In the 2021 Tokyo Olympics, climbing included three disciplines (speed-, lead-, and boulder climbing). Bouldering is performed on low walls (<6 m) with few, difficult, and often highly explosive moves (White and Olsen, 2010), whereas lead climbing is performed on higher walls (10–30 m) and consists of 20 to 50 moves with repeated sub-maximal force generation (Stien et al., 2021a). Speed climbing is performed on a slightly overhanging 15 m wall with a standardized route (Levernier et al., 2020). Success in climbing requires psychological, technical, and physical components (Vigouroux and Quaine, 2006; Baláš et al., 2012; Philippe et al., 2012). Among the physiological requirements, coaches and researchers highlight upper-body strength, power, and endurance as primary factors underpinning performance (MacLeod et al., 2007; Draper et al., 2011; Baláš et al., 2012). Despite noteworthy differences (Fanchini et al., 2013; Ozimek et al., 2017; Stien et al., 2019; Levernier et al., 2020), the three disciplines likely require partly overlapping requirements (Medernach et al., 2016). However, the tests and procedures used to measure these skills vary. Therefore, this mini-review aims to provide an overview of the climbing-specific tests, procedures and outcomes used to assess climbing performance and training effects.

A literature search was conducted, including the search terms “Climbing,” “Test,” “Assessment,” “Endurance,” “Strength,” “Force,” “Intermittent,” “Forearm” and “Finger.” Twenty-five relevant studies including climbers and describing at least one experimental testing procedure used to assess the physical characteristics of the study population were included in this mini-review. Please see Supplementary Materials 1, 2 for a more detailed description of the search strategy and screening process.

Climbing is characterized by intermittent contractions of the finger flexors. This leads to oxidative and non-oxidative metabolic demands which have been associated with climbing performance (Fryer et al., 2018; Michailov et al., 2018; Giles et al., 2021). The most common endurance tests used to examine these capacities include intermittent or sustained contractions of the finger flexors using climbing-holds (Philippe et al., 2012; Fryer et al., 2015; Michailov et al., 2018; Stien et al., 2019) or handheld dynamometers (Mermier, 2000; Limonta et al., 2015). In addition to finger-specific tasks, number of pull-ups using different holds (Vigouroux et al., 2018), and trunk muscle tests (Saeterbakken et al., 2018; Draper et al., 2021) have been used to examine climbing-related endurance. Recently, testing procedures mimicking climbing have been examined, including motorized climbing ergometers (treadwalls), bouldering, campus board, and lead climbing (Medernach et al., 2015b; Hermans et al., 2017; Baláš et al., 2021; Stien et al., 2021b). Additionally, oxygen-uptake and -saturation have recently been measured as local aerobic capacity of the finger flexors (Baláš et al., 2021). Finally, the critical-force model has recently been introduced to assess the break point of isometric finger flexor work and time to exhaustion (Giles et al., 2021).

The current consensus states that maximal and explosive strength in the fingers and upper-body are crucial factors for climbing performance (Horst, 2016; Sanchez et al., 2019; Saul et al., 2019). However, there is no agreement on how strength in the fingers and upper-body should be assessed. The applied methods vary in hold types, contraction form, body positioning, measuring techniques [e.g., time periods for calculating rate of force development (RFD)], execution (e.g., verbal instructions and duration), and joint angle and number of included joints.

Climbing performance is measured using graded boulders or routes which categorize the performance levels (Draper et al., 2016). However, concurrent improvements in climbing-tests and -performance are poorly described in the literature (Hermans et al., 2017), whereas the association between climbing-specific tests and climbing performance has been examined (Baláš et al., 2012; Fryer et al., 2015; Laffaye et al., 2016). Several climbing-specific tests and procedures have not been validated and reliability measurements of the tests are rarely reported. Furthermore, the current findings indicate that reliability data are more frequently reported than validity data. This presents a gap in the knowledge which should be addressed in future research. In addition, and despite the differences in climbing-style and physiological requirements (Fanchini et al., 2013; Fryer et al., 2017; Stien et al., 2019), specific tests for individual disciplines do not exist.

The available literature is challenged by the vast variety of applied endurance-, strength-, and power tests (Ozimek et al., 2016; Michailov et al., 2018; Levernier and Laffaye, 2019; Torr et al., 2020; Stien et al., 2021c). For example, this review revealed 13 trials that had implemented the intermittent forearm endurance test, and these provided nine different combinations of work-to-rest ratios and force thresholds (Table 1). Moreover, the study populations in various investigations range from non-climbers to higher-elite athletes. Hence, a very small portion of the possible climbers-and-tests combinations have been thoroughly examined. It is paramount that researchers strive to provide detailed descriptions of the included population and validity, reliability, and sensitivity measures of the tests applied in future research.

Although researchers may argue that some test set-ups are superior to others regarding reliability or specificity to climbing, the current available evidence could be too fragmented to support either position. Moreover, it is possible that choosing to optimize conditions for either specificity or validity will come at the cost of the other. For example, complex tests may provide conditions that mimic climbing more closely but could also increase the difficulty of reproducing similar results. Importantly, the complex nature of climbing renders it challenging to argue which test set-up is more climbing-specific. More descriptive studies such as biomechanical- (Cha et al., 2015), motion- (White and Olsen, 2010), and workload-analyses (Michailov, 2014) in climbing are needed to provide a basis for test recommendations.

