Validity, Reliability, and Sensitivity of the Brazilian Jiu-Jitsu Cardiorespiratory Fitness Test: A Methodological Approach Based on Combat Specificity

Featured Application

The BJJ-CRFT can be integrated into annual training programs to monitor cardiorespiratory performance throughout the competitive season. Its application provides coaches with a practical tool to adjust training load, plan tapering strategies, and prevent aerobic decline. In addition, the progressive structure of the BJJ-CRFT enables its use as a sport-specific aerobic training modality adaptable to different technical and physical levels. This versatility supports its application for individualized conditioning and evidence-based training management in combat sports.

Abstract

Brazilian jiu-jitsu (BJJ) is a combat sport that requires intermittent high-intensity actions, strong technical skills, strength, and aerobic capacity. Yet, there is limited evidence of validated sport-specific field protocols. This study aimed to determine the validity, reliability, and sensitivity of the BJJ Cardiorespiratory Fitness Test (BJJ-CRFT). Twenty-three trained practitioners (20 men and 3 women; age 34.4 ± 7.6 years) performed a cardiopulmonary exercise test on a treadmill and completed the BJJ-CRFT on two occasions, one week apart. Construct validity was examined using receiver operating characteristic (ROC) analysis, while concurrent validity was tested against maximal oxygen uptake (VO₂max) and maximal aerobic speed (MAS). Intra-session reliability was determined through the intraclass correlation coefficient (ICC) and the coefficient of variation (CV%). Main results showed a good discriminative ability (ROC: 0.82; 95% CI: 0.64–0.99, p = 0.001). Total repetitions in the BJJ-CRFT showed a large positive correlation with VO₂max (r = 0.66; 95% CI: 0.35–0.85, p = 0.0006) and a very large positive correlation with MAS (r = 0.72; 95% CI: 0.44–0.87, p = 0.0001). Key performance metrics, including guard passes and test duration, demonstrated excellent relative reliability (ICC = 0.99) and good absolute reliability (CV% = 4.4% and 3.6%), being sensitive to small changes. These results confirm that the BJJ-CRFT is a valid, reliable, and sensitive field test for monitoring aerobic adaptations and guiding training prescription in BJJ.

1. Introduction

Brazilian jiu-jitsu (BJJ) is a hand-to-hand combat sport that aims to takedown and subdue the opponent through grappling (levers), strangulations, dislocations, and immobilizations [1,2]. To achieve this, BJJ athletes engage in repeated high-intensity static and dynamic efforts interspersed with active pauses. In fact, through simulated combat tests, BJJ athletes maintained anaerobic heart rate, oxygen consumption, and perception of effort during the bouts [3,4]. Although anaerobic energy systems primarily support these efforts, aerobic metabolism plays a crucial role in sustaining performance during successive high-intensity bouts.

Maximal aerobic capacity—typically quantified as VO2max—is primarily determined by cardiorespiratory fitness (CRF), which reflects the integrated function between the cardiovascular, respiratory, and muscular systems during exercise. In BJJ, a higher CRF contributes to faster recovery between efforts, reduced accumulation of metabolic fatigue, and improved physiological efficiency [5]. Therefore, elevated CRF may become a key determinant of performance in BJJ, enhancing technical–tactical execution and delaying fatigue across multiple rounds [6,7].

Several sports-specific tests have been developed to evaluate BJJ performance. The Jiu-Jitsu Anaerobic Performance Test (JJAPT) [8,9] requires athletes to perform the maximum number of butterfly sweeps within a fixed period, making it highly demanding anaerobically but also prone to a progressive deterioration of technical execution as fatigue accumulates. The Brazilian Jiu-Jitsu Fitness Test (BJJFT) [10] is a general fitness-oriented protocol designed for BJJ. It consists of three consecutive exercises that mimic key components of combat: (i) double-leg takedowns performed for 20 s with 10 s rest, (ii) guard passes executed for 20 s with 20 s rest (repeated three times), and (iii) armlocks carried out for 20 s with 20 s rest. This entire sequence was repeated three times. Across all intervals, the objective is to achieve the maximum possible number of repetitions within each work period. Although the BJJFT captures multiple technical actions relevant to BJJ, its design primarily emphasizes anaerobic power and muscular endurance. The Guard Passing Test (GPT) is a protocol designed to elicit the maximal number of guard passes across six 1-minute bouts, interspersed with short recovery periods ranging from 10 to 30 s. This format aimed to replicate the intermittent and high-intensity nature of BJJ. The GPT demonstrated practical efficiency, emerging as a useful training tool. Moreover, it elicited physiological and perceptual responses comparable to those observed during simulated BJJ combats, supporting its ecological relevance [1]. More recently, the Special Brazilian Jiu-Jitsu Fitness Test–Takedown Zone (SBJJFT-TZ) [11] incorporated projection and locomotion drills and was specifically designed to evaluate special fitness in high-level competitive BJJ practitioners. The validity of this test was assessed by correlating its outcomes with a competition success ranking, showing very high associations across performance metrics (r = 0.73–0.88, p < 0.001). However, its reliability profile was limited, with a moderate variation in heart rate (CV < 10%) and high variation in throws and the performance index (CV = 20.82–32.25%). Despite its capacity to discriminate between practitioners with different competitive levels, it remains a non-incremental, anaerobic-oriented test.

