Identifying the performance characteristics explanatory of fight outcome in elite Pencak Silat matches

ABSTRACT

This study determined the performance characteristics associated with successful exponents in Pencak Silat combat, as well as examined the relationship between performance indicators and match outcome (2015 South-East Asian Games and the 2016 Pencak Silat World Championship) using both linear (binary logistic regression) and non-linear (conditional interference (CI) classification tree) statistical techniques. A magnitude-based approach was used to compare for differences in movement characteristics between semi-finalist (SF) and non-semi-finalist (NSF). Results showed that SF preferentially utilised the single front kick (effect sizes; ES = 0.29; likely) in offence. Also, SF performed more defensive side (ES = 0.27; possibly) and front kick (ES = 0.41; very likely) during defensive manoeuvres. During counter-attacks, SF executed more roundhouse kick (ES = 0.31; likely), scissor takedown (ES = 0.33; likely) and throws (ES = 0.40; very likely). The binary logistic regression and CI classification tree analyses identified “total takedown success”, “defending takedown success” and “catch success” as the most influential in explaining match outcome, with both models achieving 75% classification accuracy. The present findings highlight key fighting characteristics and performance indicators associated with match success. The incorporation of the non-linear CI classification tree may provide coaches with a guided and practical tool to aid with decision-makings and devising match strategies.

KEYWORDS:

• Combat sport
• machine learning
• performance analysis
• classification tree
• non-linear analysis

Disclosure statement

No potential conflict of interest was reported by the authors.