PERCEIVED AND BIOMECHANICAL EFFECTS OF RUNNING SHOE CENTER OF GRAVITY SHIFTING IN AMATEUR MALE RUNNERS
Abstract
This study examined the influence of running shoe center of gravity relative position shifting forward and backward in sagittal axis on male amateur runners. Twenty-three adult male runners were recruited through social media with paid to participate in this study. The experimental shoe used was the Li Ning Feidian Challenger 3. Forward center of gravity (FCG), defined as the shoe center of gravity located at 10.8cm (15% before midpoint) from shoe toe to heel. Intermediate center of gravity (ICG), defined as the shoe center of gravity, is located at 14.8cm (midpoint) from shoe toe to heel. Backward center of gravity (BCG), defined as the shoe center of gravity, is located at 23.1cm (15% after midpoint) from shoe toe to heel. Questionnaire collection was used to assess the perception of the center of gravity shifting. Ground contact temporal, peak force/pressure of plantar and kinetics indicators data were simultaneously captured by motion capture system and force platform. Three participants (13.04%) correctly perceived the shoe center of gravity shifting forward and backward simultaneously. Shoes ICG peak force underneath Meta 1 increased significantly than BCG by 7.59% (p<0.05). Shoes FCG peak force underneath Meta 2 decreased significantly compared to ICG and BCG by 13.62% and 8.96% (p<0.05). Shoes BCG peak force underneath Meta 5 decreased significantly compared to ICG and FCG by 18.18% and 23.78% (p<0.05). Shoes FCG peak pressure underneath Meta 2 decreased significantly compared to ICG and BCG by 13.02% and 9.19% (p<0.05). Shoes FCG peak pressure underneath Meta 2 decreased significantly compared to ICG and BCG by 11.18% and 9.16% (p<0.05). However, there are no significant differences in kinetic indicators. The findings suggest that a fraction of participants can correct perceived shoe center of gravity shifting. Shoes’ FCG reduces force and pressure in the middle metatarsal regions. Shoes’ BCG reduces force in the lateral and medial metatarsal region. Healthcare professionals can optimize the design of footwear accordingly to improve rehabilitation outcomes and reduce injury risks in runners.
1. Introduction
Multiple factors intertwine to influence the impact on consumer purchasing decisions for a particular running shoe is a result of multiple factors intertwined together, including comfort, fit, design style, brand, color, and style.1,2,3 For many running enthusiasts, the competitive performance of shoes is a key consideration. This belief stems from the widely held notion that lighter shoes can improve running efficiency, although recent studies have yielded inconsistent conclusions.4,5,6,7 When choosing running shoes, consumers employ various methods to test them, such as comparing weight by hand, trying them on, and wearing and walking in them for a short period. However, our current understanding of how the feet and lower limbs perceive the quality of running shoes and whether individuals can distinguish quality differences between common running shoe models is insufficient. Mechanical receptors in the feet have a higher activation threshold than those in the hands, primarily due to skin thickness.8,9,10 Early research on lower limb mass perception mainly used the method of additional weights or targeted amputees.11,12,13 More recent studies have asked participants to evaluate the relative mass of running shoes while wearing them. The results show that the feet have a poor ability to perceive the common range of running shoe mass (approximately 220–360g),13 and practice may not improve the accuracy of quality perception. Additionally, no significant difference in quality perception accuracy between males and females was observed.14 Furthermore, hands were found to be much more accurate in perceiving shoe quality than feet.13 Hands could detect small quality differences, even when given shoes that exceeded the normal quality range participants would encounter in daily life.14
Previous research has demonstrated that different types of running shoes, such as Vibram FiveFingers, lightweight shoes, and minimalist shoes, can impact running biomechanics, thus influencing the risk of injury.15,16,17 With design variations, the quality of running shoes also varies. For instance, marathon and Vibram FiveFingers running shoes weigh usually around 150–200g, neutral running shoes weigh 200–300g, and traditional running shoes weigh 360g. Studies have shown that if the combined mass of two shoes is less than 440g, the shoe quality does not harm metabolic costs. However, if the combined mass exceeds 440g, it is positively correlated with metabolic costs.6 It is well known that different running shoe designs can cause changes in foot impact mode during running. Most runners wearing regular shoes adopt a rear foot strike, while minimalist or barefoot runners tend to use a front foot strike or middle foot strike, respectively.18,19 In the past, few studies have explored the independent effects of running shoe mass or center of gravity position, and their impact on running biomechanics remains unclear. Changes in key indicators of running biomechanics can directly affect exercise performance, injury prevention, and even foot strike patterns. Current sports shoe manufacturers have not explained the center of gravity of the shoes to consumers. For consumers, using the “suspension method” could quickly and objectively confirm the position of the shoe’s center of gravity without relying on perception. Unfortunately, the potential impact of different center of gravity positions on running shoes is not yet widely known.
