Northumbria Research Link

Citation: Vickery, Will, Dascombe, Ben and Duffield, Rob (2017) The Association Between
Internal and External Measures of Training Load in Batsmen and Medium-Fast Bowlers
During Net-Based Cricket Training. International Journal of Sports Physiology and
Performance, 12 (2). pp. 247-253. ISSN 1555-0273

Published by: Human Kinetics

URL: http://dx.doi.org/10.1123/ijspp.2015-0770 <http://dx.doi.org/10.1123/ijspp.2015-

0770>

This version was downloaded from Northumbria Research Link:

http://nrl.northumbria.ac.uk/id/eprint/27020/

Northumbria University has developed Northumbria Research Link (NRL) to enable users
to access the University’s research output. Copyright © and moral rights for items on
NRL are retained by the individual author(s) and/or other copyright owners. Single copies
of full items can be reproduced, displayed or performed, and given to third parties in any
format or medium for personal research or study, educational, or not-for-profit purposes
without prior permission or charge, provided the authors, title and full bibliographic
details are given, as well as a hyperlink and/or URL to the original metadata page. The
content must not be changed in any way. Full items must not be sold commercially in any
format or medium without formal permission of the copyright holder. The full policy is
available online: http://nrl.northumbria.ac.uk/policies.html

This document may differ from the final, published version of the research and has been
made available online in accordance with publisher policies. To read and/or cite from the
published version of the research, please visit the publisher’s website (a subscription
may be required.)
Title: The association between internal and external measures of training load in

batsmen and medium-fast bowlers during net-based cricket training.

Submission Type: Original investigation

Authors: Will Vickery1,2, Ben J. Dascombe2,3,4, and Rob Duffield5

Affiliations:
1
Department of Sport, Health and Rehabilitation, Faculty of Health and Life Sciences,

University of Northumbria: Newcastle Upon Tyne, Newcastle Upon Tyne, UK

2
Applied Sports Science and Exercise Testing Laboratory, School of Environmental

and Life Sciences, Faculty of Science and IT, University of Newcastle, Ourimbah,

NSW, Australia
3
Priority Research Centre in Physical Activity and Nutrition, University of

Newcastle, Callaghan, NSW, Australia

4
Department of Rehabilitation, Nutrition and Sport, School of Allied Health, LaTrobe

University, VIC, Australia

5
Sport and Exercise Discipline Group, UTS: Health, University of Technology

Sydney, NSW, Australia

Author for Correspondence:

Will Vickery

Department of Sport, Health and Rehabilitation,

Faculty of Health and Life Sciences,

University of Northumbria: Newcastle Upon Tyne,

Newcastle Upon Tyne, UK

NE1 8ST

Phone: 0191 243 7377

Email: [email protected]
Preferred Running Head: Training load during elite cricket training

Abstract Word Count: 243

Text-Only Word Count: 3411

Number of Figures: 2

Number of Tables: 2
 1 ABSTRACT

2 Purpose: To examine the relationship between session Rating of Perceived Exertion (sRPE)

3 and measures of internal and external training load (TL) within cricket batsmen and medium-

4 fast bowlers during net-based training sessions. Methods: The internal (heart rate), external

5 (movement demands, Player Load™) and technical (cricket-specific skills) loads of thirty,

6 male cricket players (age: 21.2 ± 3.8 y, height: 1.82 ± 0.07 m, body mass: 79.0 ± 8.7 kg) were

7 determined from net-based cricket training sessions (n = 118). The relationships between

8 sRPE and measures of TL were quantified using Pearson’s product moment correlations,

9 respective to playing position. Stepwise multiple regression techniques provided key internal

10 and external load determinants of sRPE in cricket players. Results: Significant correlations

11 were evident (r = -0.34 – 0.87, P < 0.05) between internal and external measures of TL and

12 sRPE, with the strongest correlations (r ≥ 0.62) existing for GPS-derived measures for both

13 playing positions. In batsmen, stepwise multiple regression analysis revealed that 67.8% of

14 the adjusted variance in sRPE could be explained by Player Load™ and high-intensity

15 distance (y = 27.43 + 0.81 Player Load™ + 0.29 high-intensity distance). For medium-fast

16 bowlers, 76.3% of the adjusted variance could be explained by total distance and mean heart

17 rate (y = 101.82 + total distance 0.05 + HRmean -0.48). Conclusion: These results suggest that

18 sRPE is a valid method of reporting TL amongst cricket batsmen and medium-fast bowlers.

19 Position specific responses are evident, and should be considered when monitoring the TL of

20 cricket players.

