Do Physical Abilities of Brazilian Jiu Jitsu Athletes Differ According To Their Fighting Style?


Afranio de Sousa¹, Emerson Franchini², and Renato Barroso¹

1 School of Physical Education, Department of Sport Sciences, University of Campinas, Campinas, Brazil
2 Martial Arts and Combat Sports Research Group, Sport Department, School of Physical Education and Sport, University of São Paulo, São Paulo, Brazil

Do physical abilities of Brazilian jiu-jitsu athletes differ according to their fighting style?

Brazilian jiu-jitsu athletes are classified according to two fighting styles: pass fighters (PF) and guard fighters (GF). The aim of this study was to compare the flexibility, upper and lower limbs muscle power, maximal strength, and strength-endurance, and anaerobic power and capacity. Thirty-two BJJ athletes were divided into two groups according to their self-determined fighting style. Independent Student t-tests were used to compared data from GF and PF, and the level of significance was set at p ≤ 0.05. GF (32.7 ± 4.1 cm) presented higher (p = 0.03) CMJ height than PF (29.5 ± 4.0 cm). No other between-group significant differences were observed. Some GF jump upon their opponents to avoid standing fight while PF use jumps as a way to pass the guard which are mechanically different from the CMJ and SJ. Motor ability development is a consequence of long-term training process and is influenced by years of training, thus results suggest that strength and conditioning training was likely similar between GF and PF, showing that the combat itself does not affect physical abilities of BJJ fighters according to their fight style.

Keywords: grappling; combat sports; performance; physical fitness


Brazilian jiu-jitsu (BJJ) is a grappling combat sport, in which athletes are categorized considering their sex, body mass, age and grade (belt color). Although BJJ athletes start the match in the standing position, most of the combat is developed with both athletes on the ground. Athletes aim at having their opponent’s submission or at scoring more points than their opponent (Del Vecchio et al., 2007; Andreato et al., 2016; IBJJF, 2018)
Additionally, BJJ athletes can be classified according to their fighting styles: pass fighters or guard fighters. Pass fighters (PF) usually stand or kneel over their opponents, usually attacking with taking downs or passing the guard, whereas guard fighters (GF) prefer sitting or lying beneath their opponents, usually attacking swiping or submitting. In spite of athletes’ needs to perform movements according to their fighting style, few studies investigated the differences between PF and GF 蜉(Báez et al., 2014; Del Vecchio, Gondim, & Arruda, 2016; Lima et al., 2017).

Andreato, Lara, Andrade, & Branco (2017) suggested that maximal strength may be required for attacks and defenses, power for taking downs or actions performed on the ground, and the strength-endurance is important to maintain positions or to grip the gi (uniform). In consonance to Andreato et al. (2011), hamstrings and trunk flexibility is fundamental for BJJ athletes because it is needed for a good performance in some specific techniques and to contribute for the learning process of innumerous techniques and positions in the sport. However, so far it is not known whether PF and GF present differences in physical capacities that seem to be important for performance and different patterns or fight styles.

Therefore, the main goal of the present study was to compare hamstrings and trunk (lumbar) flexibility, muscular power of upper and lower limbs, isometric and dynamic strength-endurance of upper limbs, maximum strength of upper and lower limbs and anaerobic fitness of upper and lower limbs between PF and GF.

2. Materials and Methods
Thirty-two male BJJ athletes (age = 18 – 35 years old) volunteered to participate in this study. They were purple to black belt and had a training routine of 3-5 weekly training sessions, and at least two years of competitive experience. Athletes were separated into two groups according to self-selected fight style (PF or GF). This study was approved by the local research ethics committee and after being informed about all the risks, benefits and procedures all the participants signed as informed consent.

Table 1. Characteristics of BJJ athletes

2.1 Design

There was a minimum interval of 48 hours between testing days. There was a minimum interval of 5 minutes between each test except for the anaerobic fitness tests which were separated by 20 minutes.