Considering the importance of consumer perception of footwear in making purchasing decisions,2,20 and the impact of these decisions on subsequent sports performance and safety,21,22,23 it is important to understand better the effect of shoes center of gravity position on kinematics. However, there is a lack of research exploring the perception of shoe center of gravity position by feet, and the influence of shoe center of gravity position on running performance and injury prevention. This study examines the influence of running shoe center of gravity relative position shifting forward and backward in sagittal axis on male amateur runners.
2. Materials and Methods
2.1. Participants
Twenty-three adult male runners (28.57±6.46 years, 173.31±5.64cm, 65.86±4.38kg, 12.89±3.78% body fat, 9 US shoe size) were recruited through social media to participate in this study. These runners were classified as Amateur First Class Runner and Amateur Elite Runner according to “compilations of road running management documents of Chinese Athletic Association”. Runners who have visited high altitude or experienced lower limb injuries in the past six months are prohibited from participating in the study. The subjects were informed to use a comfortable daily foot strike pattern throughout the entire study participation process. One participant dropped out as he struggled to run on a treadmill, one participant right knee pain during treadmill measurement, meaning 21 participants were finally included for analysis in this study.
Power analyses of results from previous variations of this protocol indicated that a sample of 18 participants was sufficient to detect group-wise differences as large as those reported between these variations with a power of 0.8.
2.2. Shoe condition
The experimental shoe used was the Li Ning Feidian Challenger 3 (Item number: ARMT037, length: 30.8cm, weight: 209g). Three shoe center of gravity conditions were used in this study (Fig. 1); for each condition, a single shoe was 359g (2×75g lead mass). Forward center of gravity (FCG), defined as the shoe center of gravity located at 10.8cm (15% before midpoint) from shoe toe to heel. Intermediate center of gravity (ICG), defined as the shoe center of gravity, is located at 14.8cm (midpoint) from shoe toe to heel. Backward center of gravity (BCG), defined as the shoe center of gravity located at 23.1cm (15% after midpoint) from shoe toe to heel. Lead mass was distributed on the two sides of the shoe evenly, to ensure that the center of gravity on the coronal axis not changed. The suspension method is used to measure the center of gravity of shoes. All shoes for testing are disguised with black tape, and participants are not allowed to move shoes or lift them off the ground with their hands while wearing and removing shoes. Shoehorns provided by the researcher are used to wear shoes.

Fig. 1. Shoe center of gravity diagram. The FCG, ICG, BCG shoe center of gravity adjustment method is shown in diagram a, b, c, respectively.
2.3. Procedure
This study comprised two visits for each subject.
First Visit: A detailed introduction to the research content was provided to participants, then signed informed consent. Subsequently, subjects wore their shoes during a 15-min warm-up at the subjects’ self-selected running speed. After that, subjects wore FCG, ICG, and BCG shoes at a speed of 4m/s, respectively, running over the force plate. The speed error range is allowed±0.1m/s.
Second Visit: Participants were completed randomly, with at least 48 hours between visits. Body weight (BW) and weight are measured before warm-up. Subsequently, subjects completed a single 4×5-min interval running series on the treadmill. Two specified test sequences (ICG-FCG-ICG-BCG and ICG-BCG-ICG-FCG) are provided to participants for random selection. At the end of FCG and BCG shoe test, participants were asked to provide perception about shoes center of gravity immediately.