21

22 KEY WORDS: batsman, bowler, internal training load, external training load, GPS
23 INTRODUCTION

24 Within the confines of the high performance team sport environment, as a result of training

25 load monitoring practices, it is common to prescribe more individualised player training

26 programs specific to their respective match demands[1]. As reviewed previously [2, 3] there

27 are numerous methods currently available for monitoring an individual’s training load (TL),

28 though these are generally classified as either internal or external in nature[4]. Internal-TL,

29 particularly via Rating of Perceived Exertion (RPE) using Borg’s Category Ratio 10 [CR-10]

30 scale, is calculated by multiplying an individual’s RPE by the duration of a training session

31 (in minutes) [5]. Research has demonstrated the sRPE method to be a valid indicator of TL

32 when compared to other internal measures across an array of sports and activities [6-10]. In

33 addition to this, the advancements in micro-technology that allow global positioning system

34 (GPS) and accelerometer devices to measure external-TL mean that it is now ubiquitous in

35 many sports. Furthermore, recent studies show evidence of strong relationships between

36 measures of external-TL and sRPE, particularly within field-based team sports [4, 11, 12].

37

38 Training programs are traditionally prescribed on external-TL, as is determined by the work

39 performed by the athlete (ie. distance/speed from GPS devices), while the internal-TL

40 represents the psycho-physiological stress imposed on individual athletes [13]. As noted by

41 Impellizzeri and colleagues [14], the internal load experienced by an athlete is associated

42 with the extent of the external load placed upon them during training or match-play. Recent

43 evidence suggests a system that combines internal- and external-TL measures may be the

44 most appropriate method to holistically quantify TL [11, 15]. By comparison however, the

45 activity profile of cricket players during either training or match-play differs to that of other

46 field based team sports, as typically the durations are much longer and a larger proportion of

47 time is spent performing low-intensity activities (<3.5 m.s-1 and <75% maximum heart rate
48 [HRmax]) [16]. As such it is unclear whether measures of external-TL would be useful when

49 prescribing training sessions based on TL for cricket players, especially given such unique

50 and subtle movement characteristics of the sport and varying positions.

51

52 Given the high technical load specific to cricket and the various playing positions within a

53 single team, the use of external measures based on the technical demands of a specific sport

54 may be one way in which the TL’s of athletes could be monitored. Few studies have

55 examined the relationship between sRPE and technical measures of a specific sport, despite

56 the large number of studies which have compared the sRPE derived TL’s to common internal

57 (HR) and external (GPS) measures. Recently however, both Lovell et al. [11] and Weaving et

58 al. [15] have reported significant correlations between rugby league specific GPS-derived

59 technical measures, i.e. body load and number of impacts and sRPE during skills-specific (r

60 > 0.24) and skills-conditioning (r > 0.43) training sessions. Each respective study provided

61 evidence that the sport specific technical load measures in combination with other internal

62 and external measures of TL accounted for a predominance of the variance in sRPE.

63 Additionally, Murphy et al. [17] recently used shot count and the number of unforced errors

64 as a measure of reporting load amongst tennis players to determine player’s concepts of what

65 constitutes sRPE following a training session. Despite suggesting the use of external-TL

66 measures such as shot count may be useful when prescribing unsupervised practice; these

67 same measures were unable to explain the variance in sRPE within junior tennis players.

68

69 Collectively, the above research findings appear to suggest that the use of more sport specific

70 external measures of TL may be unique to each individual sport. In cricket, the training dose

71 of a net-based session is typically dictated by the restrictions of medium-fast bowlers, as

72 evidenced by a number of national organisations limiting the number of deliveries a medium-

73 fast bowler can perform in training [18, 19]. As such, coaching staff are more likely to

74 develop training programs based on this measure of TL load as opposed to internal HR-based

75 or external GPS-derived measures. Accordingly, the purpose of this study was to determine

76 the association between sRPE and previously established measures of internal- and external-

77 TL in cricket, and secondly, to determine what internal and external load markers are

78 determinants of position specific RPE responses in batsmen and medium-fast bowlers during

79 net-based training.

80

81 METHODS

82 Subjects

83 Thirty elite, male cricket players (age: 21.2 ± 3.8 y, height: 1.82 ± 0.07 m, body mass: 79.0 ±

84 8.7 kg; batsmen n =10; medium-fast bowlers n = 9) currently all playing at a minimum

85 standard of first-class cricket and with a minimum of 10 years playing experience

86 volunteered to participate in the study. All players provided verbal and written informed

87 consent prior to the commencement of the study. Players were familiarised with Borg’s CR-

88 10 RPE scale [20] and the exact procedures of the study prior to data collection. The Ethics

89 Committee of the University of Newcastle granted approval for the study (H-2010-1288).

90

91 Study Design

92 Whilst attending a pre-season training camp at the Australian National Cricket Centre the

93 internal- and external-TLs of batsmen and medium-fast bowlers were measured over a period

94 of 12 weeks during typical training sessions. During this time a total of 27 net-based training
 95 sessions were completed, with 118 individual sessions being used for analysis. A typical net-

96 based training session was similar to that previously reported in the studies of Vickery et al.