Currently, reliability data has only been reported for a handful of protocols. Isometric endurance tests with sustained force generation (e.g., finger-hang or bent-arm hang) may be easy to conduct, but do not mimic the locomotion in climbing. Isometric intermittent tests to failure have a greater ecological validity, but there is no consensus in work-relaxation ratio, force threshold, hold size, or grip position. In addition to the varying work-to-rest ratios and force thresholds, different hold depths (10–30 mm), hold types (jug, gym bar), and grip positions (half-crimp or open-hand) have been used. The more promising tests are climbing to fatigue tests using reproducible routes or standardized walls (Medernach et al., 2015b; Baláš et al., 2021; Stien et al., 2021b). However, these tests suffer from limited research and the findings may not be generalizable to other disciplines or performance levels.

Based on the previously reported reliability data, one could speculate that hold size greatly influences the reliability of a test, regardless of task complexity and contraction form. For example, some of the smallest CV-values reported for power and isometric strength (1.0–6.6%) have been collected from tests that used either jug holds (Laffaye et al., 2014; Stien et al., 2021b) or a gym bar (Levernier et al., 2020). For maximal strength, Stien et al. (2021b) reported a 1.1% CV using jug holds, compared to 4.7% using a 23 mm rung. Shallower holds (~10–20 mm) have displayed CV-values between 7.8 and 31.3% (López-Rivera and González-Badillo, 2012; Ozimek et al., 2016; Stien et al., 2021c). Indeed, the fingers are likely the weakest link in the pulling apparatus and hold depth influences the biomechanical arm action during pulling movements (Vigouroux et al., 2018). Future studies should identify whether the climbing-specificity of a test is compromised by using large holds, or if large holds can maintain validity while increasing reliability.

Finally, dynamic tests are underrepresented in the literature (Table 2). Although climbing is characterized by isometric contractions of the finger flexors, the movements in the elbows and shoulders are often dynamic to produce vertical propulsion. Hence, one could argue that future research should focus more on dynamic strength in the upper-limbs of climbers, in addition to isometric strength in the finger flexors. Indeed, investigations using dynamic tests have demonstrated that such test set-ups are (1) reliable, (2) able to differentiate between performance levels and disciplines, and (3) sensitive enough to detect within- and between-groups differences following a training intervention (Laffaye et al., 2014; Ozimek et al., 2016; Levernier et al., 2020; Stien et al., 2021b).

Some recommendations can be made based on the finding of this mini-review. Importantly, the scarcity of relevant studies should be considered when interpreting the results, as well as the conflicting findings between studies. For isolated endurance tests, the force-time integral might be more useful compared to simply reporting the total work time (Rokowski et al., 2021). Moreover, the time to fatigue during sustained endurance tests has displayed moderate-to-strong correlations with red-point climbing performance, whereas the few studies that examined intermittent tests reported weak correlations with climbing performance (Baláš et al., 2021; Rokowski et al., 2021). For strength, more reliable results might be achieved by using isometric dynamometer tests (CV ≤ 10%) compared to fingerboard tests (CV ≤ 22.9%). The validity will likely differ depending on the test set-up (e.g., elbow angle, body positioning, and grip type) and study population (e.g., performance level or preferred discipline), but in general the seated, 90° constrained elbow set-up displayed the highest correlations with climbing performance (r = 0.60–0.84) (Philippe et al., 2012; Marcolin et al., 2020). Dynamic upper-body strength tests (e.g., pull-up) also revealed a high reliability (CV ≤ 7.7%), but such tests could potentially lack specificity to climbing compared to tests focusing on the finger flexors (Ozimek et al., 2016). Finally, high intensity, upper-body tests (i.e., powerslap and campus board reach) were investigated in two studies (Draper et al., 2011; Stien et al., 2021b) and displayed a strong relationship with climbing performance (r = 0.69–0.73).

Establishing reliable and valid testing procedures is essential for the field of climbing research. Today, no consensus exists regarding preferred sport-specific performance assessments. This study provides a brief overview of the applied endurance-, strength-, and power-tests, and highlights gaps in the literature. The findings of this mini-review revealed that numerous approaches to measuring climbing-related performance have been applied, but few have reported the reliability and validity of the tests. Hence, the current knowledge is fragmented as very few findings have been re-tested in subsequent studies with similar methodology. Moreover, poor descriptions of populations challenge performance level- and discipline-specific testing recommendations. Importantly, the pioneer work by Draper et al. (2016) needs to be acknowledged for first providing a numerical scale, making it possible to compare climbing performance across continents, and more recently attempting to establish a test battery (Draper et al., 2021). Hopefully, this mini-review will provide a useful overview of the scientific literature and inspire researchers to work toward agreeing upon common tests and procedures.

NS and AS wrote the first draft of the article and extracted the relevant data from the included studies. VA reviewed the data extraction and settled any disagreements between NS and AS. All authors contributed to the conceptualization and methodology. All authors provided critical reviews of the paper.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.