See discussions, stats, and author profiles for this publication at: [Link]

net/publication/357785300

Long-Term Adaptations in the Squat, Bench Press and Deadlift: Assessing

Strength Gain in Powerlifting Athletes

Article in Medicine and Science in Sports and Exercise · January 2022

DOI: 10.1249/MSS.0000000000002858

CITATIONS READS

7 1,162

6 authors, including:

Christopher Latella Patrick J Owen

Edith Cowan University Eastern Health Australia
91 PUBLICATIONS 992 CITATIONS 162 PUBLICATIONS 2,145 CITATIONS

SEE PROFILE SEE PROFILE

Timothy B. Davies Jemima Spathis

The University of Sydney Australian Catholic University
28 PUBLICATIONS 1,158 CITATIONS 26 PUBLICATIONS 815 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Patrick J Owen on 27 January 2022.

The user has requested enhancement of the downloaded file.

 1 Long-term adaptations in the squat, bench press and deadlift: Assessing

2 strength gain in powerlifting athletes

3 Short title: Upper- and lower-body strength adaptation

4 Medicine and Science in Sports and Exercise

5 ORIGINAL INVESTIGATION

7 Christopher Latella1,2, Patrick J. Owen3, Timothy Davies4, Jemima Spathis5, Alistair

8 Mallard6, Daniel van den Hoek5

1
9 Centre for Exercise and Sports Science Research (CESSR), School of Medical and Health

10 Sciences, Edith Cowan University, Joondalup, Western Australia, Australia.

2
11 Neurophysiology Research Laboratory, Edith Cowan University, Joondalup, Western

12 Australia, Australia.

3
13 Institute for Physical Activity and Nutrition (IPAN), School of Exercise and Nutrition

14 Sciences, Deakin University, Victoria, Australia.

4
15 Discipline of Exercise and Sport Science, Faculty of Medicine and Health, University of

16 Sydney, Camperdown, New South Wales, Australia.

5
17 School of Behavioural and Health Sciences, Australian Catholic University, Banyo,

18 Queensland, Australia.

6
19 Australasian Kidney Trials Network, Centre for Health Services Research, Faculty of

20 Medicine, University of Queensland, Brisbane, Queensland, Australia.

1
21 Corresponding Author:

22 Dr Christopher Latella

23 Centre for Exercise and Sports Science Research (CESSR)

24 School of Health and Medical Sciences

25 Edith Cowan University

26 270 Joondalup Drive, Joondalup,

27 Western Australia 6027

28 AUSTRALIA

29 Email: [Link]@[Link]

30 Phone: +61 (08) 6034 3637

31

32

2
33 ABSTRACT

34 Purpose: Understanding strength changes with resistance training is important in human

35 performance. It also enables better understanding into the expected magnitude of strength

36 increase and factors that influence this change over time. Methods: Squat (SQ), bench press

37 (BP) and deadlift (DL) scores were collated from 407 powerlifting (PL) meets (n=1896 unique

38 competitors: ~625 females, ~1270 males) between 2003-2018. Absolute (kg) and relative

39 starting strength (kg/body weight) for each lift type was expressed for both sexes. Maximum

40 and overall strength gain/day and /year (kg) was calculated by comparing first and final, or

41 maximum scores for each lift, respectively, and considered based on strength quartile

42 classification. Paired and independent T-tests compared strength changes from baseline and

43 between sexes. One-way ANOVAs compared strength changes between quartiles. Pearson

44 correlations assessed relationships between strength changes over time, and baseline strength,

45 number of competitions and total days competing. Results: Maximum strength adaptations

46 were greater for SQ (20.2-25.4 kg/year) and DL (18.1-21.1 kg/year) compared to BP (10.5-

47 12.8 kg/year, P≤0.001). However, the change in absolute (all lifts: P=0.247-0.379), and relative

48 strength (all lifts: P=0.641-0.821) did not differ between sexes. For females, maximum strength

49 gain/day did not differ by quartile (all lifts: P=0.091-0.746), nor did overall strength gain/day

50 (P=0.151-0.575). Conversely, males in the fourth quartile generally displayed lower maximum

51 and overall strength gain/day. Conclusion: These findings show differences in strength gain

52 between upper- and lower-body lifts, but not sex differences in the change in strength. In line

53 with previous research, the strongest males likely gain strength more slowly than weaker

54 counterparts. Professionals should consider this information in the training, assessment and

55 long-term benchmarking of athletes whose sports require a focus on muscular strength.

56 KEYWORDS

3
57 Performance analysis; sport; resistance-training; neuromuscular; muscle

58

59 INTRODUCTION

60

61 The importance of muscular strength for overall health (1,2) and human performance (3) is

62 well documented. Indeed, strength adaptations to resistance training have predominantly been

63 investigated in short-to-medium term studies, with less evidence examining the long-term (i.e.

64 >1 year or longer) effects (see 4-6, for general population example see 7) due to, for example,

65 logistical and resource limitations. However, strength athletes already training and competing

66 in respective sports (e.g. powerlifting [PL] and weightlifting) can provide a feasible and an

67 ecologically valid model to assess longer-term adaptations to resistance training aimed at

68 improving maximum strength or power (8-12). Moreover, PL offers the unique potential to

69 examine ‘real-world’ adaptations in individuals whose primary focus is on upper- and lower-

70 body maximum strength devlopment (8,13).