Table 2. Tests order

2.2 Sit-and-reach Test

To evaluate hamstrings and trunk flexibility athletes were positioned seated with their knees completely extended and bare feet slightly apart and fully backed against a platform scale. After four maximum bending attempts, keeping the knees, elbows and wrists extended, they were instructed to reach and hold the position achieved for 5 seconds (Wells and Dillon, 1952).

2.3 Bench Press Throw

A linear encoder (20Hz, Peak Power, Cefise®, Nova Odessa, Brazil) was attached to the bar. The protocol used to identify optimal power load was that proposed by Loturco et al. (2017). In short, athletes were lying supine holding the bar with elbows fully extended, after the verbal command to begin they should perform four repetitions as fast as possible. The initial load was set at 30% of the individual’s body mass limited to a minimum of 20 kg (bar mass). In subsequent sets, 5% of the individual’s body mass was added. The test was stopped when mean power and peak power decreased compared to the previous set. The higher values observed in each set were retained for analysis.

2.4 Squat (SJ) and Countermovement Jump (CMJ)

In the SJ, a static position with hands on the hips and knees flexed at 90º was maintained for 2-s before the jump without any downward movement. For the CMJ, participants were told to keep their hands on the hips and to perform a downward movement followed by a complete extension of the lower limbs. All jumps were executed with the hands on the hips. Five attempts at each jump were performed interspersed by 15-s intervals. Jumps were performed on a contact platform (Jump System, Cefise®, Nova Odessa, Brazil). Average of five jumps was used for further analysis.

2.5 Maximal Isometric and Dynamic Pull-up Gi Endurance

Two strength-endurance tests were performed: isometric – athletes were required to hold on the gi rolled around the bar, with the elbow flexed and chin above their handgrip, and to sustain this position during the maximal possible time; dynamic – the same grip position was applied, but athletes performed the maximal number of repetitions from a fully extended to a fully flexed elbow position.

2.6 Maximum dynamic strength (1-RM) in Bench Press and Leg Press

Tests were performed according to recommendation of Brown and Weir (2001). After a specific warm-up consisting of 8 repetitions of the estimated load of 50% followed by 3 repetitions of 70% of the estimated load for 1RM, participants performed single attempts with progressive loads until they were not able to lift the load. For the bench press test, participants started lying on a bench, holding the bar with elbows fully extended, then they should lower the bar to touch the sternum and fully extend their elbows (CCI = 0.98 and Typical Error = 3.59 kg). For the leg press, athletes were positioned on the equipment, started with knees extended and should flex their knees until a 90º of flexion followed by complete extension (CCI = 0.99 and typical Error = 7.19 kg).

2.7 Upper-body and Lower-body Wingate Test

Each test was preceded by a 4-min warm-up with a 3-5-second sprints at the end of the each of the last three minutes on an arm ergometer (EB4100, Cefise®, Nova Odessa, Brazil) and a bicycle ergometer (Excalibur Sport, Lode®, Groningen, The Nederlands), for upper and lower limbs, respectively, with 1% of body mass. After the warm-up, participants rested for 3 minutes. All subjects remained seated and started from a complete standstill. After 3-second countdown the test was started, and athletes pedaled as fast as possible for 30 seconds. Loads for upper and lower limbs for the test were set at 5% and 10%, respectively. Verbal feedback as to the time remaining was provided at 15, 10, and 5 seconds remaining. Power output was measured at 1 Hz and stored for later analysis. Peak power was taken as the highest power value observed during the test and mean power the average value over the entire test duration.

2.8 Statistical analysis

Data normality was assured with Shapiro-Wilk test. Data are shown as mean and standard deviation. Student’s t-test for independent samples was used to compare PF and GF in all variables assessed. The 95% confidence intervals (CI) were calculated and a significance level was set at p ≤ 0.05. Effect size (ES) was calculated according to Cohen and classified based on thresholds <0.2, 0.2, 0.50, 0.8, 1.2,and 2.0, for very small, small, moderate, large, very large and huge 蜉(Cohen, 1988; Kolmogorov, 1933; Rhea, 2004; Sawilowsky, 2009; Smirnov, 1948; Student, 1908).