97 [21] and Petersen et al. [22], whereby batsmen batted against medium-fast bowlers (n = 2-3

98 bowlers per net) who rotated between deliveries as opposed to completing 6 ball overs on a

99 turf cricket pitch, which was surrounded by netting. Batsmen batted in pairs and were

100 instructed to rotate the strike by completing a single as typical of match-play, as often as

101 possible during their allotted batting period. When rotating the strike batsmen were

102 encouraged to perform this at typical match intensity. Training sessions were designed to

103 allow players to practice isolated technical aspects of cricket match-play [22]. Players were

104 instructed to train as per the instructions of their coaching staff.

105

106 Measures of Internal Training Load

107 Heart Rate

108 Heart rate (HR) was collected simultaneously from each player using heart rate monitors

109 (Polar Team2 System, Polar Electro Oy, Kemple, Finland) that sampled at 5 s intervals

110 throughout each training session. Due to limitations with the number of HR devices available

111 and restrictions made by coaching staff, the number of participants who wore HR devices

112 varied from 4-10 each session. Heart rate data was stored within the GPS device worn by the

113 player and download using Logan Plus 4.6 software (Catapult Innovations, Scoresby,

114 Australia) following each training session for analysis. As in previous studies [4, 12, 14]

115 mean HR (HRmean), HRmax and the amount of time spent above 75% HRmax were determined

116 during analysis [17]. Additionally, Edwards TRIMP method [23] for quantifying internal-TL

117 was determined as:

118 Internal-TL = (Zone 1 duration x 1) + (Zone 2 duration x 2) + (Zone 3 duration x 3) + (Zone

119 4 duration x 4) + (Zone 5 duration x 5)

120 where Zone 1 = 50-60% HRmax, Zone 2 = 60-70% HRmax, Zone 3 = 70-80% HRmax, Zone 4 =

121 80-90% HRmax and Zone 5 = 90-100% HRmax.

122

123 Each individual’s HRmax was determined from the HRmax achieved prior to exhaustion from

124 the performance of a Yo-Yo Intermittent Recovery Test Level 1 that was completed at the

125 commencement of the training camp.

126

127 Session-RPE

128 The perceived intensity of each specific training session was quantified using Borg’s CR-10

129 RPE scale [20] for each player following a training session as has been used previously [6, 8,

130 10, 24] . Player’s provided separate RPE scores for each of the separate training session

131 sections ie. Batting and bowling. Training load was then calculated by multiplying each

132 player’s RPE by the duration (min) of each specific part (i.e. batting and/or bowling) training

133 session [5]. To ensure that consistent ratings of perceived intensity were recorded, as

134 previously reported [24] sRPE scores were recorded 30 min following the conclusion of each

135 separate section of the training session (eg. 30 min following net batting, 30 min following

136 net bowling) to minimise any bias from the final stages of the session. Although as is typical

137 of net-based cricket training, each player was allowed to continue training if they completed

138 their batting and/or bowling session before other players which may have been led to some

139 limitations with regards to data analysis.

140
141 Measures of External Training Load

142 Global Positioning Systems

143 Similar to recent research [4, 12, 15] the movement patterns of each player during all training

144 sessions were recorded simultaneously via MinimaxX GPS devices (v6.65, Catapult

145 Innovations, Scoresby, Australia) sampling at a frequency of 10 Hz to determine the external-

146 TL of players. As with HR, limitations due to equipment availability and coaching

147 restrictions meant the number of players wearing a GPS device varied from 4-10 per session.

148 Each GPS unit was situated between the shoulder blades of each player using a specially

149 designed harness. Following each training session, data was downloaded to determine

150 measures of external-TL (distance covered [which included distance at a low-intensity:<3.5

151 m.s-1 and high-intensity: ≥3.5 m.s-1] [21], movement characteristics and Player Load™) using

152 Logan Plus 4.6 software (Catapult Innovations, Scoresby, Australia). To limit inter-unit

153 variability, each player was fitted with the same GPS device (where possible) during each

154 training session. To ensure spurious information was not included, data was removed when a

155 horizontal dilution of position value of greater than 5 was indicated, or when the number of

156 connected satellites was less than 5 [25].