71

72 In cross-sectional studies, the relationship between physical characteristics (e.g. body

73 composition and anthropometry measures) have been associated with overall strength in PL

74 athletes (14), and more specifically, with squat (SQ), bench press (BP) and deadlift (DL)

75 performance (15). Of the limited intervention-based research, Colquhoun et al. (16) reported

76 that strength in the SQ, BP and DL improve by ~11.8-12.2%, ~6.8-7.5% and ~7.8-8.9%,

77 respectively, over a 9-week flexible or traditional undulating training period in strength-trained

78 males. In a subsequent study, also in strength trained males, Colquhoun et al. (17) reported a

79 similar absolute (kg) and percentage increase in strength for the SQ (16.6-16.7 kg, 12.0-12.2%),

80 BP (7.8-9.7 kg, 7.7%-8.6%) and DL (19.0-21.0 kg, 11.8-12.6%) when training either three or

4
 81 four days per week for six weeks. However, potential differences in the magnitude of strength

82 adaptation between each type of lift were not explored. Nevertheless, the initial results (e.g.

83 ~4-5% greater increase in strength for the SQ and DL compared to the BP, with calculated

84 effect sizes derived from the results ranging from Cohen’s d=0.69 [medium; i.e. 75.5% of

85 values are greater than BP strength gains] to d=1.24 [i.e. large; 89.3% of values are greater than

86 BP strength gains]) suggest that disparate adaptations may occur between strength tasks,

87 especially if evaluated over longer durations. Furthermore, as both cohorts consisted of males

88 only, it is unclear if a similar trend would be observed in females. Indeed, a recent review has

89 suggested that any similarity in adaptive trends between sexes, at least in general resistance

90 training practice, may be confined to the lower limbs only (18).

91

92 To our knowledge, only two studies have attempted to examine long-term adaptations in

93 muscular strength directly in PL athletes; our previous work (8) and Solberg et al. (9). The

94 results of our previous analysis (8) demonstrate that, for the most part, the whole body strength

95 gain (per day and per year) is similar across all PL athletes regardless of sex and baseline

96 starting strength. One exception to this was the strongest males (i.e. top 25%), who had a slower

97 increase in strength compared to males in the bottom 25% possibly due to a ceiling effect. As

98 far as we are aware, only the study by Solberg et al. (9) has attempted to document specific

99 long-term strength adaptations in each of the SQ, BP and DL in PL athletes. The authors report

100 that in the five years preceding peak performance age (mean age range across all categories =

101 30.7-40.6 years of age), strength improved by 12%, with improvements in the SQ and BP

102 reported to be greater than the DL. Moreover, the authors also suggest that females may

103 possibly gain more strength over this period than males, which does not support our previous

104 findings (8). Further, the magnitude of strength increase for each lift (i.e. SQ and BP > than

105 DL), appears to be dissimilar to that reported by Colquhoun et al. (17) in trained males where

5
106 strength increases were lowest for the BP. Thus, further investigation into long-term muscular

107 strength adaptations between sexes and lift types in PL athletes is warranted.

108

109 Therefore, the aim of this study was to investigate longitudinal upper- and lower-body strength

110 adaptations. Specifically, we aimed to retrospectively analyse the magnitude and strength gain

111 (per day and year) for the SQ, BP and DL in both male and female PL athletes over a 15-year

112 period. Based on the current literature and our previous analysis (8) examining whole body

113 strength changes in a similar cohort, we hypothesized that strength adaptations for each lift

114 type would occur similarly across PL competitors and between sexes. However, athletes in the

115 uppermost strength quartile were expected to gain strength in each lift at a slower rate. Based

116 on previous research, we also hypothesized that rates of strength adaptation would be greater

117 in the lower-body dominant lifts (SQ and DL).

118

119 METHODS

120 Experimental approach to the problem

121 We collated available Powerlifting Australia PL competition records from 407 meets (7084

122 competition entries) consisting of local, national and international “classic” competitions

123 between 2003-2018. Competition records were extracted from the publicly available

124 OpenPowerlifting Project database: [Link]. In order to be transparent, we

125 note that the data set used for this analysis was similar to our previous study (8), but updated

126 where required, and the aims and analysis differ. Approval for this project was granted by the

127 Edith Cowan University Human Research Ethics Committee (project no. 21408).

128

6
129 Subjects

130 Collectively, there were 1896 unique competitors for DL, 1895 for BP and 1894 for SQ (~625

131 females, ~1270 males). Mean (SD) age of all unique competitors at their first recorded

132 competition was 28 (9) years (range: 15-74 years). On average, female competitors were four

133 years older (P<0.001) than male competitors. Mean (SD) body weight of all unique competitors

134 at first competition was 82.3 (19.2) kg (range: 39.3-187.5). Females (mean [SD]: 69.5 [15.7]

135 kg, range: 39.3-155.9 kg) had lower (P<0.001) bodyweight than males (88.6 [17.6] kg, range:

136 53.0-187.5 kg).