2.4. Plantar pressure
Footscan pressure plates (Footscan SCSI 2, RsScan International, Belgium, sampling at 126Hz) were using measure dynamic plantar pressure, where length at 1 m captures the data at a frequency of 126Hz. Before testing began, subjects wore their running shoes to familiarize stride to stride on plates. After that, three valid left and right stance phases were measured, and participants were asked to run at the same speed as during the race. Running speed during testing was measured with infrared gates (Time Tronics, Belgium). A maximal deviation of 10% between the trials was accepted. Running pattern to strike the pressure plate was not allowed to change. For each trial, ten anatomical zones were automatically identified by the software (Footscan software 7.0 Gait 2nd Generation, RsScan International) and, if necessary, manually corrected by adjusting the pixels per zone. These areas were defined as medial heel, lateral heel, midfoot, metatarsal 1 to 5, the hallux and toe 2–5 (Fig. 2).

Fig. 2. The location of ten anatomical zones on the peak pressure footprint.
2.5. Biomechanical measurements
A total of 22 reflective markers (diameter 14mm) were placed over the pelvis and the right leg of the participants (Fig. 3): four pelvis markers (left and right anterior superior spine and Posterior superior iliac spine), medial and lateral epicondyles of the femur, medial and lateral malleolus, three calcaneus markers Paste on the corresponding position of the shoe (posterior upper, posterior lower of the calcaneus and lateral aspects of heel counter), three foot tracking markers paste on the corresponding position of the shoe (medial side of the first metatarsal head, an upper side of the second metatarsal head and lateral side of the fifth metatarsal head) and two four-marker rigid clusters which were attached onto thigh and leg segments. Eight infrared cameras (Vicon T40, Metrics Ltd, Oxford, UK, sampling at 200Hz) were placed around the force plate treadmill (Fully Instrumented Treadmill, BERTEC, Columbus, USA, sampling at 1000Hz) in circular manner, which used for collecting biomechanical indicators.

Fig. 3. Reflective markers scheme: (a) right view; (b) left view.
2.6. Center of gravity perception
To minimize the potential perceptual impact of socks, participants were provided running socks of 95% polyester and 5% spandex composition to wear with experimental shoes, so that the sensitivity of their feet would not be affected by sock thickness and texture. At the end of the second and fourth shoe biomechanical measurements in Visit 2, participants were asked to answer “if the experimental shoe center of gravity variability in sagittal axis than the last shoe”. Four answer options were separated for alternatives: No; Yes, forward; Yes, backward; Yes, uncertain.
2.7. Data collection and analysis
Regarding biomechanical data, the force and marker trajectory of the participants in this study were recorded synchronously by operating Vicon Nexus software (Oxford Metrics Ltd, Oxford, UK). After the marker naming process, export the file to Visual 3D software (C-Motion Inc., Germantown, USA) to calculate and output all the required indicators. Joint angle, angular velocity, RoM, moment, power, and work are calculated using Visual 3D software. The moment, power, and work are calculated using the inverse dynamic model method and standardized by body height (BH) and BW.24,25,26 The kinematic data were processed with a 12Hz cutoff Butterworth fourth-order low-pass filter.27
For the shoe center of gravity variability responses, participant responses were translated to “correct” or “incorrect” perceive experimental shoe center of gravity variability in sagittal axis, then an accuracy percentage was calculated.
Data were analyzed with SPSS software (SPSS 22.0, SPSS Inc., Chicago, USA). Statistical comparison of plantar pressure and biomechanical parameters in running difference shoes center of gravity was repeated measures ANOVA. If a significant main effect was identified, post hoc analysis was performed using least significant difference (LSD), a significance set with p<0.05. Partial eta squared (ηp2) was calculated to determine effect size.