Collectively, these protocols provide ecological validity by reproducing combat-specific techniques [12]. However, their reliance on maximum repetition counts within fixed time intervals often sacrifices technical execution and fails to capture the progressive cardiorespiratory demand. In contrast, the BJJ-CRFT introduces an incremental cadence-based design, ensuring movement quality while progressively increasing physiological stress, a structure that mirrors standardized laboratory assessments (e.g., treadmill CPET). This methodological distinction positions the BJJ-CRFT as a promising tool for assessing cardiorespiratory fitness in a sport-specific setting, potentially filling a critical gap and providing practical utility for monitoring training adaptations and prescribing individualized aerobic workloads.

To address this methodological gap, the present study aimed to determine the criterion validity, reliability, and sensitivity of a novel field-based protocol named BJJ-CRFT, specifically progressive physiological demand and practical applicability, to offer a reliable estimate of cardiorespiratory fitness sensitive to a sport’s unique effort profile. Beyond addressing this gap, this study contributes academically by expanding the evaluation paradigm in combat sports through the development of an incremental aerobic-oriented test. This finding has practical implications, as it offers coaches a tool for individualized monitoring and prescription in real training contexts.

2. Materials and Methods

2.1. Participants

This study included 23 BJJ practitioners (≥18 years), with a mean age of 34.4 ± 7.6 years (range: 18.7–50.5 years), height of 172.4 ± 6.9 cm (157.5–186.8 cm), and a body mass of 73.4 ± 12.4 kg (50.0–114.5 kg). Among them, 20 were males (34.3 ± 7.6 years, 173 ± 6.4 cm, 75.2 ± 11.8 kg) and 3 were females (33.9 ± 12.1 years, 168 ± 10.6 cm, 61.4 ± 11.5 kg). Based on their training orientation, the participants were classified as recreational (n = 12, including two females) or competitive (n = 11, including onefemale). The belt-rank distributions were: recreational (white: 2, blue: 5, purple: 1, brown: 0, black: 4) and competitive (white: 0, blue: 5, purple: 1, brown: 3, black: 2). The inclusion criteria were as follows: (i) Competitive practitioners: ≥2 years of BJJ experience, ≥3 weekly BJJ training sessions, and participation in at least one official national or international competition within the six months preceding testing; and (ii) Recreational practitioners: ≥1 year of BJJ practice and ≥2 weekly BJJ training sessions. The exclusion criteria for all practitioners included the presence of cardiovascular or respiratory diseases, active musculoskeletal injuries, or the use of medications or illicit substances. Additionally, records from participants who failed to meet at least two criteria of maximal effort during cardiopulmonary exercise test (CPET) or BJJ-CRFT were excluded: (i) no stabilization of oxygen consumption despite increased workload; (ii) failure to reach a maximum heart rate within 10 beats per minute of the theoretical value (220–age); (iii) a respiratory exchange ratio (RER) <1.10; (iv) a peak blood lactate concentration ([Lac]peak) <8 mmol/L; and (v) a rating of perceived exertion (RPE; Borg’s scale [0–10]) ≥ 9 [13,14,15,16]. All exercise protocols were approved by the Scientific Ethics Committee of the Universidad de Playa Ancha (CEC-UPLA), Approval Act No. 20-2023. This study adhered to the principles of the Declaration of Helsinki for human studies. Informed consent was obtained from all participants after a detailed explanation of the study.

2.2. Procedures

2.2.1. Anthropometric Measurements

Basic anthropometric measures were recorded: body mass was assessed using an electronic scale (SECA, model 803, Hamburg, Germany) with a sensitivity of 0.1 kg, and height was measured using a portable stadiometer (SECA, model 213, Hamburg, Germany) with a sensitivity of 0.1 cm.

2.2.2. Cardiopulmonary Exercise Test

A CPET was performed on a treadmill (Runrace, Technogym, Cesena, Italy) integrated with a metabolic cart (Cortex Metalyzer 3B, Cortex Biophysik, Leipzig, Germany). Resting heart rate (HRrest) was measured by averaging the last 5 min of a 10-min supine semi-Fowler position using a validated heart rate monitor (H10, Polar, Kempele, Finland). During CPET, the following parameters were continuously recorded: treadmill speed and incline, gas exchange (O₂ and CO₂), ventilation, heart rate, and RPE (Borg’s scale [0–10]) [17]. In addition, capillary blood lactate concentrations, at rest ([Lac]rest) and at the first minute post-exercise ([Lac]peak), were measured (Lactate PRO2, Arkray, Kyoto, Japan). The exercise began after 1 min of unloaded rest on the treadmill in a standing position. The initial exercise load was at a speed of 6 km/h with 1% positive incline. The load was only increased by 1 km/h every minute until voluntary exhaustion. During recovery, vital signs and exertional symptoms were monitored until normalization. VO₂max was the highest oxygen uptake average over a continuous 30-s interval. Maximal aerobic speed (MAS) was defined as the lowest speed at which VO₂max was achieved [18].