137

138 Procedures

139 Strength data were extracted for all competitors from each competition by taking the highest

140 successful weight lifted out of three attempts for the SQ, BP and DL, respectively, and used

141 for further analysis. Each lift type was categorised separately. At least one successful lift out

142 of the three attempts for each of the SQ, BP and DL were required in order to record a total

143 score during a competition. Database errors were removed by manual screening and outliers

144 were determined as per methods detailed in our previous study (8). Following database

145 cleaning, individuals who competed (i) only once, (ii) in BP or DL only competitions, (iii) in

146 equipped competitions (i.e. those that permit the use of knee wraps, SQ and DL suits, or BP

147 shirts), or (iv) those who failed to record at least two scores for any of the SQ, BP or DL from

148 separate competitions were excluded. The following exclusions occurred: three participants for

149 SQ (male: n=2), two participants for BP (male: n=1) and one participant for DL (male: n=1) as

150 scores were only recorded from either lift at one competition only. Absolute strength

151 (expressed in kilograms of weight lifted) and starting relative strength (expressed as first

152 competition score [kg]/first competition bodyweight [bw]) for each lift type (i.e. SQ, BP and

7
153 DL) was expressed for both males and females regardless of age. Competitors were then

154 stratified into starting strength quartiles for each sex and each lift independently. For example,

155 quartile one (Q1) indicated competitors with relative strength in the bottom 25% at baseline,

156 while quartile four (Q4) indicated competitors with the highest relative strength at baseline, as

157 similarly done in our previous study examining whole body strength adaptations in PL athletes

158 (8). Additionally, the maximum and overall strength gain/day (kg) for each lift type was

159 calculated using the following formulae:

160 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ 𝑔𝑎𝑖𝑛/𝑑𝑎𝑦 (𝑘𝑔)

[𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑠𝑢𝑐𝑐𝑒𝑠𝑠𝑓𝑢𝑙 𝑎𝑡𝑡𝑒𝑚𝑝𝑡] − [ℎ𝑖𝑔ℎ𝑒𝑠𝑡 𝑠𝑢𝑐𝑐𝑒𝑠𝑠𝑓𝑢𝑙 𝑎𝑡𝑡𝑒𝑚𝑝𝑡 𝑎𝑡 𝑓𝑖𝑟𝑠𝑡 𝑐𝑜𝑚𝑝𝑒𝑡𝑖𝑡𝑖𝑜𝑛]

161 =
[𝑑𝑎𝑦𝑠 𝑖𝑛 𝑏𝑒𝑡𝑤𝑒𝑒𝑛]

162 𝑇𝑟𝑢𝑒 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ 𝑔𝑎𝑖𝑛/𝑑𝑎𝑦 (𝑘𝑔)

[𝑓𝑖𝑛𝑎𝑙 𝑠𝑢𝑐𝑐𝑒𝑠𝑠𝑓𝑢𝑙 𝑎𝑡𝑡𝑒𝑚𝑝𝑡] − [ℎ𝑖𝑔ℎ𝑒𝑠𝑡 𝑠𝑢𝑐𝑐𝑒𝑠𝑠𝑓𝑢𝑙 𝑎𝑡𝑡𝑒𝑚𝑝𝑡 𝑎𝑡 𝑓𝑖𝑟𝑠𝑡 𝑐𝑜𝑚𝑝𝑒𝑡𝑖𝑡𝑖𝑜𝑛]

163 =
[𝑑𝑎𝑦𝑠 𝑖𝑛 𝑏𝑒𝑡𝑤𝑒𝑒𝑛]

164

165 Where: (i) ‘maximum successful attempt’ is the maximum successful weight in kilograms

166 recorded during any competition for each competitor for each lift type, (ii) ‘highest successful

167 attempt at first competition’ is the best recorded score for each lift type in kilograms for each

168 competitor in their first competition, (iii) ‘final successful attempt’ is the best recorded score

169 for each lift type in kilograms for each competitor in their last documented competition within

170 the dataset, and (iv) ‘days in between’ is the number of days in between the first competition

171 and either the maximum or final successful attempt, for maximum and overall strength gain,

172 respectively. This was done to ensure the ‘days in between’ value was appropriate for each

173 metric. Maximum and overall strength gain per year was calculated by multiplying the value

174 derived from each formula by 365.

175

8
176 Statistical analyses

177 All analyses were conducted using Stata statistical software version 16 (College Station, TX).

178 Independent t-tests were used to compare outcomes between sexes, whereas paired t-tests were

179 used to compare maximum and final strength with baseline strength. One-way analysis of

180 variances (ANOVAs) with Tukey post-hoc adjustments were used to compare strength changes

181 between quartiles. Strength gain per day (maximum and overall) and starting absolute baseline

182 strength (i.e. first recorded score), number of competitions performed and total days competing

183 for each competitor was used to assess the magnitude and direction of associations for the entire

184 sample and defined with Pearson correlation coefficients (r). Additionally, the strength of linear

185 relationships for maximum and overall strength gain across males and females and within

186 quartiles was also assessed. r values were classified as neglible (0.00-0.10), weak (0.10-0.39),

187 moderate (0.40-0.69), strong (0.70-0.89) or very strong (0.90-1.00) (19). An alpha of 0.05 was

188 adopted for all statitical tests.