Table 2 displays the results for all the variables measured. No significant differences were observed for most of the physical abilities, except for CMJ, which GF jumped higher than PF.

Table 3. Flexibility, muscle power, strength-endurance and maximal strength in pass fighters (PF) and guard fighters (GF) Brazilian jiu-jitsu athletes


CMJ: Countermovement Jump; SJ: Squat Jump; 1RM: 1 Repetition Maximum; CI: Confidence Interval.


The aim of this study was to compare the flexibility, upper and lower limbs muscle power and strength, isometric and dynamic upper limbs strength-endurance, and upper- and lower-body anaerobic power and capacity between two BJJ fighting styles, PF and GF. Results indicated higher level of lower limbs muscle power in GF, as indicated by the CMJ height. Although no between-group difference was observed in SJ and in number of pull-ups performed (i.e., dynamic strength-endurance), the moderate ES favors GF. Both groups showed similar values for flexibility, upper limbs muscle power and isometric strength-endurance, upper- and lower-body maximum strength and anaerobic power and capacity.
GF and PF presented distinct combat behavior. GF athletes prefer to fight lying on the floor with flexed hips, while PF athletes maintain a standing position. In spite of the different positions that may impose a greater demand on the flexibility of hamstrings, both groups performed similar in the sit-and-reach test. Although flexibility seems to discriminate BJJ athletes of different competitive levels (Coswig, Neves, & Del Vecchio, 2011; Marinho, Andreato, Follmer, & Franchini, 2016), it did not distinguish GF from PF. It is possible that the sit-and-reach test was not specific to BJJ, as during combat athletes need dynamic and passive flexibility (i.e., the opponent applies force in the legs) and the test assess active (i.e., participants has to contract their muscles) and static flexibility (i.e., no movement). Alternatively, it may indicate low strength levels of hip flexors.

Upper limbs muscle power was not different between GF and PF. Even though few studies investigated upper limbs muscle power in BJJ athletes, 蜉da Silva, Simim, Marocolo, Franchini, & da Mota (2015) concluded that the peak power observed in a bench press power test was not different between beginners and advanced BJJ fighters. Considering that upper limbs muscle power does not discriminate athletes of different levels, it is not surprising the comparable muscle power level in athletes of similar level, but with different fighting style. It is also possible that the bench press (e.g., elbow extension and horizontal shoulder adduction), which was the exercise used in the test, does not reflect the mimic movements from combat (e.g., elbow flexion, shoulder extension and horizontal shoulder abduction). We were not able to change the exercise, but it is possible that the row exercise provides better estimates of muscle power specific to BJJ.

GF presented better results in CMJ than PF. It has been demonstrated that advanced athletes performed better in CMJ than beginners, which suggests that muscle power may play a role in athlete progression from beginner to proficient (Diaz-Lara, Monteiro, Garcia Garcia, & Abián-Vicén, 2014). Interestingly, in spite of moderate ES favoring GF, SJ was not different between fighting styles. Some GF jump upon their opponents to avoid standing fight, to decrease the risk of being taken down or to anticipate any GF opponent’s move to assume the guarding position first. On the other hand, PF also use jumps during the combat, but as a way to pass the guard. These jumps are usually performed projecting their own trunk towards the ground or opponent’s body with the hands usually placed on them with some kind of grip. Thus, in spite of also performing jumps during combats, they are mechanically different from the CMJ and SJ.

Even though upper limbs isometric and dynamic strength endurance levels can differentiate athletes from different competitive levels 蜉(Corrêa da Silva, Junior, de Moura Simim, Franchini, & da Mota, 2014; da Silva, Ide, de Moura Simim, Marocolo, & da Mota, 2014; da Silva et al., 2014; da Silva et al., 2013), results of these tests were similar between GF and PF. Lima, et al. (2017) revealed that PF presented higher isometric trunk extension endurance than GF because they need to tolerate their upper limbs and trunk being constantly pulled towards the ground by their opponents (generally a GF), i.e., GF fighting style seems to affect isometric trunk extension endurance of PF. In spite of no between-group difference, ES moderately favored GF.