2.2.3. BJJ-Cardiorespiratory Fitness Test

The BJJ-CRFT was designed as an intermittent, multistage fitness test based on specific movements from Brazilian Jiu-Jitsu (https://www.protocols.io/view/validation-and-reliability-protocol-of-a-new-test-4r3l2q213l1y/v1, accessed on 24 November 2024). The protocol replicated progressive physiological demands like multistage shuttle run tests but adapted to BJJ’s technical and functional context [19,20]. The test consisted of repeated guard-passing sequences starting from a standardized position, with each sequence comprising three phases: (a) initial position, (b) forward guard pass, and (c) return to the starting position (see Figure 1). The test was divided into five stages with increasing cadences of 25, 30, 35, 40, and 45 steps per minute, corresponding to cumulative totals of 40, 60, 80, 160, and 180 repetitions, respectively. Each stage differed in duration (106, 135, 156, 280, and 279 seconds), thereby ensuring a progressive rise in physiological demand. The detailed per-stage progression (cadence, repetitions, and duration) is presented in Appendix A,

Table A1. Stage-by-stage structure of the Brazilian Jiu-Jitsu Cardiorespiratory Fitness Test.

Stage	Cadence (rep/min)	Guard Passes (n)	Duration (s)
1	25	40	106
2	30	60	135
3	35	80	156
4	40	160	280
5	46	180	279

.

To ensure standardization, a target zone was marked on the floor between 110 and 130 cm from the start line, which had to be reached with the lead leg on each step. This zone was delineated using a durable white adhesive tape (5 cm wide) to withstand repeated contact and provide a strong visual contrast against the gray tatami surface under controlled indoor lighting conditions. The visible strip creates a clear and consistent boundary for foot placement, facilitating reliable termination decision-making. Transitions between stages included 5-s recovery pauses. The test was terminated if the participant (i) could not maintain the cadence imposed by the auditory signal; (ii) failed to reach the target zone in three consecutive attempts; or (iii) voluntarily stopped due to fatigue. During execution, HR (H10 sensor, Polar, Kempele, Finland), [Lac] peak (Lactate PRO2, Arkray, Kyoto, Japan), and RPE (Borg’s scale [0–10]) were measured. The test was considered maximal if HRmax was ≥90% of the theoretical value (220-age), [Lac]peak was ≥8.0 mmol/L-1, or RPE (Borg’s scale [0–10]) was ≥9 [15,16,17].

2.3. Experimental Design

Participants underwent 3 consecutive assessments, spaced 6–7 days apart: two BJJ-CRFT sessions and one CPET. The order of the CPET and BJJ-CRFT was randomized. Before the evaluations, a 3-day familiarization period was conducted on the mat, which included a simulated test covering the protocol instructions (e.g., stage transitions), perceived exertion monitoring, and technical execution details (i.e., guard-passing steps). Each familiarization session lasted approximately 90 minutes and focused on movement standardization, adherence to cadence with auditory signals at different rhythms, and correction of guard pass length across the three required phases (initial position, forward pass, and return). The participants received continuous feedback to ensure consistent performance. As the guard pass drill is a commonly used technique in Brazilian Jiu-Jitsu training, the participants were already familiar with its mechanics. No pre-test dataset was collected to quantify potential learning effects during the familiarization period.