189

190 RESULTS

191 Descriptive statistics

192 On average, males competed 61 days longer than females (mean [SD]: 643 [610] vs

193 582 [566] days, P=0.036). The mean (SD) total number of competitions did not significantly

194 differ between sexes (both: 4 (3) competitions; male range: 2-25 competitions; female range:

195 2-32 competitions, P=0.890). Males did not significantly differ in days competing by quartile

196 for SQ (P=0.585), BP (P=0.082) or DL (P=0.302). For females, the number of days competing

197 did not significantly differ between quartile for SQ (P=0.689), BP (P=0.068) or DL (P=0.202).

198

9
199 A total of 151 males (8.0%) were within the highest quartile (i.e. Q4) for all three lift types,

200 whereas 72 females (12%) met this criteria. Approximately 23% of males (n=298) and females

201 (n=145) were in Q4 for two of three lift types, whereas 39% of males (n=501) and 40% of

202 females (n=251) were in Q4 for at least one lift type. Moreover, 32% of males (n=411) and

203 34% of females (n=214) were in Q3 or Q4 for all three lift types.

204

205 Sex differences in strength and strength gain

206 Absolute and relative strength, and maximum and overall strength gain/day and gain/year,

207 regardless of starting strength classification (i.e. quartiles) for males and females are shown in

208 Table 1. Absolute and relative first, maximum and final strength were significantly greater in

209 males compared to females for all lift types (all: P<0.001). Absolute and relative maximum

210 and final strength were also significantly greater than starting strength for all lift types (all:

211 P<0.001). Changes in absolute strength over time did not significantly differ between sexes for

212 SQ (P=0.247), BP (P=0.353) and DL (P=0.379). Similarly, changes in relative strength over

213 time did not differ between sexes for SQ (P=0.821), BP (P=0.661) and DL (P=0.641).

214 Absolute first, maximum and final strength by quartile, sex and lift type are shown in Table 2.

215 Maximum and final strength was significantly greater than baseline strength for all quartiles,

216 regardless of sex or lift type (all: P<0.001).

217

218 Strength gains between the SQ, BP and DL

219 Among the total sample, the change in absolute strength was greater for SQ compared to BP

220 and DL (both: P<0.001), as well as for DL compared to BP (P<0.001). For females, the change

221 in absolute strength was greater for SQ (P=0.001) and DL (P<0.001) compared to BP (Table

222 1). For males, the change in strength was greater for SQ (P<0.001) and DL (P=0.001) compared

10
223 to BP, as well as for DL compared to BP (P<0.001; Table 1). No differences in overall strength

224 gain between the SQ, BP and DL were observed among the total sample or either sex.

225

226 <Insert Table 1 here>

227

228 Effect of strength level on maximum and overall strength gains

229 Changes in strength (i.e. maximum minus first score) for each quartile, lift type and sex

230 can be found in Figure 1A-1F, respectively. For females, maximum strength gain/day did not

231 significantly differ by quartile for SQ (P=0.091), BP (P=0.630) or DL (P=0.746). Similarly,

232 overall strength gain/day for females did not significantly differ by quartile for SQ (P=0.151),

233 BP (P=0.431) or DL (P=0.575) (Table 3). Conversely, maximum strength gain/day for males

234 significantly differed by quartile for SQ (P=0.004) and DL (P<0.001), yet not BP (P=0.216)

235 (Table 3). Post-hoc analyses showed that males in Q4 displayed significantly less maximum

236 strength gain/day compared to those in Q1 for SQ (P=0.002) and DL (P<0.001). Overall

237 strength gain/day for males also significantly differed by quartile for SQ (P=0.027), BP

238 (P=0.004) and DL (P=0.006). Further analyses revealed that males in Q4 experienced

239 significantly less overall strength gain/day compared to those in Q1 for SQ (P=0.018), BP

240 (P=0.012) and DL (P=0.004), as well as those in Q2 for BP only (P=0.007).

241

242 <Insert Table 2 here>

243 <Insert Table 3 here>

244 <Insert Figures 1A-1F here>

245

246 Relationships between strength gain and time

11
247

248 Maximum strength gain over time showed moderate linear relationships for the SQ,

249 BP and DL across all males (Figure 2A-C) and all females (Figure 2D-F). Overall strength

250 gain over time showed weak to moderate linear relationships across all males (Figure 3A-3C)

251 and all females (Figure 3D-3F). More detailed examination of strength quartiles generally

252 showed weak to moderate relationships between maximum or overall strength gain and time

253 and is displayed in Table 4.

254

255 <Insert Figures 2A-2F here>

256 <Insert Figures 3A-3F here>

257 <Insert Table 4 here>

258

259 Relationship between starting strength, competition history and strength gain

260

261 The relationships between maximum and overall strength gain/day for the SQ, BP and

262 DL and starting strength, number of competitions performed and total number of days

263 competing for males and females are displayed in Table 5. Greater starting strength was

264 associated with less overall strength gain/day in males (r = -0.062 to -0.135, P≤0.028), yet not

265 females (r=-0.012 to 0.006, P≥0.522). A greater number of competitions was also associated

266 with less maximum strength gain/day in males only. Greater total days competing was

267 associated with less maximum strength gain/day in both males and females, yet this

268 relationship was not observed when overall strength gain/day was considered (see Table 5).

