Large-Scale Method for Identifying the Relationships between Racket Properties and Playing Characteristics

KARDITSAS, Helen Elizabeth (2020). Large-Scale Method for Identifying the Relationships between Racket Properties and Playing Characteristics. Doctoral, Sheffield Hallam University.

Karditsas_2020_PhD_Large-ScaleMethod.pdf - Accepted Version
Creative Commons Attribution Non-commercial No Derivatives.

Download (4MB) | Preview
Link to published version::


The application of advanced engineering in tennis has seen vast changes in playing styles, racket materials and racket design. Although previous researchers have investigated the effects of racket properties during and post ball-racket impacts, the studies focused on limited variation within racket properties. As regulators of the game, the International Tennis Federation monitor racket performance, however, standard laboratory test methods do not exist. The establishment of appropriate testing standards would further the understanding of the effect of racket properties, or racket property combinations, whilst reducing discrepancies between studies. This work aims to identify racket properties resulting in distinct behavioural characteristics through the development of a test protocol accurately simulating different forehand conditions found within the field of play. Classification of the raw player testing data, previously collected from the 2006 Wimbledon Qualifying tournament, identified the characteristics of three specific forehand shots used within the field of play. The forehand shots were identified as either topspin or slice, each possessing different defining characteristics. The results from the player shot classification, five impact positions varying along the longitudinal and transverse axis, and a restrictive torque value representative of hand grip were used in the development of a laboratory test protocol capable of realistically and accurately simulating different forehand shots. Using the developed test protocol for a typical topspin and slice forehand, a total of 39 rackets of varying properties and property combinations, were repeatedly impacted at the relative impact positions. A three-dimensional analysis, through the use of two Phantom High-Speed video cameras, recorded the experimental outputs within a fully calibrated control volume. Reducing the complexity of the data, the experimental outputs were interpreted using clustering techniques, identifying clusters of rackets sharing similar behavioural characteristics. A total of four clusters of distinct behavioural characteristics were identified for both the topspin and slice forehand. Analysis of these clusters revealed that rackets of diverse property combinations can produce similar behavioural characteristics, indicating the importance of varying racket property combinations in this area of research. The relationships between the behavioural clusters and subsequent racket properties were identified using multinomial logistic regression. (MNLR). Investigations revealed a complex dynamic relationship between racket properties and racket behaviour, such that racket behaviour, or performance, is dependent on its physical properties as both individual and interacting entities and are specific to shot type. Therefore, to gain a complete understanding regarding the effects of racket properties on the nature of the game, investigations consider the combined effects of racket properties and their relationship(s) to specific shot types found within the field of play.

Item Type: Thesis (Doctoral)
Thesis advisor - Goodwill, Simon [0000-0003-0638-911X]
Thesis advisor - Choppin, Simon [0000-0003-2111-7710]
Thesis advisor - Kelley, John [0000-0001-5000-1763]
Additional Information: Director of studies: Simon Goodwill / Thesis supervisors: Simon Choppin & John Kelley
Research Institute, Centre or Group - Does NOT include content added after October 2018: Sheffield Hallam Doctoral Theses
Identification Number:
Depositing User: Colin Knott
Date Deposited: 07 May 2021 14:42
Last Modified: 03 May 2023 02:07

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics