Assessing the validity of the zero-velocity update method for sprinting speeds

The zero-velocity update (ZUPT) method has become a popular approach to estimate foot kinematics from foot worn inertial measurement units (IMUs) during walking and running. However, the accuracy of the ZUPT method for stride parameters at sprinting speeds remains unknown, specifically when using sensors with characteristics well suited for sprinting (i.e., high accelerometer and gyroscope ranges and sampling rates). Seventeen participants performed 80-meter track sprints while wearing a Blue Trident IMeasureU IMU. Two cameras, at 20 and 70 meters from the start, were used to validate the ZUPT method on a stride-by-stride and on a cumulative distance basis. In particular, the validity of the ZUPT method was assessed for: (1) estimating a single stride length attained near the end of an 80m sprint (i.e., stride at 70m); (2) estimating cumulative distance from ∼20 to ∼70 m; and (3) estimating total distance traveled for an 80-meter track sprint. Individual stride length errors at the 70-meter mark were within -6% to 3%, with a bias of -0.27%. Cumulative distance errors were within -4 to 2%, with biases ranging from -0.85 to -1.22%. The results of this study demonstrate the ZUPT method provides accurate estimates of stride length and cumulative distance traveled for sprinting speeds.

Modeling the Early and Late Acceleration Phases of the Sprint Start in Elite Long Track Speed Skaters

ABSTRACT

Zukowski, MH, Jordan, MJ, and Herzog, W. Modeling the early and late cceleration phases of the sprint start in elite long track speed skaters. J Strength Cond Res 38(2): 236-244, 2024-This study established the reliability of an exponential function to model the change in velocity during the speed skating sprint start and the validity of associated model parameters in a group of subelite and elite long track speed skaters. Long track speed skaters ( n = 38) performed maximal effort 50-m on-ice accelerations from a standing start while tethered to a horizontal robotic resistance device that sampled position and time data continuously. An exponential function was applied to the raw data to model the change in velocity throughout the acceleration phase and compute the maximal skating speed (MSS), maximal acceleration capacity (MAC), maximum relative net horizontal power ( PMax ), and an acceleration-time constant ( τ ). All constructed models provided a sufficient fit of the raw data ( R -squared > 0.95, mean bias <2%). Intraday reliability of all model parameters ranged from good to excellent (intraclass correlation coefficient >0.8 and coefficient of variation <5%). Strong negative correlations ( r : -0.72 to -0.96) were observed between MSS and PMax and the 10 and 20 m split times measured with the robotic resistance and with 100 split times obtained from 500 m races. Moderate-to-large between-group differences were observed in MSS, MAC, and PMax between the elite vs. subelite speed skaters (Cohen d effect sizes: 1.18-3.53). Our results indicate that monoexponential modeling is a valid and reliable method of monitoring initial acceleration performance in elite level long track speed skaters.

Comparison of kinematics and kinetics between unassisted and assisted maximum speed sprinting

Producing comparable/greater ground reaction forces (GRFs) at faster running speeds is beneficial for sprint performance, and assisted sprint training is used to induce faster running speed conditions. This study aimed to demonstrate the characteristics of assisted sprinting at the maximal speed phase and investigate acute differences to control sprinting. Fifteen sprinters completed control and assisted (5 kg) sprints over force platforms. Assisted sprinting increased running speed (9.3% mean difference), while propulsive mean force (-4.3%) and impulse (-12.4%) decreased, suggesting that running speed improvements were caused primarily by assisted pulling force rather than improvements in anteroposterior force production of athletes. In addition, vertical mean force increased (4.2%), probably due to braking mean force (34.2%) and impulse (32.5%) increases. Magnitude of control trial maximum speed was achieved earlier (during acceleration) in assisted trials, and net anteroposterior (includes both braking and propulsive components) mean force (67.2%) and impulse (67.9%) increased at this matched speed, suggesting that assisted sprints could be used to practice producing greater GRFs at comparable speeds. Running speed improvement by pulling force was associated with contact time decreases (r = -.565), suggesting that shortening contact time may be important for effective assisted sprinting.