269

270 <Insert Table 5 about here>

271

12
272

273 DISCUSSION

274 The aim of this investigation was to assess longitudinal changes in upper- and lower-body

275 maximum strength in a large cohort of PL athletes. To achieve this aim, we analysed

276 competition results from 1896 unique PL competitors (~625 females, ~1270 males) over a 15-

277 year period. Specific analyses were performed to determine changes in maximum and overall

278 strength gain (per day and year) from the first recorded competition, between sexes and lifts

279 (i.e. SQ, BP and DL) and starting strength levels (i.e. quartiles). The relationship between

280 maximum and overall strength gain/day and baseline strength, the number of competitions

281 performed and the total number of days competiting were also explored. Collectively, the

282 results show that strength (absolute and relatative) significantly improved for all lifts, that

283 maximum rates of adaptation may differ between lifts, and generally, adaptations occur at a

284 slower rate for the strongest males. However, the change in strength (i.e. maximum and overall

285 strength) did not differ between sexes. Maximum strength gain/day displayed weak negative

286 correlations with total days competing (both sexes), and also with baseline strength and the

287 number of competitions performed (males only). These findings provide detailed information

288 regarding upper- and lower-body maximal strength changes over a considerable time period in

289 strength sport athletes.

290

291 No influence of sex on strength adaptation

292 Despite differences in strength at baseline, and the maximum and final score achieved, the

293 change in strength did not differ between sexes for any lift. These results reaffirm our previous

294 observations that the the change in whole-body strength (i.e. when all lifts are combined) does

295 not differ between sexes in PL athletes (8). Admittedly, our previous work was based largely

13
296 on the same cohort of athletes and so comparison to other PL cohorts and general literature is

297 also warranted in future work. Indeed, the study by Solberg et al. (9) analysed PL athlete

298 performances over a similar time period (15 years) to our current and previous study (8)

299 although the exact number of PL atheltes incorporated into their analysis is not entirely clear.

300 In disagreement to our current findings, Solberg et al. (9) suggests that female PL athletes may

301 ‘possibly’ improve more (2.7±3.8%) over the 5 years prior to peak performance compared to

302 males. Further literature suggests that the similarity of strength adaptations between sexes may

303 be confined to the lower-limbs only (18). At first thought, this assumption appears warranted

304 given earlier research (20) and Table 1 of the current analysis which indicates that, based on

305 kilograms lifted, that there may be a larger difference in absolute strength at baseline (i.e. first

306 score) between sexes for the BP, compared to the SQ and DL (107% greater compared to 73%

307 and 68% greater, respectively) and thus, different strength adaptive potential between the lifts

308 for each sex. However, differences in the change in strength (both maximum and overall) for

309 each lift did not differ between sexes. Some support for this finding is provided by work in

310 general populations where changes in muscle strength and thickness are similar between males

311 and females following 12 weeks of resistance training (21). However, it is unclear how well

312 general population findings would translate to PL athletes. Thus, although the underpinning

313 mechanisms of these observations are outside the scope of this paper, and the retrospective

314 nature of the study cannot account for all possible influencing factors that may be present, they

315 are in contrast to ongoing anecdotal thought and other previous literature in PL athletes (9).

316

317 Possible differences regarding strength adaptation between lifts

318 The current analysis showed that strength gain differed between lifts. Overall, rates of

319 adaptation were greatest for the SQ (~20.2-24.4kg/year), with the BP being the slowest (~10.5-

14
320 12.8kg/year). When sex was explored independently, a similar result was observed. In support

321 of our previous work (8), males in Q4 tended to show slower rates of maximum (SQ and DL)

322 and overall strength gain (all lifts) compared to athletes in Q1-Q2 (see Table 3 for detailed

323 results). Furthermore, our results partially support the findings of Solberg et al. (9) who also

324 reported greater improvements in the SQ compared to other lifts. However, they do not support

325 their findings that gains in strength for the BP are greater than that for the DL. Indeed, our

326 findings appear to be more in line with Colquhoun et al. (17) who showed that in trained males,

327 strength increases were lowest for the BP. The potential reasons underpinning why greater rates

328 of maximum strength adaptation occur for lower-body dominant lifts provokes interesting

329 thought. One possibility is that because greater absolute loads can be lifted for the SQ and DL,

330 there is also more potential to improve (especially when expressed as a change in kilograms).

331 However, when the percentage increase (maximum score/first score x 100) is calculated from

332 the current dataset for each lift there appears to be minimal difference; SQ: 11.7%-13.6%, BP:

333 9.6-12.3%, and DL: 9.4-10.9%, possibly due to the greater percentage point improvement per

334 kilogram in the lift type with the lowest load (i.e. BP). Further, this suggests that despite a

335 similar percentage increase in strength over the analysis period, this change likely occurs in

336 less time for the SQ and may be an important consideration given the contribution of the SQ to

337 competition success (22). Another possible contributing factor is based on data from Nigro and

338 Bartolomei (23) indicating that strength transfer occurs between SQ and DL tasks after only 6

339 weeks of training; that is, performing SQ or DL only training can improve DL or SQ

340 performance by up to 50% and 27% of the strength increase observed in the trained lift,

341 respectively. Unfortunately, concurrent effects on BP performance were not explored. That

342 being said, there is mixed evidence on global ‘heterologous’ muscle adaptations following