157

158 Technical Skill

159 Recent evidence [17] suggests sports-specific technical skills are associated with athlete’s

160 perception of effort (r = 0.63), and hence may be an avenue for exploration in cricket

161 environments to determine TL. Consequently, during each net-based training session a fixed

162 video camera (HDV 1080i/mini DV Handycam, Sony, Japan) was placed behind the batsmen

163 (opposite end to where the ball was delivered by bowlers) to record the technical skills of
164 batsmen and bowlers during each net-based training session. Following data collection, the

165 footage was viewed by the lead researcher and notational analysis was used to quantify the

166 volume of the technical skills performed by batsmen (number of balls faced, number of balls

167 hit, number of defensive shots, number of attacking shots) and medium-fast bowlers (number

168 of balls bowled).

169

170 Statistical Analyses

171 Pearson’s product moment correlation was used to calculate the association between

172 measures of internal- and external-TL. Only those correlations that were statistically

173 significant (P<0.05) were reported. Similar to previous research [12] ratio measures for 90%

174 limits of agreement were also calculated using a customised spreadsheet [26]. Correlation

175 coefficients categories were to quantify the strength of the association based on Hopkins [27]

176 (trivial= 0-0.1, small= 0.1-0.3, moderate= 0.3-0.5, large= 0.5-0.7, very large= 0.7-0.9, almost

177 perfect= 0.9-1). Using both internal and external measures of TL, stepwise multiple

178 regression was used to determine a predictive equation for sRPE. Additionally, for each

179 playing position, partial correlations, standardized coefficients, and level of significance were

180 reported for sRPE. Collinearity tolerance statistics established correlations between predictor

181 variables, where values <.10 were considered beyond an acceptable tolerance level and

182 removed from the model. All statistical analyses were completed using SPSS (v. 22, IBM

183 Corporation, Somers, New York, USA) with the level of statistical significance set at P<0.05.

184 As in Scott et al. [4] the amount of data available from individual players was a limitation of

185 the study design and as such the following correlation coefficients reflect the relationship

186 between measures of TL from the pooled data, rather than the mean of intra-subject

187 correlations.
188

189 RESULTS

190 Measures of internal- and external-TL are presented in Table 1. Mean and 90% confidence

191 intervals of correlation coefficients between the measures of internal- and external-TL and

192 sRPE shown in Figure 1 for (a) batsmen and (b) medium-fast bowlers, respectively. Mean

193 duration of individual batting sessions was 21 ± 10 min (range: 13 - 61 min), whereas mean

194 duration for medium-fast bowling sessions was 28 ± 13 min (range: 13 – 72 min). Across

195 each of the playing positions, the mean sRPE TL for individual sessions were: batsmen 82 ±

196 39 Arbitrary Units (AU) (range: 43 – 179 AU) and medium-fast bowlers 124 ± 57 AU (range:

197 44 - 279 AU) (Table 1). Heart rate based internal measures of TL within batsmen showed a

198 small negative correlation with sRPE (r = -0.28 - -0.24, P < 0.05); whereas, moderate

199 correlations between external-TL and sRPE within batsmen were evident (r = -0.34 – 0.47, P

200 < 0.02) (Figure 1a). Specifically, measures of external-TL associated with physical demands

201 (distance covered, number of efforts, player load) demonstrated large to very large

202 correlations with sRPE (r = 0.60 – 0.74, P < 0.01) (Figure 1a). Alternatively, technical skill

203 (number of balls faced, hit and defensive shots played) displayed moderate negative

204 correlations with sRPE (r = -0.34 - -0.33, P < 0.02) (Figure 1a). In regards to medium-fast

205 bowlers, a small negative correlation existed between HRmean and sRPE (r = -0.29, P < 0.02),

206 with all other measures of internal-TL not significantly associated (P > 0.05) (Figure 1b).

207 Moderate to very large correlations (r = -0.54 – 0.87, P < 0.01) were evident between all

208 measures of external-TL and sRPE (Figure 1b). Notably a moderate negative correlation was

209 seen between work-to-rest ratio and sRPE (, r = -0.54, P < 0.01). Additionally, the number of

210 balls bowled demonstrated a strong association with sRPE (r = 0.68, P < 0.01,) (Figure 1b).

211 ***INSERT TABLE 1 AROUND HERE***

212 ***INSERT FIGURE 1AROUND HERE***

213 The results of the stepwise multiple regression analysis are presented in Table 2. A total of

214 67.8% of the adjusted variance in batsmen’s sRPE could be explained by Player Load™ and

215 the total distance covered performing at a high-intensity (y = 27.43 + 0.81 Player Load™ +

216 0.29 High-intensity distance; adjusted R2 = 0.68; F = 54.76; P < 0.001). The collinearity of

217 this equation was acceptable for both variables with tolerance levels of 0.530. In regards to

218 the medium-fast bowlers, total distance and HRmean accounted for 76.3% of adjusted variance

219 in sRPE (y = 101.82 + Total distance 0.05 + HRmean -0.48; adjusted R2 = 0.76; F = 100.97; P

220 < 0.001). Similar to batsmen, the collinearity of this equation was acceptable for both

221 variables with a tolerance level of 0.979.

