleball experience, there are still limits on how these results should be generalized—these results should not be generalized to higher levels of play beyond recreational/casual players who are mostly older, and they should not be generalized to singles pickleball.

5. Conclusions

The conclusion regarding our first research question is that pickleball players do exhibit learning effects on the CMJ test, even between trials within the first set, especially for parameters that may indicate technique of the jump, like braking time and RSImod. For our second research question, and contradictory to our hypothesis, we did not detect a fatigue effect after a single match of doubles pickleball, or even after multiple matches of doubles pickleball. Instead, we detected a significant and large learning effect for multiple outcomes from the CMJ test, both within the first set of jumps, and especially between the first and second jump sets (i.e., after one match). These results indicated both learning effects from repeated performance of the CMJ test, as well as potentiation effects on CMJ performance from playing one match of doubles pickleball. For our third research objective, the data did indicate the adequate reliability of CMJ performance after the first match of doubles pickleball. In answering our final objective, we found that age but not sex influenced CMJ performance. Overall, this study provides evidence of the need to measure more than just jump height or power when examining changes in CMJ performance in people playing doubles pickleball. Furthermore, it provides a methodological basis for future studies that can examine fatigue during pickleball, highlighting a potential need to control play and rest times to help control match load.