343 specific exercise. For example, early studies report no change in handgrip force after

344 quadriceps training (24), no change in handgrip strength after knee extensor and elbow flexor

15
345 training (25), and more recently that strength adaptations are confined to the upper- or lower-

346 body only after upper- or lower-body resistance training, respectively (26). However, some

347 studies suggest that global strength adaptations do occur (27,28). Specifically, a prior study by

348 Othman et al. (28) in youth suggests that strength transfer to an elbow flexor and handgrip task

349 is 3.7-12.3% of the strength increase observed for the leg press after high load, low repetition

350 unilateral leg press training. Based on the mixed evidence to date it is unclear if BP

351 performance would also be positively affected by SQ and DL training specifically. However,

352 PL athletes routinely perform high load, low repetition training which may facilitate the global

353 transfer effect demonstrated by Othman et al. (28). Additionally, in PL the BP involves, and is

354 usually accompanied by instruction to drive the legs and feet hard into the ground. Thus, it can

355 be speculated that increased lower limb strength may also assist in PL style BP performance.

356 However, we acknowledge that this is speculative discussion only and that specific evidence

357 for this effect is not available. Nevertheless, despite some uncertaintly about the causes, the

358 current study is one of the first to demonstrate differences in maximum change in strength

359 between the SQ, BP and DL over a considerable period of time in a large cohort of strength

360 athletes.

361

362 Relationship between starting strength and competition history on strength gain

363 Further examination of potential factors influencing the strength adaptation showed mixed, and

364 inconsistent relationships for maximum and overall strength gain. For example, baseline

365 strength displayed a weak, but significant negative relationship with maximum and overall

366 strength gain/day for males in the SQ, BP and DL. However, this relationship was not observed

367 for females (see Table 5). Conversely, weak negative relationships were observed between the

368 total number of days competing and maximum strength gain/day only, for each lift in both

16
369 males and females. Lastly, the maximum, but not overall strength gain/day appeared to be

370 negatively impacted by having partaken in more competitions for males only. The possible

371 factors contributing to these findings warrant at least some narrative discussion. For example,

372 the data may have incorporated some athletes whose latter stages and end of competitive

373 careers were captured in the analysis. In turn, the negative relationship between total days

374 competing and maximum strength gain/day may suggest that the greatest rates of strength

375 adaptation occur earlier on. Such a claim does not appear unreasonable as strength is thought

376 to improve quickly at the onset of training due to various neural and physiological adaptations

377 (see 29-31). Indeed, Abe et al. (21) report that maximum strength in the knee extension and

378 chest press exercises improve significantly by two and six weeks, respectively, in male

379 subjects. In a more recent study, Davies et al. (32) demonstrated that strength in the BP exercise

380 significantly increased by ~4%, and ~10-12% after only three and seven weeks of training,

381 respectively. Another possibility is that injury or personal factors resulting in less training time

382 or time away from training may have contributed to a reduced capacity to gain and reach

383 maximum strength quickly. However, we highlight that these factors are only speculative and

384 combined with the weak correlational values should be interpreted with caution.

385

386 Limitations and future directions

387 We acknowledge that the findings presented in the article are retrospective in nature and

388 occurred outside of a tightly controlled research environment. However, this is also a novel

389 aspect of the analysis whereby strength adaptations were obtained from strength athletes

390 training for, and participating in real competitions. Although, factors affecting performance(s)

391 of individual athletes, or on a competition-to-competition basis could not be determined in this

392 study, the large sample of PL athletes incorporated into the analysis increases the strength of

17
393 the findings. Moreover, the variability of ages and length of time competing, illness or time off

394 training, or even data regarding variations in training practices and competition factors (e.g.

395 quality of competitors, number of athletes in each competitive class) that could not be captured,

396 in our opinion, provides a conservative but realistic and robust representation of strength

397 adaptive potential in this setting. A caveat, however, is that strength adaptations do not display

398 strong linear relationships over time, and this should be considered when using these findings

399 to plan, progress and assess the effectiveness of strength training programs. Future research

400 may also seek to more closely explore statistical models that help to explain the non-linear

401 relationship(s) between strength gain and time. Additionally, we acknowledge that athlete

402 experience and level (i.e. local, national or international competitor) was not considered in the

403 current analysis. Based on the recent review by Petre et al. (33), concurrent endurance training

404 negatively affects lower body maximum strength development in trained, but not moderately

405 or untrained individuals. That being said, based on the criteria for ‘trained’ being at least three

406 months of structure training in Petre et al. (33), it seems that interference effects are a possibility

407 given the ongoing and longitudinal nature of training and participation for many PL athletes in

408 the current analysis. However, although it was not possible to capture further details about the

409 training history and practices of the included athletes, recent evidence by Shaw et al. (34)

410 suggests that it is unlikely that PL athletes spend considerable time participating in other

411 training modalties (e.g. endurance training). Specifically, Shaw at al. (34) reported that a

412 majority of PL athletes (58% of all PL competitors and 69% of international PL competitors)

413 resistance train as much as 5-6 times per week, with 90-120 minutes the most common training

414 duration. Thus, although this training frequency can not be generalised to every athlete included

415 in the current analysis (e.g. recreational competitor who may also participate in other sporting

