Recently my twitter feed was filled with the discussion on jump testing methodologies, measurement accuracy, arm swing and a plethora of other related topics to vertical jump testing. Trying my hardest to keep up (remaining somewhat entertained throughout), I resisted the temptation to join the debate and instead sat back and tried to piece it all together. There were some questionable lines of thought (in regards to the questions being asked) though we will approach this as there is no such thing as a stupid question…
This post will focus on a twitter poll questioning the most accurate measure of jump height.
Most accurate measure of jump height
Comment = Other
— Official Strength Debates (@StrengthDebates) January 21, 2023
(also included is a pretty funny ongoing twitter battle between @RUGBY_STR_COACH and @ByrdSportsPF to further exacerbate the issue of vertical jump testing – But we will leave that alone for now…).
These threads seemingly contained similar lines of thought with some good experts in the field providing valuable contributions. Within this blog posts I look to provide some additional perspectives to the following questions, which are what I gathered as the key areas of discussion from the threads:
Unpacking what is the “most accurate” measure of jump height?
Can a force plate accurately identify jump height?
Can we determine the most accurate device by considering errors amongst other factors?
DISCLAIMER: I am not a biomechanist by trade. However I have a keen interest in applied methods of jump testing, and have subsequently utilized all the methods listed in the original poll.
UNPACKING THE QUESTION OF ACCURACY OF MEASURES OF JUMP HEIGHT:
Firstly let’s start with some conceptual thinking. Any measure of jump height is an estimation of the true value ± error (regardless of device/method used to collect it – though the device/method may be important as we will discuss later). The error represents the uncertainty in the measure (which we can also view as a measure of the “accuracy” of the device used to compute the jump performance). In an ideal scenario we want our testing equipment to correctly identify what we are measuring and carry low amounts of error, reducing our uncertainty and consequently providing more confidence in the precision of the estimate (e.g. so we can ultimately say this is a good representation of vertical jump height performance).
From the aforementioned twitter threads a discussion entailed of which a key theme was what device offers the best (most accurate) measure of jump height? Unfortunately this was somewhat a loaded question, mainly because it didn’t detail specifically what was meant by jump height. In this sense Dr Peter Mundy and some others detailed this as the need to better provide operational definitions to the question;
If the user is looking to measure jump height as a result of total reach during a vertical jump task, with the aim that the specificity is similar to that displayed in sports like basketball and volleyball etc… (jump + touch representing the highest point an athlete can reach vertically), then the Vertec offers a good practical solution to measure this.
If a user is interested in a means to assess the vertical displacement of the system’s center of mass and estimate the jump height from this accurately using principles of motion, then a force plate calculating vertical ground reaction forces offers a viable solution.
Therefore an important aspect of this process first starts with THE WHAT & THE WHY.
The what, specifically detailing “what it is you are trying to measure” in the first place, and the why, being the reason(s) it is important to measure it in the context of your “what” performance question is and what you’re ultimately hoping to get out of it.
As I briefly touched on above, a coach may be looking to specifically measure something that may directly relate to the sports specific task/skill that relates to competitive performance (i.e. the Vertec representing a maximal jump height + reach that may relate to an athlete’s capability to perform a block in both basketball and volleyball).
A coach alternatively may be interested in understanding kinetic contributions as a result of kinematic execution of a jump and thus would require a means to measure forces acting about the body (i.e. the use of a force plate). This may inform on what training goals the athlete may need to complete and provide a means of measuring the progression of training. This same coach may also want to use this as a measure of neuromuscular function, with “the why” attempting to detect athletes progression or fatigue in relation to training and competition. They may find it easier to standardize this process with the use of a force plate providing further metrics beyond just that of traditional jump height (which may not inform on strategy, execution or fatigue). These hypothetical questions and scenarios are likely common among many applied sport performance environments.