2.4. Statistical Analysis

Data symmetry and normality assumptions were verified using skewness and kurtosis coefficients, complemented by the Shapiro–Wilk test. Variables meeting normality criteria were reported as mean ± standard deviation with minimum and maximum values, whereas non-normally distributed variables were presented as median and interquartile range (IQR). For between-group comparisons (competitive vs. recreational), independent-sample t-tests were applied when normality was confirmed; otherwise, the Mann–Whitney U test was used for non-normal data. For repeated-measures comparisons (e.g., CPET vs. BJJ-CRFT within the same participants), paired-samples t-tests were applied when data met normality assumptions, and the Wilcoxon signed-rank test was used otherwise. Effect sizes (ES) were calculated according to the test employed: Cohen’s d (with Hedges’ g correction for small samples) for independent-sample t-tests, Cohen’s d for paired-sample t-tests, and rank-biserial correlation (r_Sb) for Wilcoxon and Mann–Whitney tests. Cohen’s d (and Hedges’ g) were interpreted as trivial (<0.2), small (0.21–0.60), moderate (0.61–1.20), large (1.21–2.0), and very large (2.1–4.0), whereas r_Sb thresholds were small (0.10), moderate (0.30), and large (0.50) [9,21]. Construct (discriminative) validity was assessed using Receiver Operating Characteristic (ROC) curve analysis, with an area under the curve (AUC) greater than 0.70 considered indicative of acceptable discriminative capacity, and 95% confidence intervals were reported to emphasize estimation precision [9,12]. Concurrent validity was evaluated using Pearson’s correlation coefficient, given that all bivariate relationships between repetitions, VO₂max, and MAS satisfied normality assumptions according to the Shapiro–Wilk test (p > 0.05). Correlation magnitudes were interpreted using established thresholds (trivial < 0.10, small 0.10–0.29, moderate 0.30–0.49, large 0.50–0.69, very large 0.70–0.89, and excellent > 0.90) [16,22]. In addition, to address concerns regarding statistical power, a priori sensitivity analyses were performed using G*Power 3.1.9.7 [23], indicating that with n = 23 and α = 0.05, the minimal detectable effect at 80% power corresponded to r ≈ 0.49 for one-tailed Pearson correlations (directional hypothesis: higher aerobic capacity associated with a greater number of guard passes), d_z ≈ 0.61 for paired-sample t-tests, and Hedges’ g ≈ 1.23 for independent-sample comparisons. These thresholds indicate that the study was sufficiently powered to identify moderate-to-large effects, although the power to detect small effects was limited. Relative reliability was determined via the intraclass correlation coefficient (ICC), classified as poor (<0.50), moderate (0.50–0.75), good (0.75–0.90), and excellent (>0.90). Absolute reliability was assessed using the coefficient of variation (CV%), with CV% < 5.0 considered good, values <10% acceptable, and values >10% regarded as poor [20,22,24]. Sensitivity was analyzed via the typical error of measurement (TE) and the smallest worthwhile change (SWC). The SWC was derived as 0.2, 0.6, and 1.2 times the between-subject standard deviation from the first trial, corresponding to small, moderate, and large effect thresholds. The classification was defined as marginal (TE > SWC), acceptable (TE = SWC), or good (TE < SWC) depending on the ability to detect meaningful changes [12,25,26]. The minimum detectable change (MDC) was calculated using the formula MDC = TE × 1.96 × √2, where the multiplier 1.96 corresponds to the 95% confidence interval, a conventional threshold to ensure that the observed changes exceed random error with high certainty. This approach has been previously applied in sports science to quantify the smallest detectable changes in performance [9,22]. Statistical significance was set at p < 0.05. All analyses were performed using JASP software (version 0.19.3, Amsterdam, Netherlands), Microsoft Excel 365 (Microsoft Corp., Redmond, WA, USA), and GraphPad Prism (version 10.5.0, GraphPad Software, San Diego, CA, USA) for figure generation.

3. Results

3.1. Sample Characteristics, Performance, and Internal Load: BJJ-CRFT vs. CPET

The participants had an average BJJ training experience of 7.0 ± 4.9 years and were categorized by belt rank as follows: 2 white belts, 10 blue belts, 2 purple belts, 3 brown belts, and 6 black belts. Comparative analysis between CPET and BJJ-CRFT (

Table 1. Cardiovascular, Ventilatory, and Metabolic Variables at Rest and During Exercise in CPET and BJJ-CRFT.

Variable	CPET	BJJ-CRFT	Statistic (t/Z)	p-Value	ES (d/rSb)	ES Rating
HRrest (bpm)	63.9 ± 8.6 (44.0–82.0)	64.3 ± 9.2 (41.0–81.0)	−0.42	0.68	−0.08	trivial
HRmax (bpm)	185 ± 11.4 (152–203)	182 ± 10.3 (156–204)	1.99	0.06	0.15	trivial
HRmean (bpm)	155 ± 12.2 (126–170)	162 ± 9.9 (138–184)	−4.42	0.001 **	−0.60	small
∆HRR60s (bpm)	38.9 ± 8.6 (23–61)	38.9 ± 11.3 (9.4–55.5)	−0.06	0.96	−0.01	trivial
[Lac]rest (mmol·L−1)	1.3 [1.2–1.4]	1.3 [1.3–1.7]	−1.16	0.26	−0.35	moderate
[Lac]peak (mmol·L−1)	10.1 ± 2.5 (6.6–14.9)	8.7 ± 1.8 (6.1–13.4)	2.47	0.02 *	0.67	moderate
VO2max (L·min−1)	3.6 ± 0.6 (2.29–4.39)	—	—	—	—	—
VO2max (ml·kg−1·min−1)	49.2 ± 6.1 (31.0–62.0)	—	—	—	—	—
RER	1.13 ± 0.1 (1.06–1.28)	—	—	—	—	—
Test duration (s)	532.5 ± 92.9 (322–730)	431.3 ± 112.4 (210–734)	5.80	0.001 **	0.92	moderate
RPE (au)	9.5 ± 0.5 (9.0– 10.0)	8.8 ± 1.0 (6.0–10.0)	3.76	0.001 *	0.91	moderate
MAS (km·hr−1)	13.6 ± 1.7 (10.0–17.0)	—	—	—	—	—
Peak velocity (km·h−1)	14.2 ± 1.6 (11.0–17.0)	—	—	—	—	—
Guard passing drill (rep)	—	202.8 ± 62.3 (87–380)	—	—	—	—