Single Leg Lateral and Horizontal Loaded Jump Testing: Reliability and Correlation With Long Track Sprint Speed Skating Performance

ABSTRACT

Zukowski, MH, Jordan, MJ, and Herzog, W. Single leg lateral and horizontal loaded jump testing: reliability and correlation with long track sprint speed skating performance. J Strength Cond Res 37(11): 2251-2259, 2023-This study examined the intraday reliability of 2 novel unilateral loaded jump protocols designed for long track speed skaters. Highly trained ( n = 26), national level athletes performed single leg jumps with a horizontal robotic resistance across 3 external load conditions (10 N, 7.5% of body mass and 15% of body mass) using their dominant limb. Jumps were performed in both the horizontal (Jump Horz ) and lateral (Jump Lat ) direction to replicate the body position and line of force application observed during the running and gliding phases of on-ice acceleration. Subjects completed 2 consecutive trials of the same jump protocol to examine the intraday reliability of the peak velocity achieved for each loading condition. Peak velocity across each jump type and loading condition had good reliability (intraclass correlation coefficient >0.8, coefficient of variation <5%). Significant positive relationships ( r = 0.5-0.8, p < 0.05; n = 22) were observed between all jump conditions and on-ice sprint race split times obtained including 100, 400, and 500 m. Our results indicate that unilateral loaded jump tests are reliable in speed skating athletes and may help practitioners diagnose and monitor lower-limb maximal muscle power capacity in a sport-specific manner.

Reliability of phase-specific outcome measurements in change-of-direction tests using a motorized resistance device

This study aims to determine test-retest reliability of phase-specific information during initial acceleration, deceleration, and re-acceleration phases of different change-of-direction (CoD) tests using a motorized resistance device (MRD). A total of 21 participants (16 males and five females, with mean age of 22.3 ± 3.9 years, body mass of 75.2 ± 6.9 kg, height of 177.9 ± 6.8 cm) completed the modified 505 (m505), 10-0-5, and 15-0-5 CoD tests on four different test sessions while exposed to an external load (3 kg) provided by the MRD. Outcome variables included overall and phase-specific kinetic (force, power, and impulse) and kinematic (time, distance, velocity, and acceleration/deceleration) data during the initial acceleration, deceleration, and re-acceleration phases. The deceleration and re-acceleration phases were further divided into two subphases, namely, early and late subphases, using 50% of maximum velocity. Reliability was assessed using an intraclass correlation coefficient (ICC), coefficient of variation (CV), typical error (TE), and minimal detectable change (MDC). Good to excellent ICC values (>0.75) and acceptable (<10%) to good (<5%) CV values were observed for most outcome measurements. Specifically, 80.1% (822 out of 1,026) of all variables showed good or better relative reliability (i.e., ICC ≥ 0.75), while 97.0% (995 out of 1,026) of all variables showed acceptable or better absolute reliability (i.e., CV < 10%). In conclusion, the present study demonstrates that the MRD can obtain reliable phase-specific outcome measurements across different CoD tests, providing coaches and researchers with new opportunities to advance our understanding of CoD ability and inform more advanced CoD training prescriptions.

A Load-Velocity Relationship in Sprint?

The aims were to compare predicted maximal velocity from load-velocity relationships established with different resisted and assisted loads by different regression analyses to the measured maximal velocity during sprint running, and to compare maximal velocity measured between a robotic pulley system and laser gun. Sixteen experienced male sprinters performed regular 50 m sprints, a 50 m with 5-kilogram-assisted sprint, and 10, 20, 30, and 30 m resisted sprints with, respectively, 65, 50, 25, and 10% calculated reduction in maximal velocity. Maximal velocity obtained by laser gun during the regular sprint was compared with predicted maximal velocity calculated from four trendlines (linear and polynomial based upon four resisted loads, and linear and polynomial based upon four resisted and one assisted load). Main findings demonstrate that the robotic pulley system and laser measure similar maximal velocities at all loads except at the load of 10% velocity reduction. Theoretical maximal velocity based upon calculated predictions were underestimated by 0.62-0.22 m/s (2.2-0.78 km/h; 6.7-2.3%) compared to measured maximal velocity. It was concluded that different regression analyses underestimated measured maximal velocity in regular sprinting and polynomial regression analysis (with resisted and assisted loads) estimation was closest to measured velocity (2.3%).