222 ***INSERT TABLE 2 AROUND HERE***

223

224 DISCUSSION

225 This study aimed to determine the relationship between measures of internal- and external-TL

226 and sRPE amongst cricket batsmen and medium-fast bowlers. As in recent previous research

227 in tennis as well as field based team sports such as rugby and football [11, 14, 17, 24, 28], the

228 current research reported strong relationships between GPS-derived measures of load and

229 sRPE within both playing positions. However, HR-based measures of internal-TL typically

230 demonstrated weaker relationships with sRPE within both playing groups in the current

231 study. Also unique to this study was the use of cricket-specific skills as a measure of

232 external-TL, which indicated a moderate to strong relationship with sRPE dependent upon

233 playing position. The findings suggest that a collective of TL measures best explains sRPE
234 amongst cricket players, and it was interesting that the inclusion of cricket-specific skills was

235 not observed.

236

237 With the exception of HRmean (batsmen: r = -0.28; medium-fast bowlers: r = -0.29), the

238 present results showed no significant correlation between any measures of HR and sRPE

239 (HRmax, percentage time >75%HRmax and Edwards’ TRIMP) for either playing position. As

240 stated above, this contrasts with previous research that has reported strong relationships

241 between HR measures and sRPE in field based team sports [1, 12, 28]. For example,

242 Impellizeri and colleagues [14] found large to very large correlations (r = 0.50 – 0.85)

243 between sRPE and measures of HR amongst young soccer players when performing a soccer-

244 specific training program. More recently, Lovell et al. [11] reported moderate to large

245 correlations (r = 0.45 – 0.75) between sRPE as a measure of TL and Banister’s TRIMP

246 across a range of rugby league training activities amongst professional players. The weak

247 correlation between sRPE and HR-based measures of load in the present study may be

248 explained by the unique nature of cricket training, especially given such large proportions of

249 time are spent performing low-intensity activities, particularly compared to the more likely

250 higher-intensity sessions undertaken by the football codes [21, 29]. Previous research within

251 cricket [21, 30, 31] highlights that despite the intermittent nature of net-based training, a large

252 percentage of time is spent at an intensity below 75%HRmax during net-based cricket training

253 sessions (batsmen: 43 ± 38%; medium-fast bowlers: 48 ± 37%). Given Edwards’ TRIMP

254 method places a greater weighting on more intense activity when calculating internal-TL, this

255 may explain the lower correlation when compared to other team sports requiring longer

256 periods at higher intensities [32, 33]. Surprisingly though, the current results demonstrated a

257 negative correlation for HRmean and sRPE for both playing positions. As such, alternative

258 methods other than those based on internal measures (HR, sRPE) may be required to monitor
259 the TL of cricket batsmen and medium-fast bowlers. As sRPE includes the duration of the

260 training session it is possible that the considerable portion of time spent at low-intensities

261 invokes low cardiovascular load, yet can amplify the calculated TL. This suggests that

262 coaches who develop net-based training sessions which are designed based around internal

263 measures of TL may need to consider this information regarding the cardiovascular responses

264 of cricket players. The limited relationship which exists between internal-TL and sRPE

265 amongst cricket batsmen and medium-fast bowlers suggests that external measures may

266 possibly have a stronger relationship with sRPE.

267

268 Similar to recent studies [4, 11, 12] strong correlations were present between sRPE and

269 measures of external-TL. Specifically for batsmen, large correlations existed between sRPE

270 and total distance (r = 0.74), total low-intensity distance (r = 0.74) and Player Load™ (r =

271 0.73). Further, moderate correlations were observed between sRPE with total high-intensity

272 distance (r = 0.62) and the number of high-intensity efforts (r = 0.60). As in previous studies,

273 weaker correlations were reported with an increase in running speed [4, 11], however as in

274 the current study, these correlations were still considered moderate to large (r ≥ 0.43). Due to

275 the limited movement that occurs when batting in the nets (as highlighted by the proportion

276 of low-intensity activity in Table 2), it is not surprising that the strongest relationship existed

277 between movement performed at low speeds and sRPE within batsmen. A similar result

278 occurred with medium-fast bowlers, with large correlations existing between sRPE and all

279 GPS-derived external measures of load (r = 0.76 – 0.87) apart from work-to-rest ratio

280 (r = -0.58). Although still largely low-intensity activity, the increased proportion of high-

281 intensity activity performed by medium-fast bowlers during net-based training explains the

282 greater correlation to sRPE than when compared to batsmen. In regards to work-to-rest ratio,

283 a high ratio (more time between high- and low-intensity efforts) is likely to result in a lower
284 perceptual response due to the increased recovery time, which likely explains the negative

285 correlation to sRPE. Based on this, coaches may consider decreasing the work-to-rest ratio if

286 wanting to increase the resulting TL of medium-fast bowlers during a net session. Regardless

287 of playing position, GPS-derived external measures appear to correlate to sRPE during

288 cricket training, highlighting that measures of external-TL may prove more useful for cricket

289 coaches when monitoring the loads of batsmen and medium-fast bowlers when compared to

290 internal measures. Unlike that of previous research [12, 15], this study found stronger

291 relationships between external measures of TL and sRPE as opposed to internal-TL

292 measures.