With this in mind the method chosen to measure the jump height is also dependent on the questions of what and why in addition to the specific case of the environment. For example if an environment has 60 athletes and access to one of the following: Just Jump Mat, Vertec, Force Plate, Camera Based App like MyJumpApp, it is likely that the most time efficient testing protocols would consist of the Just Jump Mat and Force Plate for this number of athletes, followed by Vertec and finally MyJumpApp which require some pre/post processing respectively. As such a practitioner also has to think about what practical solution works best within their environment while yielding the right information for the questions mentioned above.
The combined answer of what Is needed to provide a practical time efficient means to measure jump height in a reliable way may consequently determine the specific use of what technology/device method is used to calculate the jump performance. The practitioner is then left with a decision on what method they wish to pursue.
Once this is determined what we really want to know is how valid and reliable is that method (i.e. does it measure what it says it measures, is the measure repeatable and does the measure carry a small level of error acceptable to detect change over time). Much of this is covered in a reliability analysis, whereby one may compare to a criterion (typically a gold standard measure) or assess the repeatability of the device across a specific timeframe. Oftentimes within research this consists of looking at correlations (typically intraclass correlations ICC coefficient), mean differences / effect sizes, Bland Altman analysis with limits of agreement and calculating the standard error of the measurement (which may also be displayed as coefficient of variation value CV%).
It should also be noted that we are dealing with human movement and thus there is additional error introduced to any analysis of this type due to biological variation in addition to the precision specification of the method used. Therefore, it is important that we assess these qualities to determine whether the method is viable to be used to detect the measure accurately in relation to the variation invoked through biological factors.
Firstly we will start with a brief dive force plate technology and why this has been readily adopted as a gold standard measure of vertical jumping.
CAN A FORCE PLATE ACCURATELY MEASURE JUMP HEIGHT?
The simple answer here is yes. But we will provide a short dissemination of key literature as to why. Firstly let’s start with a quick overview of the specifications of a force plate.
Typically force plates consist of a configuration of either strain gauges, load cells or piezoelectric elements that create an electrical output when a force is applied directly into the plate. This can be seen as akin to a household scale, in which we would measure body weight. The difference being that a force plate typically offers a higher sampling frequency (~100 to ~1000Hz) and a detection method which allows for dynamic actions (and resulting forces) to be quantified. From a simple perspective when we stand still on a force plate the resulting force = mass * gravity, which consequently provides an estimate of weight for the person standing on the plate. Unlike a force plate a household scale would lack sensitivity to detect such movements like jumping. However, due to these sampling frequencies being greater than the typical contact times seen for vertical jumping, we can effectively measure the key movement signatures of jumping tasks.
A breadth of research exists detailing the accuracy within calibration and sampling processes of force plate technology. The early research I am more familiar with was that of Major et al (1998) and Cross (1999) though the use of force plate technology dates back to the 1950-60s when it comes to vertical jumping (Payne et al 1962). Typically force plate calibration has been performed by using an object with a known mass and assessing the force plate’s ability to accurately determine this mass in relation to gravity, as a resulting output of force (N) (Major et al, 1998). As such when a person steps onto a force plate the resulting F = M*g (Cross, 1999). This creates a Given that a force plate can accurately determine force of any given object mass placed upon it we can then use principles of Newton’s laws to determine jump height.
As detailed in Newton’s law of inertia, motion occurs as a result of a change in momentum of a body when a force is acted on it. When it comes to vertical jumping, this force is applied over time within different phases of the jump with the resulting center of mass moving vertically (Linthorn, 2001). We can quantify this as impulse, with the formula:
The use of impulse makes more sense when we consider Newton’s law of acceleration (the impulse momentum relationship). As this law states the impulse is equal to the change in momentum of the body. Therefore we can assume that impulse when calculated correctly is able to explain jump performance.