Quantification of horizontal force for the EXER-GENIE® resisted sprint training device

Sport performance coaches use a range of modalities to apply a horizontal force (Fh) to athletes during resisted sprint training (RST). These modalities include parachutes, weighted vests, pulley devices, motored tethered devices, and, most notably, weighted sleds. Despite the widespread use of these devices, the resistance forces of the pulley devices have not been evaluated for reliability and accuracy. Therefore, the primary aim of this study is to quantify the Fh of a commercially available pulley device (EXER-GENIE®) and determine how resistance force is related to the load settings on the device. The secondary aim is to identify the differences in the Fh values between three EXER-GENIE® devices that use 36 m and 60 m ropes. The Fh values in the Newtons (N) of the three EXER-GENIE® devices were analyzed using a motorized winch, a lead acid battery, and an S-beam load cell. Four 10 s winch-driven trials were performed using 15 different EXER-GENIE® loads, ranging from 0.028 kg to 3.628 kg, employing two different 36 m devices and one 60 m device. The mean ± standard deviation for Fh was reported across the four trials for each load setting. All devices produced similar Fh values across lighter load settings (loads ≤0.141 kg). However, at heavier loads (loads ≥0.226 kg), the 60 m device had Fh values 50-85 N greater than those of the 36 m device. The coefficient of variation across the four trials was extremely high at light loads but sharply decreased to <10% at heavy loads. Absolute reliability was high for each device [intraclass correlation coefficient (ICC) = 0.99]. A regression analysis for Fh values and EXER-GENIE® load indicated a strong positive relationship between load and Fh values across all devices (R2 = 0.96-0.99). Caution should be exercised when using identical loads on the different-length pulley devices, as the 60 m device produced greater Fh values than the 36 m devices at load settings higher than 0.226 kg. These results can provide coaches and practitioners with a better understanding of the magnitude of resistance that is applied when prescribing EXER-GENIE® devices for higher training loads.

Relevance of force-velocity and change of direction assessments for the ranking position in elite junior tennis players

Purpose

This study aimed to correlate sprint mechanical parameters (SMP) of a linear sprint (LS) and a tennis specific modified 505 (Tm505) change of direction (CoD) test obtained with a motorized resistance device (MRD) to the current tennis ranking position (RP).

Methods

107 male and 86 female elite junior tennis players nationally ranked in the German Tennis Federation between 10 and 18 years participated in the study. According to their age at peak height velocity (PHV), players were divided into pre-PHV, circa-PHV, and post-PHV groups. SMP were derived from instantaneous time-velocity data of two 20 m all-out LS measured with 333 Hz. Further, mean values from two Tm505 trials with constant 3 kg loading over acceleration-deceleration (1a) and reacceleration (1b) phases were measured with an MRD. SMP of LS and CoD measurements were partially correlated with the current RP in the overall national ranking by controlling for biological maturation.

Results

Low to moderate correlations (r_s  = -0.1 to -0.3) were found between SMP and the RP in all male and female age groups. Correlations of the CoD measurements were overall more pronounced, particularly in girls (r_s  = -0.44). All linear SMP, like maximal theoretical force (F₀; N/kg), and maximal theoretical velocity (v₀; m/s), maximal power (P_max; W/kg), improved over maturation for both genders with P_max being most important for sprint performance. Further, P_max was shown to correlate with the girls ranking position (r_s  = -0.31). During the Tm505, matured players achieved significantly faster overall total and CoD times. Positioning of CoM before CoD enlarged over maturation and was found to correlate to the RP in both sexes. In addition, nearly all SMP significantly correlated to the primary performance outcomes in the Tm505 test in both genders (r = -0.3 to -0.6).