416 activities), we can be reasonably confident that concurrent training activities were minimal

417 given the similarity of the cohorts between studies. With that being said, appropriately designed

18
418 future prospective studies that are able to overcome logistical and resource limitations may be

419 able to address more specific questions within the area. Prospective, or even similar

420 retrospective designs as the current study could also be applied in other strength-focussed, or

421 multidisciplinary sports to provide specific and comprehensive information about

422 neuromuscular or performance adaptations over time. Lastly, we have presented information

423 about absolute and relative strength, however, although the latter accounts for bodyweight,

424 changes in muscle mass do not necessarily result in a linear increase in strength. We

425 acknowledge that using other approaches such as the change in Wilks coefficient may account

426 for this somewhat. However, such information likely holds little meaningful information for

427 those outside of PL, or for the generalisation of strength adaptations.

428

429 Collectively, these results provide information regarding real-world upper- and lower-body

430 maximal strength adaptations with considerable periods of strength-training and -sport

431 competition in a large cohort of PL athletes. Maximum rates of strength gain generally appear

432 to be lower for the BP, but adaptation rates are overall similar between sexes. However, the

433 strongest males may, for the most part, gain strength more slowly than their less strong

434 counterparts. These findings may assist strength and conditioning professionals by providing

435 anticipated or likely rates of strength adaptation for the SQ, BP and DL. In turn, this may help

436 to benchmark athletes’ upper- and lower-body strength progression and appropriately plan and

437 periodize resistance training in long-term development models, especially in PL or other

438 strength-focussed sports. In addition, this information also offers more general insight into

439 long-term muscular strength adaptations in humans with likely minimal contamination effects

440 from other training modalities.

441

442 CONFLICT OF INTEREST

19
443 The authors declare they have no conflict of interest and the results of the present study to not

444 contstitue endorsement by ACSM. The results of this study are presented clearly and honestly.

445 FUNDING

446 No funding was receive for this study.

447 ACKNOWLEDGEMENTS

448 Nil

449

20
450 REFERENCES

451 1. Artero EG, Lee D-C, Lavie CJ, et al. Effects of muscular strength on cardiovascular

452 risk factors and prognosis. J Cardiopulm Rehabil Prev. 2012;32(6):351-358.

453 2. Ortega FB, Silverntoinen K, Tynelius P, Rasmussen F. Muscular strength in males

454 adolescents and premature death: cohort study on one million participants. Brit Med J.

455 2012;345:47279.

456 3. Suchomel, TJ, Nimphius, S, Stone, MH. The importance of muscular strength in

457 athletic performance. Sports Med 46(10): 1419-1449, 2016.

458 4. Appleby, B, Newton, RU, Cormie P. Changes in strength over a 2-year period in

459 professional rugby union players. J Strength Cond Res. 2012;26(9):2538-46..

460 5. Baker, DG, Newton RU. Adaptations in upper-body maximal strength and power

461 output resulting from long-term resistance training in experienced strength-power

462 athletes. J Strength Cond Res. 2006;20(3):541-6.

463 6. Szyszka P, Czaplicki A. Analysis of record scores in the clean and its variations in the

464 long-term training of young weightlifters. J Strength Cond Res. 2021; doi:

465 10.1519/JSC.0000000000004086

466 7. Steele J, Fisher J, Giessing J, et al. Long-term time-course of strength adaptation to

467 minimal dose resistance training: Retrospective longitudinal growth modelling of a

468 large cohort through training records [Internet]. SportRxiv; 2021 Available from:

469 [Link]

470 8. Latella C, Teo W-P. Spathis J, can den Hoek D. Long-term strength adaptation; A 15-

471 year analsysi of powerlifting athletes. J Strength Cond Res. 2020;34(9):2412-8.

472 9. Solberg PA, Hopkins WG, Paulsen G, Haugen TA. Peak age and progression in world-

473 class weightlifting and powerlifting athletes. Int J Sports Physiol Perform. 2019;14:

474 1357-63.

21
475 10. Hakkinen, K, Komi, PV, Alen, M, Kauhanen, H. EMG, muscle fibre and force

476 production characteristics during a 1 year training period in elite weightlifters. Eur J

477 Appl Physiol Occup Physiol. 1987;56(4):419-27.

478 11. Hakkinen K, Pakarinen A, Alen M, Kauhanen H, Komi PV. Neuromuscular and

479 hormonal adaptations in athletes to strength training in two years. J Appl Physiol.

480 1988;65:2406-12.

481 12. Miller, JD, Ventresca, HC Bracken, LE. Rate of performance change in American

482 female weightlifters over ten years of competition. Int J Exerc Sci. 2018;11(6):290-307.

483 13. PerasonJ, Spathis JG, van den Hoek DJ, Owen PJ, Weakley J, Latella C. Effect of

484 competition frequency on strength performance of powerlifting athletes. J Strength Cod

485 Res. 2020;34(5):1213-9.

486 14. Keogh JWL, Hume PA, Pearson SN, Mellow PJ. Can absolute and proportional

487 anthropometric characteristics distinguish stronger and weaker powerlifters? J Strength

488 Cond Res. 2009;23(8):2256-65.

489 15. Ferland P-M, Pollock A, Swope R, et al. The relationship between physical

490 characteristics and maximal strength in men practicing the back squat, the bench press

491 and the deadlift. Int J Exerc Sci. 2020;13(4):281-97.