3. Results
All participants (n=21) were asked to provide feedback about their perception of the shoe center of gravity as soon as running stopped (Fig. 4). After the FCG shoes test, 10 participants (43.48%) felt shoe center of gravity change, among these participants, eight (34.78%) felt shoe center of gravity shifting forward, and two (8.70%) felt shoe center of gravity shifting backward. After the BCG shoes test, 13 (56.52%) felt shoe center of gravity change, among these participants, five (21.74%) felt shoe center of gravity shifting forward, and eight (34.78%) felt shoe center of gravity shifting backward. Only three participants (13.04%) perceived the shoe center of gravity shifting forward and backward correctly.

Fig. 4. Percentage of subjects perceive after FCG and BCG test: (a) after FCG test; (b) after BCG test.
The ground contact temporal data and peak force peak pressure of plantar for all participants with different center of gravity are provided at Tables 1–3. Running shoes with different centers of gravity did not result in a significant difference in the ground contact temporal data. Shoes ICG peak force underneath Meta 1 increased significantly than BCG by 7.59% (p<0.05). Shoes FCG peak force underneath Meta 2 decreased significantly compared to ICG and BCG by 13.62% and 8.96% (p<0.05). Shoes BCG peak force underneath Meta 5 decreased significantly compared to ICG and FCG by 18.18% and 23.78% (p<0.05). Shoes FCG peak pressure underneath Meta 2 decreased significantly compared to ICG and BCG by 13.02% and 9.19% (p<0.05). Shoes FCG peak pressure underneath Meta 5 decreased significantly compared to ICG and BCG by 11.18% and 9.16% (p<0.05).
Variable (%) | FCG (%) | ICG (%) | BCG (%) |
---|---|---|---|
Duration of ICP | 9.90 | 8.97 | 8.96 |
Duration of FFCP | 5.43 | 5.70 | 6.40 |
Duration of FFP | 39.07 | 40.54 | 38.63 |
Duration of FFPOP | 45.60 | 44.79 | 46.01 |
Plantar region (N) | FCG (%) | ICG (%) | BCG (%) |
---|---|---|---|
Lateral heel | 23.88±1.86 | 25.88±2.94 | 26.27±2.70 |
Medial heel | 24.88±1.69 | 24.82±2.03 | 24.22±1.90 |
Midfoot | 4.92±1.02 | 4.27±0.88 | 6.94±1.24 |
Meta 1 | 37.83±1.15 | 39.86±1.37# | 34.96±1.38 |
Meta 2 | 43.52±1.14# | 50.38±1.64∗ | 47.80±1.48∗ |
Meta 3 | 36.90±1.29 | 40.91±1.27 | 39.31±1.41 |
Meta 4 | 23.08±0.74 | 22.60±0.77 | 22.71±0.44 |
Meta 5 | 16.22±1.11# | 17.41±0.83# | 13.27±0.88∗ |
Toe 1 | 22.63±2.27 | 29.46±2.54 | 24.37±2.40 |
Toe 2–5 | 31.41±4.09 | 36.10±4.46 | 42.72±5.46 |
Plantar region (kPa) | FCG | ICG | BCG |
---|---|---|---|
Lateral heel | 16.32±1.32 | 17.40±1.96 | 17.73±1.81 |
Medial heel | 17.41±1.29 | 16.55±1.35 | 16.22±1.25 |
Midfoot | 3.35±0.68 | 2.84±0.59 | 4.65±0.83 |
Meta 1 | 25.27±0.78 | 26.60±0.92 | 23.57±0.90 |
Meta 2 | 29.20±0.82# | 33.58±1.09∗ | 32.16±0.92∗ |
Meta 3 | 24.05±0.76 | 27.29±0.84 | 26.47±0.89 |
Meta 4 | 15.35±0.50 | 15.08±0.52 | 15.14±0.51 |
Meta 5 | 10.80±0.74# | 11.63±0.56# | 8.95±0.58∗ |
Toe 1 | 15.24±1.54 | 19.58±1.69 | 16.51±1.62 |
Toe 2–5 | 20.89±2.72 | 24.06±2.97 | 28.88±3.40 |
The kinetics indicators and lower limb joint work for all participants with different center of gravity are provided in Figs. 5 and 6 and Table 4. From the results of lower limb joint work, it was found that, the closer center of gravity to heel, the greater joint work to proximal lower limb; the closer center of gravity to toes, the greater joint work to distal lower limb. However, there are no significantly changes found on statistically.