Upper and lower limbs maximal strength is different in BJJ athletes of different levels 蜉(da Silva, et al., 2015; Marinho, et al., 2016). Considering that a BJJ combat starts with both athletes standing, but is developed with athletes on the ground, and maximal strength is mostly used in the ground fight specially for immobilization and submission escapes, it is conceivable that both groups should present similar level of strength.

The anaerobic fitness depends on anaerobic metabolism and it is as important as the muscle power for decisive moments in a combat because it can be required in fast and vigorous movements (Inbar, Bar-Or, & Skinner, 1996). Demands of the combat seem to be similar regardless of the fighting style, it is not surprising that both groups present comparable upper and lower limbs anaerobic fitness.

Motor ability development is a consequence of long-term training process and is influenced by years of training. Although, training characteristics of the athletes in this sample were not studied, it seems reasonable to suggest based on the results observed in this study that strength and conditioning training was similar between GF and PF, showing that the combat itself does not affect physical abilities of BJJ fighters according to their fight style.

Practical application

BJJ athletes with different fighting styles do not present different physical fitness characteristics. However, if a coach considers the physical abilities to be determinant to fighting performance, they are encouraged to design a strength and conditioning training routine to enhance a specific physical ability. To date it is unclear if improving physical abilities leads to improved performance.


Andreato, L. V., de Moraes, S. F., de Moraes Gomes, T. L., Esteves, J. D. C., Andreato, T. V., & Franchini, E. (2011). Estimated aerobic power, muscular strength and flexibility in elite Brazilian Jiu-Jitsu athletes. Science & Sports, 26(6), pp. 329-337.

Andreato, L. V., Lara, F. J. D., Andrade, A., & Branco, B. H. M. (2017). Physical and Physiological Profiles of Brazilian Jiu-Jitsu Athletes: a Systematic Review. [Review]. Sports Med Open, 3(1), p 9. doi:10.1186/s40798-016-0069-5 Retrieved from

Báez, E., Franchini, E., Ramírez-Campillo, R., Cañas-Jamett, R., Herrera, T., Burgos-Jara, C., & Henríquez-Olguín, C. (2014). Anthropometric characteristics of top-class Brazilian Jiu Jitsu athletes: Role of fighting style. International Journal of Morphology, 32(3), pp. 1043-1050.

Brown, L. E., & Weir, J. P. (2001). ASEP PROCEDURES RECOMMENDATION I: ACCURATE ASSESSMENT OF MUSCULAR STRENGTH AND POWER. Professionalization of Exercise Physiology, 4(11)

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. 2nd: Hillsdale, NJ: erlbaum.

Corrêa da Silva, B. V., Junior, M. M., de Moura Simim, M. A., Franchini, E., & da Mota, G. R. (2014). Performance in kimono grip strength tests among Brazilian Jiu-Jitsu practitioners from different levels. Journal of Combat Sports & Martial Arts, 5(1)

Coswig, V., Neves, A. S., & Del Vecchio, F. (2011). Características físicas e desempenho motor no jiu-jitsu brasileiro: estudo com iniciantes e experientes na modalidade. Lecturas Educación Física y Deportes (Buenos Aires)

da Silva, B. V., Simim, M. A., Marocolo, M., Franchini, E., & da Mota, G. R. (2015). Optimal load for the peak power and maximal strength of the upper body in Brazilian Jiu-Jitsu athletes. J Strength Cond Res, 29(6), pp. 1616-1621. doi:10.1519/JSC.0000000000000799 Retrieved from

da Silva, B. V. C., Ide, B. N., de Moura Simim, M. A., Marocolo, M., & da Mota, G. R. (2014). Neuromuscular responses to simulated brazilian jiu-jitsu fights. J Hum Kinet, 44, pp. 249-257. doi:10.2478/hukin-2014-0130 Retrieved from

da Silva, B. V. C., Júnior, M. M., Rogério, F. C., Dias, I. S., de Moura Simim, M. A., & da Mota, G. R. (2014). TESTES FÍSICOS DISCRIMINAM PRATICANTES DE BRAZILIAN JIU-JITSU? Revista Brasileira de Ciência e Movimento, 22(1), pp. 90-96.