Legend: CPET: cardiopulmonary exercise test; BJJ-CRFT: Brazilian jiu-jitsu cardiorespiratory fitness test; HRrest: resting heart rate; HRmax: maximum heart rate; HRmean: mean heart rate; ∆HRR60s: heart rate recovery at 60 s; [Lac]rest: resting blood lactate concentration; [Lac]peak: peak blood lactate concentration; RER: respiratory exchange ratio; RPE: rate of perceived exertion (Borg’s scale [0–10]); VO₂max: maximal oxygen uptake; MAS: maximal aerobic speed; rep: number of repetitions. * p < 0.05; ** p < 0.001. ES: Effect size. Cohen’s d: trivial (<0.2), small (0.21–0.60), moderate (0.61–1.20), large (1.21–2.0), very large (2.1–4.0); rank-biserial correlation (r_Sb): small (0.10), moderate (0.30), large (0.50). —: variable not applicable.

) showed significant differences (p < 0.05) in several physiological variables. Higher values were recorded in CPET for lactate peak concentration ([Lac]peak, moderate ES = 0.67), test duration (moderate ES = 0.92), and perceived exertion (RPE, moderate ES = 0.91). Conversely, HRmean was significantly higher during BJJ-CRFT (small ES = −0.60). No significant differences were observed for HRrest (trivial ES = −0.08), HRmax (trivial ES = 0.15), or ∆HRR60s (trivial ES = −0.01), indicating a negligible effect sizes for these variables (p > 0.05).

3.2. Construct Validity (ROC Analysis)

Figure 2 show the ROC curves generated for VO₂max (Figure 2a) and the total number of repetitions (Figure 2b) performed in the BJJ-CRFT. The area under the curve (AUC) was higher for the number of repetitions (AUC = 0.82; 95% CI: 0.64–0.99, p = 0.001), while VO₂max showed an AUC of 0.59 (95% CI: 0.35–0.83, p = 0.417). Between-group comparisons (Appendix B,

Table A2. Between-group comparisons (recreational vs. competitive) for CPET and BJJ-CRFT variables.

Variable	Recreational (n = 12)	Competitive (n = 11)	Test t/U	p-Value	ES (95% CI)
CPET variables
VO2max (ml·kg−1·min−1)	50.0 (46.5–51.3)	50.0 (49.0–51.5)	78.5	0.46	rSb = 0.19 (−0.28 to 0.59)
MAS (km·h−1)	12.7 ± 1.6	14.5 ± 1.4	2.91	0.008 *	g = 1.17 (0.27–2.05)
Test duration (s)	484.2 ± 78.9	574.6 ± 83.0	2.68	0.014 *	g = 1.08 (0.19–1.95)
HRmax (bpm)	184.5 ± 12.0	183.6 ± 11.9	−0.17	0.864	g = −0.07 (−0.89–0.75)
[Lac]peak (mmol·L−1)	10.7 ± 2.9	9.6 ± 2.1	−1.07	0.297	g = −0.43 (−1.25–0.40)
BJJ-CRFT variables
Guard passing drill (reps)	180 (136.3–200.0)	220 (200.5–250.0)	108.0	0.010 *	rSb = 0.64 (0.26 to 0.84)
Test duration (s)	378.4 ± 104.8	489.0 ± 93.1	2.67	0.014 *	g = 1.07 (0.18–1.94)
HRmax (bpm)	183 ± 11.2	181.7 ± 9.7	−0.29	0.774	g = −0.12 (−0.93–0.70)
[Lac]peak (mmol·L−1)	8.8 ± 2.2	8.5 ± 1.4	−0.37	0.718	g = −0.15 (−0.97–0.67)

Depending on distribution, data are expressed as mean ± SD or median (IQR). Statistical tests: Student’s t-test (t) or Mann–Whitney U test (U), as appropriate. Effect sizes (ES) are reported as Hedges’ g for parametric comparisons and rank-biserial correlation (r_Sb) for non-parametric tests, with 95% confidence intervals (CI). * p < 0.05 was considered statistically significant. Hedges’ g was interpreted as trivial (<0.20), small (0.21–0.60), moderate (0.61–1.20), large (1.21–2.0), and very large (2.1–4.0), whereas r_Sb thresholds were small (0.10), moderate (0.30), and large (0.50).