293

294 A new finding from this study was the relationships observed between cricket-specific skills

295 and sRPE. Specifically, the number of balls faced and hit by batsmen during a net session

296 demonstrated a moderate but negative correlation (r = -0.34 and -0.33, respectively) with

297 sRPE. Houghton et al. [31] reported a general increase in batsmen’s RPE with an increase in

298 the number of balls faced during a simulated batting innings. However, this finding included

299 the physical work that accompanied each shot during the simulation and therefore not

300 unsurprisingly, an increase in physical work was evident alongside in the increase in

301 perceived intensity. In the current study however, it was unclear as to why sRPE and the

302 number of balls faced by batsmen shared a negative association. This conceptually differs to

303 the study of Lovell et al. [11], where a positive relationship was reported between a skill-

304 specific external measure of load (impacts) and sRPE. Within the current study it is possible

305 this negative relationship may reflect that the more balls faced during net-training results in

306 longer sessions and less movement, hence the lower RPE is a by-product of more time in the

307 nets and reduced total and high-intensity movements [34].In regards to medium-fast bowlers,

308 a large correlation (r = 0.68) was reported between the number of balls bowled and sRPE.
309 This is not surprising as completing a greater number of deliveries will lead to greater high-

310 intensity running for the run up of each delivery as well as lengthen the duration of the

311 training session. Consequently, the inverse relationship between technical activity and sRPE

312 may suggest alternate, if not expanded methods of TL monitoring are required. Therefore,

313 although it is common practice for net sessions to be based around the technical load of

314 medium-fast bowlers [18, 19], coaches need to consider the playing position when

315 developing training sessions which are based around the volume of technical skills

316 performed.

317

318 Similar to recent studies by Lovell et al. [11] and Murphy et al. [17], the use of a multiple

319 stepwise regression in the current study shows a combination of load and intensity measures

320 may explain more of the variance in sRPE than individual measures of load. Unlike these

321 previous studies, the results of the multiple stepwise regression analysis differed depending

322 on the playing position. For batsmen, Player Load™ and distance covered at a high-intensity

323 contributed to 67.8% of the adjusted variance for sRPE. Although unexpected, the inclusion

324 of distance covered at a high-intensity and Player Load™ to explain the variance in batsmen

325 sRPE suggests that the movement characteristics (e.g. running between the wickets and small

326 movements in various directions whilst batting as opposed to remaining stationary during net-

327 sessions) are influential in the perceived intensity of batsmen.

328

329 Meanwhile for the medium-fast bowlers, 76.3% of the variance in sRPE could be explained

330 by total distance and HRmean. These results suggest that within batsmen only external

331 measures of TL account for the variance within sRPE, whereas within medium-fast bowlers it

332 is a combination of internal- and external-TL measures. Interestingly within both playing
333 positions, the external measures of TL specific to technical skill did not account for any

334 variance using this analysis. Therefore, these results would suggest that a combination of

335 internal- (HR-based) and external-TL (GPS-derived and skill-specific) measures account for

336 sRPE within cricket players during net-based training sessions, although this is somewhat

337 position specific. It should be noted that these results are specific to batsmen and medium-

338 fast bowlers during net-based cricket training. Although other methods of training are

339 currently used for skill development and physical conditioning in the sport of cricket such as

340 small-sided games or conditioning based exercises, this study was limited to net-based

341 training sessions due to time and player access restrictions. Future research should consider

342 the training loads of each playing position associated with a variety of training methods

343 which are utilised by current cricket coaches

344

345 PRACTICAL APPLICATIONS

346  The use of sRPE appears to be a suitable tool for monitoring the TL of cricket

347 batsmen and medium-fast bowlers with the addition of other internal and external

348 measures for monitoring TL.

349  Coaches may need to reconsider only using of cricket-specific measures of skill

350 volume, such as medium-fast bowler’s ball count, in monitoring the TL of cricket

351 players during net-based training sessions. A combination of both internal, GPS

352 derived-external and cricket-specific measures of skill may be more superior to

353 monitor TL.

354  The data suggests the GPS-derived (external) information proves the most useful and

355 suitable for coaches for the determining of position specific TL during net-based
356 training sessions. This would minimise the amount of information required and

357 therefore impeding less on player’s practice time.