492 16. Colquhoun, RJ, Gai, CM, Walters, J, et al. Comparison of powerlifting performance in

493 trained men using traditional and flexible daily undulating periodization. J Strength

494 Cond Res. 2017;31(2):283-91.

495 17. Colquhuon, RJ, Gai, CM, Aguilar, D, et al. Training volume, not frequency, indicative

496 of maximal strength adaptations to resistance training. J Strength Cond Res.

497 2018;32(5):1207-13.

498 18. Roberts BM, Nuckols G, Krieger JW. Sex differences in resistance training: A

499 systematic review and meta-analysis. J Strength Cond Res. 2020;34(5):1448-60.

22
500 19. Schober, P, Boer, C, Schwarte, L. Correlation coefficients: Appropriate use and

501 interpretation. Anesthesia Analgesia. 2018;126(5):1763-68.

502 20. Latella, C, van den Hoek, D, and Teo, W-P. Factors affecting powerlifting performance:

503 an analysis of age- and weight-based determinants of relative strength. Int J Perform

504 Analysis Sport. 2018;18(4):532-44.

505 21. Abe, T, DeHoyos DV, Pollock, ML, Garzarella, L. Time course for strength and muscle

506 thickness changes following upper and lower body resistance training in men and

507 women. Eur J Appl Physiol. 2000;81(3):174-80.

508 22. Howells RJ, Spathis JG, Pearson J, et al. Impacts of squat attempt weight selection and

509 success on powerlifting performance. J Sports Med Physical Fitness. 2021; doi:

510 10.23736/S0022-4707.21.12140-1.

511 23. Nigro F, Bartolomei S. A comparison between the squat and the deadlift for lower body

512 strength and power training. J Hum Kinet. 2020;73:145-52.

513 24. Hortobagyi T, Scott K, Lambert J, Hamilton G, Tracy J. Cross-education of muscle

514 strength is greater with stimulated than voluntary contractions. Motor Control. [Link]

515 205-19.

516 25. Hellebrandt FA, Parrish AM, Houtz SJ. The influence of unilateral exercise on the

517 contralateral limb. Arch Phys Med. 1947;28:76-85.

518 26. Othman BA, Chaouuachi M, Makhlouf I, et al. Unilateral elbow-flexion- and leg-press-

519 training induce cross-education but not global training gains in children. Pediatr Exerc

520 Sci. 2020;32:36-47.

521 27. Green LA, Gabriel DA. The cross education of strength and skill following unilateral

522 strength training in the upper and lower limbs. J Neurophsyiol. 2018;120:468-79.

523 28. Othman BA, Behm DG, Chaouachi A. Evidence of homologous and heterologous

524 effects after unilateral leg training in youth. Appl Physiol Nutr Metab. 2018;43:282-91.

23
525 29. Moritani T, deVries HA. Neural factors versus hypertrophy in the time course of muscle

526 strength gain. AM J Phys Med. 1979;58(3):115-30.

527 30. Sale DG. Neural adaptations to resistance training. Med Sci Sports Exerc.

528 1988;20(5Suppl):S135-45.

529 31. Staron RS, Karapondo DL, Kraemer WJ, et al. Skeletal muscle adaptations during

530 early phase of heavy-resistance training in men and women. J Appl Physiol.

531 1985;76(3):1247-55.

532 32. Davies TB, Halaki M, Orr R, et al. Effect of set-structure on upper-body muscular

533 hypertrophy and performance in recreationally-trained male and female. J Strength

534 Cond Res. 2021;doi: 10.1519/JSC.00000000000039.

535 33. Petre H, Hemmingson E, Rosdahl H, Psilander N. Development of maximal dynamic

536 strength during concurrent resistance and endurance training in untrained , moderately

537 trained, and trained individuals: A systematic review and meta-analysis. Sports Med.

538 2021;51:991-1010.

539 34. Shaw MP, Andersen V, Saeterbakken AH, Paulsen G, Samnoy LE, Solstad TEJ.

540 Contemporary training practices of Norwegian powerlifters. J Strength Cond Res.

541 2020;doi: 10.1519/JSC.0000000000003584.

542

543

24
544 Figure Caption

545 Figure 1. Displays the change in maximum strength for (A) squat (SQ) for females (n=625),

546 (B) SQ for males (n=1264), (C) bench press (BP) for females (n=624), (D) BP for males

547 (n=1270), and (E) deadlift (DL) for females (n=626), (F) DL for males (n=1268). a P<0.05

548 compared to first quartile (Q1). Each data point represents a single competitor.

549

550 Figure 2. Linear relationships and Pearson’s r values for maximum strength gain over time for

551 the SQ, BP and DL across all males (Figure 2A-C) and all females (Figure 2D-F). All r values

552 are P<0.001. Each data point represents a single competitor.

553

554 Figure 3. Linear relationships and Pearson’s r values for overall strength gain over time for the

555 SQ, BP and DL across all males (Figure 3A-C) and all females (Figure 3D-F). All r values are

556 P<0.001. Each data point represents a single competitor.

557

558

559

560

561

562

563

25

View publication stats