Conclusion

CoD performance has a moderate and higher impact on tennis performance compared to LS. CoD performance as well as P_max achieved a higher relevance for the ranking position predominantly in girls compared to boys. Hence, particularly P_max as well as the transfer to on-court CoD motor skills should be a central training goal in elite junior tennis players besides technical skills and should depend on maturation status and gender.

Sprint Acceleration Mechanical Outputs Derived from Position- or Velocity-Time Data: A Multi-System Comparison Study

To directly compare five commonly used on-field systems (motorized linear encoder, laser, radar, global positioning system, and timing gates) during sprint acceleration to (i) measure velocity−time data, (ii) compute the main associated force−velocity variables, and (iii) assess their respective inter-trial reliability. Eighteen participants performed three 40 m sprints, during which five systems were used to simultaneously and separately record the body center of the mass horizontal position or velocity over time. Horizontal force−velocity mechanical outputs for the two best trials were computed following an inverse dynamic model and based on an exponential fitting of the position- or velocity-time data. Between the five systems, the maximal running velocity was close (7.99 to 8.04 m.s−1), while the time constant showed larger differences (1.18 to 1.29 s). Concurrent validity results overall showed a relative systematic error of 0.86 to 2.28% for maximum and theoretically maximal velocity variables and 4.78 to 12.9% for early acceleration variables. The inter-trial reliability showed low coefficients of variation (all <5.74%), and was very close between all of the systems. All of the systems tested here can be considered relevant to measure the maximal velocity and compute the force−velocity mechanical outputs. Practitioners are advised to interpret the data obtained with either of these systems in light of these results.

Validity of Velocity Measurements of a Motorized Resistance Device During Change of Direction

The aim of this study was to determine validity of velocity measurements of a motorized resistance device (MRD) during change of direction (CoD). Eight male (age: 22.1 ± 4.2 yrs; weight: 83.3 ± 17.1 kg; height: 181.6 ± 12.6 cm) and three female participants (age: 21.7 ± 1.5 yrs; mass: 69.7 ± 2.4 kg; height: 167.0 ± 3.6 cm) completed the modified 505 CoD test (m505) with turning off the left and right foot while exposed to external loads (3, 6, and 9 kg) provided by the MRD. Three-dimensional kinematic data were measured (200 Hz) for all tests using a full-body marker set with an additional marker placed on the pulley used to attach the carabiner (CAR) at the end of the line from the MRD to the participant. Average velocity of overall center of mass (COM_vel), pelvis (COM_{pelvis_vel}), and pulley (CAR_vel) was then calculated and compared to the velocity measured by MRD (MRD_vel) in 0.5 s intervals 1.5 s before and after CoD. Average velocities from these intervals were then compared using correlational, Bland–Altman analysis, coefficient of variation (CV), and statistical parametric mapping (SPM). Mostly, excellent correlations were observed and ranged from 0.93 to 1.00, 0.53 to 1.00 and 0.93 to 1.00 for the 3, 6, and 9 kg load conditions, respectively. CV values ranged from 0.3 to 3.2%, 0.8 to 4.3%, and 1.5 to 7.7% for the CAR_vel, COM_{pelvis_vel}, and COM_vel comparisons, respectively. The biases for CAR_vel comparisons ranged from −0.027 to 0.05 m/s, −0.246 to 0.128 m/s and −0.486 to 0.082 m/s across all load conditions and time intervals for the CAR_vel, COM_{pelvis_vel}, and COM_vel comparisons, respectively. SPM analysis indicated significant differences between MRD_vel and COM_vel and COM_{pelvis_vel} over short time periods during the CoD, but no difference between MRD_vel and CAR_vel. The velocity measurements obtained by a MRD during a m505 test are valid as low biases, low CV’s, and high correlations are observed for the MRD_vel to CAR_vel comparison. As single points of measurement (i.e., laser) has been proven useful to assess other athletic tasks (i.e., sprint running), the single point CAR_vel comparison is an appropriate comparison for validating MRD_vel measurements during the m505 test.