358  As sRPE can be explained by varying internal and external measures of TL, coaches

359 need to consider playing position when deciding upon which TL measure to use when

360 developing net-based training sessions and monitoring cricket players. This would

361 allow for more specific information to be gathered by coaches which in turn would

362 help in the development of more individualised net-based training programs.

363

364 CONCLUSION

365 This study supports the use of sRPE as a measure of TL as it was demonstrated that sRPE is

366 highly correlated with external-TL measures, particularly those derived from GPS devices in

367 cricket batsmen and fast-bowlers. However, this was not the case with HR-derived internal-

368 TL measures, which is likely explained by the intermittent nature and greater proportion of

369 low-intensity activity of cricket players during training activities. It was also evident that

370 technical skill external measures of TL were correlated to sRPE to varying levels depending

371 on playing position. Additionally, this study also showed that a number of factors could be

372 used to predict sRPE as opposed to only relying on one internal or external measure of TL,

373 although these factors differ between playing position. Overall the results of this study

374 provide cricket coaches with information regarding the use of load monitoring during net-

375 based cricket training sessions.

376

377 ACKNOWLEDGEMENTS
378 The authors would like to express their gratitude to the Cricket Australia Sport Science Sport

379 Medicine Group particularly Rian Crowther, David Beakley and Peter Blanch as well as the

380 coaching and playing groups at the National Cricket Centre, Australia.

381
382 REFERENCES

383 1. Borresen, J. and M.I. Lambert, Quantifying training load: a comparison of subjective

384 and objective methods. Int J Sports Physiol Perform. 2008; 3: 16-30.

385 2. Borresen, J. and M.I. Lambert, The quantification of training load, the training

386 response and the effect on performance. Sports Med. 2009; 39: 779-795.

387 3. Lambert, M.I. and J. Borresen, Measuring training load in sports. Int J Sports Physiol

388 Perform. 2010; 5: 406-411.

389 4. Scott, B.R., et al., A comparison of methods to quantify the in-season training load of

390 professional soccer players. Int J Sports Physiol Perform. 2013; 8: 195-202.

391 5. Foster, C., et al., Effects of specific versus cross-training on running performance. Eur

392 J Appl Physiol Occup Physio. 1995; 70: 367-372.

393 6. Coutts, A., et al., Validity of the session-RPE method for determining training load in

394 team sport athletes. J Sci Med Sport. 2003; 6: 525.

395 7. Day, M.L., et al., Monitoring exercise intensity during resistance training using the

396 session RPE scale. J Strength Cond Res. 2004; 18: 353-358.

397 8. Herman, L., et al., Validity and reliability of the session RPE method for monitoring

398 exercise training intensity. South Afr J Sports Med. 2009; 18: 14-17.

399 9. Manzi, V., et al., Profile of weekly training load in elite male professional basketball

400 players. J Strength Cond Res. 2010; 24: 1399-1406.

401 10. Wallace, L., et al., Using session-RPE to monitor training load in swimmers. Strength

402 Cond J. 2008; 30: 72-76.

403 11. Lovell, T.W.J., et al., Factors affecting perception of effort (session rating of

404 perceived exertion) during rugby league training. Int J Sports Physiol Perform. 2013;

405 8: 62-69.
406 12. Scott, T.J., et al., Validity and reliability of the session-RPE method for quantifying

407 training in Australian Football: a comparison of the CR10 and CR100 scales. J

408 Strength Cond Res. 2013; 27: 270-276.

409 13. Impellizzeri, F.M., E. Rampinini, and S.M. Marcora, Physiological assessment of

410 aerobic training in soccer. J Sports Sci. 2005; 23: 583-592.

411 14. Impellizzeri, F.M., et al., Use of RPE-based training load in soccer. Med Sci Sports

412 Exerc. 2004; 36: 1042-1047.

413 15. Weaving, D., et al., Combining internal- and external-training-load measures in

414 professional rugby league. Int J Sports Physiol Perform. 2014; 9: 905-912.

415 16. Abernethy, B., Mechanisms of skill in cricket batting. Aus J Sports Med. 1981; 13: 3-

416 10.

417 17. Murphy, A.P., et al., Comparison of athlete-coach perceptions of internal and external

418 load markers for elite junior tennis training. Int J Sports Physiol Perform. 2014; 9:

419 751-756.

420 18. Cricket Australia, Junior Cricket Policy. 2004, Cricket Australia: Melbourne,

421 Australia.

422 19. English Cricket Board, ECB Fast Bowling Directives. 2013, English Cricket Board:

423 London, UK.

424 20. Borg, G., P. Hassmen, and M. Lagerström, Perceived exertion related to heart rate and

425 blood lactate during arm and leg exercise. Eur J Appl Physiol Occup Physiol. 1987;

426 56: 679-685.