(Copied from Jordan et al, in aspetar journal)
Given that the only force acting on the person jumping is gravity and vertical velocity at the peak of the jump is 0 m/s, Jump height can consequently be estimated from the following simple equation:
Jump Height = ½ * (Takeoff Velocity2 / Gravity2)
As such the use of net impulse using a force plate has been proposed as an effective way to report on jump performance and a means to estimate jump height (Linthorne, 2001; Miziguchi, 2012).
When calculating vertical jump height there has been numerous studies conducted providing validity and reliability to force plate technology. Cormack et al (2008) also demonstrated reliability low standard error (SE) for both countermovement jumps and repeated countermovement jumps within intraday and interday (SE = ~2 cm CV% = ~5%) sessions, sampled at 200Hz. In a study by Lake & Colleagues (2018) they assessed a portable 1-D Dual force plate (Pasco) in comparison to a laboratory standard Kistler force plate. Using jump height calculated through the impulse momentum method they noted 3.7% CV% values for both systems sampling at 1000 Hz. When using a countermovement jump with arm swing CV% values ranged from 3.88 – 8.07% in Australian soldiers across 3 testing occasions on a single force plate sampling at 1000 Hz (Smith et al, 2022). Taken together these studies suggest force plates are able to provide reliable estimates of vertical jump height. Though this blog only offers a snapshot of the current literature assessing the reliability of force plates and doesn’t cover other important variables that are characteristic of analysis of force time curves.
Some practical considerations:
An important aspect to maximizing the reliability and subsequent accuracy of jump height calculation using impulse from a force plate is test standardization. Firstly a force plate always needs to be zeroed so an accurate recording of mass can be obtained. When utilizing the impulse-momentum theorem to estimate the vertical displacement of center of mass the force plate and subsequent analysis software require an accurate measure of the system mass (which is typically taken from a quiet period whereby the athlete stands still).
If the athlete is not weighed correctly there can be errors introduced into the calculation, for example for every 10N difference in body weight from actual body weight there may be a difference of 2-3 cm in flight height calculated (Linthorne, 2001). Athletes may raise heels before jumping, perform small movements of the feet and thus change the forces applied to the plate, which could influence impulse related calculations. Therefore correct instruction is imperative to minimize these errors with the athlete being instructed to be as still as possible prior to jump onset (approximately for 2 seconds; Linthorn, 2001). This will also depend on the type of detection threshold used within the specific sampling software to establish the net impulse.
Alternatively a force plate can also use the flight time method to calculate jump height (the time taken between take off and landing), which likely provides a higher estimation of jump height in comparison to the impulse method (Linthorn, 2001). However, this makes for a pretty expensive way to calculate jump height when other systems (i.e. contact mat) can provide a similar estimate utilizing the same calculation for a 5th of the cost. Notwithstanding it is likely though practitioners wish to look beyond primarily jump height as a measure when profiling athletes during a vertical jump, with a force plate providing a plethora of force-time characteristics distinct to the individual performing the jumping task.
Summary of this section:
Given that force plates can typically sample data at 1000 Hz (meaning 1000 force samples per second) it has the ability to accurately collect force profile information of jumping tasks with minimal ground contact times.
Given the principles of motion, a force plate provides a means of estimating forces that can then be used to apply impulse-momentum based calculations to calculate jump height, as a resultant of center of mass displacement accurately
A force plate is highly dependent on the standardization of the testing process to ensure higher levels of reliability. When performed correctly the estimation of jump height has been shown to be highly valid and reliable making force plates one of the gold standards for quantifying jumping performance.
CAN WE DETERMINE THE MOST ACCURATE DEVICE BY CONSIDERING TEST ERROR AMONGST OTHER FACTORS?