Fig. 5. (Color online) GRF with different shoe center of gravity. The mean GRF of FCG group, ICG group and BCG group is shown in gray, red and blue solid line. The corresponding colored dotted line shows the standard deviation of each group.

Fig. 6. Lower limb joint work with different shoe center of gravity. The hip joint’s positive and negative work is shown in a1 and a2, respectively. The knee joint’s positive and negative work is shown in b1 and b2, respectively. The Ankle joint’s positive and negative work is shown in c1 and c2, respectively. Pos: positive; Neg: negative.
Indicators (BW/s) | FCG | ICG | BCG |
---|---|---|---|
Max impact loading rate | 117.50±20.53 | 117.46±18.71 | 111.72±16.66 |
Mean impact loading rate | 71.92±0.26 | 70.73±0.25 | 69.48±0.25 |
Max overall loading rate | 59.42±19.60 | 61.51±18.78 | 56.06±20.38 |
Mean overall loading rate | 33.98±11.56 | 34.97±10.65 | 31.38±11.57 |
4. Discussion
This study investigated the runner perception and biomechanical effects, when the running shoes center of gravity located at different positions on the sagittal axis. Three main hypotheses were put forth: (1) Human foot weak to perceive the running shoes center of gravity shifting at a relative position of 15% on sagittal axis. (2) Such shift of center of gravity may cause changes in ground contact temporal data, peak force and pressure plantar on runners. (3) Shoes FCG may lead increase on peak 2, max loading rate 2, mean loading rate 2; Shoes BCG may lead increase on peak 1, max loading rate 1, mean loading rate 1. The experimental results supported part of our hypothesis.
As hypothesized, less than half of participants can correctly perceive changes in the center of gravity of the running shoes on the sagittal axis. The perception of mass includes static mass perception through touch and active mass perception through muscular components.28 Our research focuses on the center of gravity perception during exercise, so black tape was used to block vision and any limbs touching test shoes are forbidden except feet. There are rare research works on running shoes center of gravity position, only a few studies involved in shoes with varied distributions of masses. The studies by Slade and Hausler et al. suggest that subjects were poor at perceiving mass across a range of common running shoe mass from 220g to 360g.13,14 Chiu et al. found that shoes center of gravity closer to the rear end may cause participants tends to feel lighter.29 They are also inclined to choose the shoes with such design in their daily lives, whether they are male or female, when shoes weight remains same. According to our results, a higher proportion of participants report perceived shifting center of gravity in BCG (61.90%) than FCG (52.38%). Equal proportion of participants correctly perceived the direction of center of gravity (38.10%), whether shifting forward or backward. However, only 14.29% of the participants correctly reported the forward and backward shift of the shoe center of gravity. A simple binary Yes/No questions reliable for footwear evaluation, repeating measurement does not show any effect on perceived shoe mass.28 Like the perception of shoes mass, subjects were poor at perceiving across a range of running shoe’s center of gravity from 15% before to after the midpoint.
The ground contact temporal data showed no significantly differences, which did not support for our hypothesis. Using ICG as baseline, FCG lead peak force at Meta 2 peak pressure at Meta 2 and 3 decreased significantly, BCG lead peak force at Meta 1 and 5 decreased significantly. In the research hypothesis, we speculate that the shifting center of gravity of running shoes on the sagittal axis may cause a significant change in force and pressure of plantar at toe and heel regions. Surprisingly, 15% forward shift of center of gravity seems to reduce force and pressure in the middle metatarsal region, 15% backward shift of center of gravity seems to reduce force in the lateral and medial metatarsal region. It is well known that running alter force and pressure pattern of plantar.30 running with difference shoes alter force and pressure pattern of plantar either.31 In our research, shift of the center of gravity in test running shoes only on the sagittal axis, which lead plantar pressure pattern changes on the coronal axis. The change in plantar pressure pattern caused by the change in shoe center of gravity is counterintuitive, which needs finite element analysis to explain the mechanism. By understanding the effects of CG shifts in running shoes on foot pressure distribution and optimizing the footwear design accordingly, healthcare professionals can improve rehabilitation outcomes and reduce injury risks in runners. Further research should continue to explore the long-term effects of CG shifts on rehabilitation strategies to enhance the efficacy of these interventions.