) further reinforced these findings: VO₂max did not significantly differ between the recreational and competitive groups (p = 0.46; r_Sb = 0.19, 95% CI: –0.28 to 0.59), whereas the total number of repetitions was markedly higher in competitors (220 [200.5–250.0]) than in to recreational practitioners (180 [136.3–200.0]) (p = 0.010; r_Sb = 0.64, 95% CI: 0.26–0.84), indicating a large effect. This significant difference in repetitions underscores their superior discriminative capacity, which aligns with the higher AUC observed in the ROC analysis. Complete descriptive and inferential statistics are provided in Appendix B, Table A2.

3.3. Concurrent Validity

Regarding concurrent validity (Figure 3), the total number of repetitions in the BJJ-CRFT exhibited a large positive correlation with VO₂max (r = 0.66; 95% CI: 0.35–0.85, p = 0.0006). The Shapiro–Wilk test indicated bivariate normality for VO₂max and repetitions (W = 0.93, p = 0.075), justifying the use of Pearson’s correlation. Similarly, repetitions showed a very large positive correlation with the MAS (r = 0.72; 95% CI: 0.44–0.87, p = 0.0001), with Shapiro–Wilk confirming normality (W = 0.97, p = 0.67). These findings highlight that aerobic capacity, particularly MAS, is strongly associated with BJJ-CRFT performance.

3.4. Reliability and Sensitivity

The test–retest analysis (

Table 2. Reliability metrics and practical usefulness of BJJ-CRFT physiological and performance variables.

Variable	Test	Retest	ICC3,1 (95% CI)	CV% (95% CI)	TE	SWC0.2	SWC0.6	SWC1.2	MDC95%
Guard passing drill (reps)	202.9 ± 62.3	206.0 ± 65.7	0.99 (0.97 to 1.00)	4.4 (3.0 to 5.6)	6.3	12.3	36.6	73.1	17.4
Test duration (s)	431.3 ± 112.4	438.0 ± 117.5	0.99 (0.97 to 1.00)	3.6 (2.5 to 4.7)	11.0	22.0	65.9	131.9	30.4
HRrest (bpm)	64.3 ± 9.2	62.9 ± 7.8	0.80 (0.58 to 0.91)	8.3 (5.7 to 10.9)	3.8	1.8	5.4	10.8	10.4
HRmean (bpm)	162.2 ± 9.9	163.9 ± 9.5	0.85 (0.68 to 0.93)	3.1 (2.1 to 4.1)	3.6	1.9	5.8	11.6	10.1
HRmax (bpm)	182.4 ± 10.3	182.8 ± 10.6	0.95 (0.88 to 0.98)	1.8 (1.2 to 2.4)	2.4	2.0	6.0	12.1	6.6
∆HRR60s (bpm)	38.9 ± 11.4	41.2 ± 9.9	0.84 (0.65 to 0.93)	13.9 (9.6 to 18.2)	4.0	2.2	6.7	13.3	10.9
[Lac]rest (mmol·L−1)	1.4 ± 0.3	1.5 ± 0.4	0.76 (0.51 to 0.89)	15 (10.3 to 19.7)	0.2	0.1	0.2	0.4	0.4
[Lac]peak (mmol·L−1)	8.7 ± 1.8	8.9 ± 1.7	0.51 (0.13 to 0.75)	19.5 (13.4 to 25.6)	1.2	0.4	1.1	2.1	3.4
RPE (au)	8.8 ± 1.0	9.0 ± 1.0	0.62 (0.29 to 0.82)	9.4 (6.5 to 12.3)	0.6	0.2	0.6	1.2	1.6

Legend: ICC₃,₁: intraclass correlation coefficient (two-way mixed model, single measurement); interpretation: poor (<0.50), moderate (0.50–0.75), good (0.75–0.90), excellent (>0.90). CV%: coefficient of variation; interpretation: good (<5.0%), acceptable (<10.0%), poor (>10.0%). TE: typical error of measurement. SWC: smallest worthwhile change, calculated as 0.2 (SWC _0.2), 0.6 (SWC _0.6), and 1.2 (SWC _1.2) times the between-subject standard deviation. Sensitivity: TE < SWC = good; TE ≈ SWC = acceptable; TE > SWC = marginal. MDC₉₅%: minimum detectable change at 95% confidence (TE × 1.96 × √2). HRrest: Resting heart rate; HRmean: Mean heart rate; HRmax: maximum heart rate; ∆HRR60s: heart rate recovery at 60 s; [Lac]rest: resting blood lactate concentration; [Lac]peak: peak post-exercise blood lactate concentration; RPE: rate of perceived exertion (Borg’s scale [0–10]).