427 21. Vickery, W., B. Dascombe, and R. Duffield, Physiological, movement and technical

428 demands of centre-wicket Battlezone, traditional net-based training and one-day

429 cricket matches: a comparative study of sub-elite cricket players. J Sports Sci. 2014;

430 32: 722-737.

431 22. Petersen, C.J., et al., Comparison of training and game demands of national level

432 cricketers. J Strength Cond Res. 2011; 25: 1306-1311.

433 23. Edwards, S., High performance training and racing, in The Heart Rate Monitor, S.

434 Edwards, Editor. 1993, Sacramento, CA: Feet Fleet Press. 113-123.

435 24. Foster, C., et al., A new approach to monitoring exercise training. J Strength Cond

436 Res. 2001; 15: 109-115.

437 25. Elsworthy, N. and B.J. Dascombe, The match demands of Australian rules football

438 umpires in a state-based competition. Int J Sports Physiol Perform. 2011; 6: 559-571.

439 26. Hopkins, W., et al., Progressive statistics for studies in sports medicine and exercise

440 science. Med Sci Sports Exerc. 2009; 41: 3.

441 27. Hopkins, W.G., A scale of magnitudes for effect statistics. A scale of magnitudes for

442 effect statistics. 2002.

443 28. Alexiou, H. and A. Coutts, A comparison of methods used for quantifying internal

444 training load in women soccer players (2008). Int J Sports Physiol Perform. 2008; 3:

445 320-330.

446 29. Petersen, C., et al., Movement patterns in cricket vary by both position and game

447 format. J Sports Sci. 2010; 28: 1-8.

448 30. Duffield, R., M. Carney, and S. Karppinen, Physiological responses and bowling

449 performance during repeated spells of medium-fast bowling. J Sports Sci. 2009; 27:

450 27-35.

451 31. Houghton, L., et al., Movement patterns and physical strain during a novel, simulated

452 cricket batting innings (BATEX). J Sports Sci. 2011; 29: 801-809.

453 32. Bangsbo, J., M. Mohr, and P. Krustrup, Physical and metabolic demands of training

454 and match-play in the elite football player. Physical and metabolic demands of

455 training and match-play in the elite football player. 2006; 24: 665-674.
456 33. Impellizzeri, F.M., et al., Physiological and performance effects of generic versus

457 specific aerobic training in soccer players. Int J Sports Med. 2006; 27: 483-492.

458 34. Green, J.M., et al., Effect of exercise duration on session RPE at an individualized

459 constant workload. Eur J Appl Physiol. 2009; 107: 501-507.

460

461
Table 1. Internal and external measures of training load (mean ± SD) of batsmen and
medium-fast bowlers during net-based training sessions.

Batsmen Medium-Fast Bowler

(n = 24) (n = 18)
Duration (min) 21 ± 10 28 ± 13
HRmean (b.min-1) 145 ± 18 149 ± 16
HRmax (b.min-1) 171 ± 21 186 ± 19
%Time >75% HRmax 46 ± 37 46 ± 31
Edward’s TRIMP (AU) 60 ± 48 109 ± 139
Total Distance (m) 421 ± 420 2181 ± 1066
Relative Distance (m.min-1) 16 ± 11 76 ± 16
Distance at LI (m) 404 ± 394 1508 ± 730
Distance at HI (m) 14 ± 35 657 ± 356
# HIE 9 ± 15 88 ± 37
Peak Speed (m.s-1) 2.6 ± 1.8 6.2 ± 1.7
Work:Rest Ratio (1:x) 189 ± 472 7±8
PlayerLoad ™ 68 ± 35 150 ± 72
# Balls Faced 68 ± 17
# Balls Hit 56 ± 15
# Defensive Shots 18 ± 6
# Attacking Shots 39 ± 14
# Balls Bowled 30 ± 10

HR = heart rate; LI = low-intensity; HI = high-intensity; # = number of

Table 2. Partial correlations, standardised coefficients (ß) and level of significance (P) for
predictors of sRPE within batsmen and medium-fast bowlers.

Partial Correlation ß P

Batsmen sRPE
Player Load™ 0.649 0.651 0.000
Total Distance
HI Total Distance 0.296 0.236 0.035
LI Total Distance
# HIE
HRmean
# Balls Faced
# Balls Hit
Medium-Fast Bowlers sRPE
Player Load™
Total Distance 0.868 0.866 0.000
HI Total Distance
LI Total Distance
# HIE
Work:Rest Ratio
HRmean -0.248 -0.143 0.025
# Balls Bowled

HR = heart rate; LI = low-intensity; HI = high-intensity; # = number of