The majority of recreational runners are classified as heel-strikers, who generally exhibit two distinct vertical ground-reaction force (GRF) peaks: An impact peak (First Peak, Peak 1) and an overall peak (Second Peak, Peak 2).32 The impact peak associate with common running-related injuries, the overall peak associate with running performance.33 Average vertical GRF, the occurrence time of peak force and GRF loading rate in our research were no significantly differences. The result data does not support our hypothesis, which illustrate running shoes center of gravity shifting at a relative position of 15% on sagittal axis did not influence GRF related indicators. A speculation has been proposed in this study, further expanding the range of center of gravity transfer may lead to changes in GRF related indicators.
Positive/negative work on lower limb joint indicates energy generated/absorbed during running movements.34 In the result data of this study, no significant changes were found in the lower limbs joint work. However, result data reveal an interesting trend that, BCG presents maximum value energy generation/absorption in the hip joint, ICG presents maximum value energy generation/absorption in the knee joint, FCG presents maximum value energy generation/absorption in the ankle joint. The closer center of gravity to heel, the greater joint work to proximal lower limb. The closer center of gravity to toes, the greater joint work to distal lower limb. The physiological and ergonomic factors in running shoe design have a profound impact on the competitive performance of runners,35 the trend of changes in lower joint work found in this study is a supplement on physiological and ergonomic factors.
There are several limitations of this study. In the recruitment of participants, strike pattern did not been restricted. Participants are being recruited, our research could classify the effect of different shoes center of gravity on different types of strike pattern, the systematic changes can be established for a deeper interpretation. Moreover, a fraction of kinematic indicators that did not show significant changes were not listed in the results such as the maximum eversion angle of the ankle joint. On the other hand, researchers initially assumed that changes in the center of gravity of the sagittal axis would not affect these indicators.
5. Conclusions
In conclusion, a fraction of participants (14.9%) can correctly report the forward and backward shift of shoes center of gravity at same time. FCG reduces force and pressure in the middle metatarsal region, BCG reduces force in the lateral and medial metatarsal region. The findings of this study have important implications for rehabilitation efforts. Finally, there was no evidence to suggest that a 15% movement of the center of gravity along the sagittal axis at the midpoint of a running shoe can cause significant changes in the runner’s kinematic indicators.
Ethical Compliance
Research experiments conducted in this paper with participants were approved by the China University of Mining and Technology — Beijing Ethics Committee. Responsible authorities of our research organization following all guidelines, regulations, legal, and ethical standards as required for humans.
Conflicts of Interest
The authors have no relevant interests to disclose.
Acknowledgments
We thank B. G. Yu and F. Q. Fu, for providing us the technical support for adjusting the center of gravity position. This research was partially supported by Li Ning (China) Sports Goods Company (No. CYF-2023-2002271).
ORCID
Fan Yang https://orcid.org/0000-0001-7240-1936
Tianyu Han https://orcid.org/0009-0002-3822-9022
Tongtong Guo https://orcid.org/0000-0001-9552-308X
Huanhuan Meng https://orcid.org/0009-0003-4818-6019
Minghui Yang https://orcid.org/0009-0006-3668-2156
Zhaoqi Yan https://orcid.org/0009-0001-3333-7262
Zhaowei Chu https://orcid.org/0000-0003-3927-3666
Fengqin Fu https://orcid.org/0009-0002-3489-6637
Yuxi Wang https://orcid.org/0009-0006-5929-2724