) showed excellent relative reliability for the external load metrics (guard passes and test duration) with ICC values of 0.99 (95% CI: 0.97–1.00). Absolute reliability was supported by low coefficients of variation (CV% = 4.4% and 3.6%, respectively), and both variables demonstrated good sensitivity to detect small performance changes, as their typical error (TE) was lower than the smallest worthwhile change (SWC). The minimum detectable change (MDC95%) was calculated at 17.4 repetitions for guard passes and 30.4 s for test duration, indicating the minimal improvements required to be considered real beyond measurement error.

Conversely, the internal load variables exhibited greater variability and lower sensitivity. Blood lactate peak concentration ([Lac]peak) showed moderate relative reliability (ICC = 0.51), with a CV% of 19.5% and an MDC95% of 3.4 mmol·L⁻¹. The perceived exertion (RPE) rating demonstrated moderate reliability (ICC = 0.62; CV% = 9.4%), with an MDC95% of 1.6 au. Heart rate metrics presented varied reliability levels: HRmean exhibited good reliability (ICC = 0.85; CV% = 3.1%; MDC95% = 10.1 bpm), HRmax showed excellent reliability (ICC = 0.95; CV% = 1.8%; MDC95% = 6.6 bpm), and ∆HRR60s showed good reliability (ICC = 0.84; CV% = 13.9%; MDC95% = 10.9 bpm). Despite the acceptable ICC and low variability in HRmean and HRmax, none of the internal load variables demonstrated sufficient sensitivity to detect small performance changes (TE > SWC), being only capable of detecting moderate to large changes in performance over longer periods.

4. Discussion

This study aimed to determine whether the specific BJJ-CRFT can assess the cardiorespiratory fitness of BJJ practitioners, thereby addressing an important methodological gap in the functional evaluation of this discipline. Overall, the BJJ-CRFT proved to be a valid, reliable, and sensitive test, with appropriate physiological correspondence to the actual demands of combat in BJJ and the ability to discriminate between different training levels.

Regarding physiological responses, participants in the BJJ-CRFT protocol reached an HRmean of 162 ± 9.9 bpm, equivalent to 89% of HRmax, and a [Lac]peak of 8.7 ± 1.8 mmol·L⁻¹ (range: 6.1–13.4 mmol·L⁻¹). These results align with those reported by Øvretveit (2018) and Rufino et al. (2024) in simulated and real matches, where mean heart rates exceeded 160 bpm (both >85% HRmax) and lactate concentrations fell within comparable ranges to the current literature [1,3,27]. This similarity supports the BJJ-CRFT’s capacity to accurately reproduce the physiological demands of the discipline, thereby reinforcing its validity. However, it is important to note that guard passing performance strongly depends on technical execution, suggesting that BJJ-CRFT reflects a combined interaction between aerobic capacity and technical proficiency.

In terms of construct validity, the ROC curve analysis demonstrated a good discriminative capacity to differentiate between recreational and competitive participants (AUC: 0.82; 95% CI: 0.64–0.99, p = 0.001), exceeding the AUC > 0.70 benchmark proposed by Chaabene et al. (2018) for discriminative validity [12]. Furthermore, this result contrasts favorably with the VO₂max obtained via a conventional treadmill protocol (AUC: 0.59; 95% CI: 0.35–0.83, p = 0.417), suggesting that physiological variables from sport-specific tests are more sensitive in identifying athletic profiles among BJJ participants. Similarly, da Silva et al. (2022) reported analogous findings in anaerobic-specific tests, highlighting the utility of these protocols in identifying sports experience and competitive level [9]. In contrast, our results diverge from those of Taati et al. (2022), who reported that the predicted VO₂max from a taekwondo-specific field test showed a strong discriminative capacity (AUC = 0.83, p = 0.003) to distinguish between national- and regional-level athletes [15]. Importantly, this underscores that while predicted aerobic power (VO₂max) derived from a discipline-specific protocol may achieve strong discrimination in certain combat sports, conventional treadmill VO₂max failed to do so in our study, reinforcing the relevance of the BJJ-CRFT’s sport-specific design.

Regarding criterion validity (i.e., concurrent validity), the results of the present study indicate that BJJ-CRFT performance, represented by the total number of guard passes, was positively and largely correlated with VO₂max obtained from CPET (r = 0.66; 95% CI: 0.35–0.85, p < 0.001). This association highlights the relevance of aerobic capacity. However, it also suggests that specific technical execution during the test is influenced by additional factors beyond VO₂max, which is consistent with the multifactorial nature of combat sports performance Nonetheless, MAS, defined as the minimal speed required to reach VO₂max, exhibited a very large correlation with the number of guard passes (r = 0.72; 95% CI: 0.44–0.87, p < 0.001), indicating that this metric may be a stronger predictor of performance in intermittent, sport-specific protocols such as the BJJ-CRFT. To further strengthen these findings, sensitivity analyses performed with G*Power showed that, given our sample size (n = 23), the study was adequately powered (1 − β = 0.80) to detect correlations of r ≥ 0.49. Therefore, both observed associations (r = 0.66 and r = 0.72) clearly exceeded this detection threshold, supporting the validity of the reported relationships. These results align with findings from combat-specific tests such as the Karate-Specific Test by Tabben et al. (2014), who also reported a strong association (r = 0.71) between technical performance and aerobic fitness [16].