da Silva, B. V. C., Marocolo Junior, M., de Moura Simim, M. A., Rezende, F. N., Franchini, E., & da Mota, G. R. (2013). Reliability in kimono grip strength tests and comparison between elite and non-elite Brazilian Jiu-Jitsu players. Archives of Budo, 8(2), pp. 103-107.

Del Vecchio, F. B., Gondim, D. F., & Arruda, A. C. (2016). Functional Movement Screening Performance of Brazilian Jiu-Jitsu Athletes From Brazil: Differences Considering Practice Time and Combat Style. J Strength Cond Res, 30(8), pp. 2341-2347. doi:10.1519/JSC.0000000000001324 Retrieved from

Diaz-Lara, F., Monteiro, L., Garcia Garcia, J. M., & Abián-Vicén, J. (2014). Body composition, isometric hand grip and explosive strength leg – similarities and differences between novices and experts in an international competition of Brazilian jiu jitsu. ARCH BUDO, 10, pp. 211-217. Retrieved from

Lima, P. O., Lima, A. A., Coelho, A. C., Lima, Y. L., Almeida, G. P., Bezerra, M. A., & de Oliveira, R. R. (2017). Biomechanical Differences in Brazilian Jiu-Jitsu Athletes: The Role of Combat Style. Int J Sports Phys Ther, 12(1), pp. 67-74. Retrieved from

Inbar, O., Bar-Or, O., & Skinner, J. S. (1996). The Wingate anaerobic test: John Wiley & Sons.
Kolmogorov, A. (1933). Sulla determinazione empirica di una lgge di distribuzione. Inst. Ital. Attuari, Giorn., 4, pp. 83-91.

Loreto, I., Nakamura, F. Y., Winckler, C., Bragança, J. R., da Fonseca, R. A., Moraes-Filho, J., . . . Kitamura, K. (2017). Strength-power performance of visually impaired Paralympic and Olympic judo athletes from the Brazilian National team: a comparative study. The Journal of Strength & Conditioning Research, 31(3), pp. 743-749.

Marinho, B. F., Andreato, L. V., Follmer, B., & Franchini, E. (2016). Comparison of body composition and physical fitness in elite and non-elite Brazilian jiu-jitsu athletes. Science & Sports, 31(3), pp. 129-134.

Rhea, M. R. (2004). Determining the magnitude of treatment effects in strength training research through the use of the effect size. J Strength Cond Res, 18(4), pp. 918-920. doi:10.1519/14403.1 Retrieved from
Sawilowsky, S. S. (2009). New effect size rules of thumb.

Smirnov, N. (1948). Table for estimating the goodness of fit of empirical distributions. The annals of mathematical statistics, 19(2), pp. 279-281.

Student. (1908). The probable error of a mean. Biometrika, pp. 1-25.

Wells, K. F., & Dillon, E. K. (1952). The sit and reach—a test of back and leg flexibility. Research Quarterly. American Association for Health, Physical Education and Recreation, 23(1), pp. 115-118.

Afranio “Fran” De Sousa Silva earned his Master’s Degrees in Physical Education from UNICAMP (Universidade Estadual de Campinas), and his Post-Graduate in Exercise Physiology & Functional Training from FMU (Faculdades Metropolitanas Unidas) in Brazil.

He earned his First Degree Blackbelt in Brazilian Jiu-Jitsu and Blackbelt in Judo, and recently won gold in the 2019 IBJJF Miami Spring International Open, also taking silver in open class

Comments Closed