In terms of reliability, the BJJ-CRFT demonstrated excellent reliability for performance metrics, with intraclass correlation coefficients (ICC > 0.99) and low coefficients of variation (CV% < 5) for the total test time and number of guard passes [24]. Similar results have been reported for repetition metrics in other BJJ-specific tests, such as the JJAPT, where da Silva et al. (2022) observed high relative reliability (ICC = 0.93) and low variability (CV% = 2.7) [9]. Additionally, HRmax and HRmean exhibited adequate reliability (ICC = 0.95 and 0.85, respectively) and low variability (CV% = 1.8 and 3.1, respectively). These findings confirm that the BJJ-CRFT is a reliable test with high stability and precision for specific metrics such as total time and repetitions.

Another notable aspect is the high sensitivity of BJJ-CRFT in detecting small changes in performance (TE < SWC). The MDC calculation indicated that changes exceeding 30.4 s in total test time and 17.4 guard pass repetitions can be considered objective performance improvements. This capability enables the longitudinal monitoring of aerobic training, aiding coaches and athletes in applied decision-making contexts. Conversely, variables such as HRmean, HRmax, ΔHRR60s, [Lac]peak, and RPE exhibited lower sensitivity, limiting their utility in detecting small changes. This limitation has been noted in previous studies, where RPE, despite being a widely used subjective marker of effort, did not reflect minor performance variations in specific tests [9].

Finally, while some variables (e.g., HRmean, HRmax, HRR60s, and RPE) demonstrated good reliability, they showed limited sensitivity in detecting small performance changes (TE > SWC). This limitation must be acknowledged, as it restricts the utility of fine-grained monitoring in short-term interventions. However, these variables effectively identified moderate-to-large changes, which are meaningful for monitoring medium- and long-term training programs. Considering that cardiorespiratory adaptations typically require at least 4 weeks to manifest, it is reasonable that HRmean and HRR60s may not capture acute variations but can still provide valuable information about sustained improvements over time [28,29,30].

Despite these promising results, this study had certain limitations. The sample size was small and unbalanced between male and female participants (n = 20 men; n = 3 women), which may restrict generalizability to broader BJJ populations, particularly women, adolescents, and older adults. In addition, although the BJJ-CRFT elicited physiological responses consistent with combat simulations, the lack of direct metabolic measurements (e.g., breath-by-breath gas exchange) limited comparisons with CPET in terms of ventilatory efficiency and anaerobic thresholds. Environmental factors such as room temperature, humidity, and tatami surface hardness were not systematically controlled and may also have influenced execution and physiological responses. However, these aspects reflect the complexity of simulating real training conditions and highlight the ecological validity of the test.

Sensitivity analyses using G*Power 3.1.9.7 indicated adequate power to detect moderate-to-large effects, although the ability to capture small effects wasa limitation. Furthermore, the absence of multivariate models precluded the simultaneous analysis of technical, physiological, and perceptual variables, which could enrich our understanding of performance determinants. Finally, the cross-sectional design prevented the evaluation of the BJJ-CRFT’s responsiveness to training-induced adaptations, underscoring the need for future longitudinal and sex-stratified studies. Such research, ideally with larger and more diverse samples and incorporating metabolic assessments, would further strengthen the construct and physiological validity of this protocol.

From an academic and applied perspective, BJJ-CRFT advances the methodological framework of sports science by introducing an incremental, field-based protocol specifically tailored for combat sports. Unlike traditional anaerobic-oriented tests, it offers a structured means of assessing cardiorespiratory fitness in a discipline where aerobic profiling has been systematically overlooked. Beyond its novelty, the test is practical, accessible, and closely aligned with the ecological demands of BJJ practice. As such, the BJJ-CRFT provides researchers, coaches, and practitioners with a valuable tool to bridge anaerobic and aerobic evaluations in combat sports, supporting both applied practice and methodological rigor in the sports sciences.

5. Conclusions

The BJJ-CRFT is a valid, reliable, and sensitive tool for assessing cardiorespiratory fitness in BJJ practitioners. Its performance metrics demonstrated high reproducibility and the ability to detect small changes, while its discriminative validity outperformed conventional tests. Beyond its methodological contribution, the BJJ-CRFT offers a practical instrument to monitor aerobic adaptations, individualize training loads, and track readiness across different preparation phases. Its incremental cadence-based structure ensures technical quality while progressively eliciting physiological stress, making itfor integrating physiological monitoring into sport-specific training environments. From an applied perspective, this protocol can support evidence-based decision-making in athlete development, periodization, and return-to-sport strategies.