There were a number of common jump methods detailed in the previously mentioned twitter poll. As we already covered force plate information (above) we will focus on other methods mentioned to determine if a most accurate device can be identified…
The system that seemingly got the second most love after Force Plates (at least from the outcomes of the aforementioned twitter poll) was the Vertec Standing Reach Test. This is a common method employed in NFL and NBA combine testing procedures, and has been a practical cost effective means to measure vertical jump height in athlete populations. For those unfamiliar with this type of test the Vertec comprises of a number of panes
When evaluating the current body of literature, Leard et al (2007) compared the Vertec in relation to 3 Camera Motion capture (utilizing a reflective marker placed at the sacrum to measure vertical displacement). When estimating jump height, the mean differences between the Camera System (43.8 cm) and the Vertec (39.5 cm) was shown to underestimate jump height. Similarly in comparison to a force plate, the Vertec was again shown to underestimate jump height (50.3 cm vs 47.9 cm; Buckthorpe et al, 2012).
Alternatively Cheah et al (2016) showed that compared to a force plate (using the flight time method of calculating jump height) the Vertec overestimated jump height (36 ± 9 cm vs 55 ± 12 cm). Similarly, Petushek & Colleagues also showed overestimated differences between force plate jump height (also calculated by flight time) and Vertec performance ( 40.2 ± 9 cm and 55.4 ± 10 cm). More recently McClelland et al (2022) determined Vertec performance was overestimated jump height (more specifically displacement of center of mass COM) in comparison to force plate method using impulse to derive Jump height (54 ± 14 cm vs 40 ± 11 cm).
In the case of Buckthorpe et al (2012) study some explanation for these differences may be the specific jump protocol used, as these jumps were not all performed at the same time (which means some differences may be due to typical variation of jump trials despite the authors attempts to provide randomization). There is also a performance factor of the Vertec in which the person is expected to hit the panes at the peak of their vertical displacement, however in practice this takes some familiarity with the task and may also be another reason why errors may arise in comparison to other criterion devices, which may be apparent in the populations studied in each example.
Additionally the Vertec protocol requires prior calibration by estimating reach height, and therefore this will be highly dependent on the protocol used to determine standing reach (Ferreira et al, 2010). As such it is suggested that performance execution of the Vertec jump is equally about the ability to reach as it is about jumping performance e.g. displacement of center of mass (McClelland et al 2022). From a practical perspective despite best attempts to standardize reach distance prior to testing, some athletes may display exaggerated asymmetrical positions (i.e. hip hike) that may account for greater jump heights shown recorded during Vertec jumping.
Despite this the Vertec device may still be considered to be a reliable means of jump testing. In an investigation utilizing university students Nuzzo et al (2011) showed both the intraday and interday reliability to be acceptable in both male and female cohorts (intraday; SEM= 2.1-2.3 cm, CV% = 4.6-7.6%). Such values are similar to those we displayed earlier for force plate technology. As such despite the discrepancies between estimating COM displacement, Vertec can still offer a repeatable means to assess jump height + the skill of reaching. Though it should be noted that this can only be measured to a precision of 0.5 inches / 1.25 cm and may also be influenced by learning effects (i.e. the participants may jump higher with more familiarity; Nuzzo et al, 2012).
As we previously mentioned the Nuzzo et al (2011) study this also provided the reliability estimates for a jump contact mat (in this case Just Jump, which is a popular piece of apparatus in applied sport environments). The Just Jump showed similar error across intra and interday sessions (SEM = 1.6-2.3cm, CV% = 4.2-5.2%) across males and females. These authors also suggested that this may be a more appropriate device to test female athletes in comparison to Vertec due to lower error estimates.
Jump mat technology lends its effectiveness from the simplicity of estimating jump height from flight time (i.e. the time taken between final contact and landing of the jump). As such this method is closer related to that of COM displacement estimation using camera based tracking (Leard et al, 2007). This was shown to be closely related to the criterion measure (Just Jump = 44.2 cm, Camera system = 43.7 cm). Suggesting that this system is more closely related to camera based measures in comparison to the use of a Vertec. Which is likely due to the methods of calculating the resulting jump height (as mentioned previously).
Continuing the theme of Just Jump McMahon & Colleagues (2016) showed excellent reliability and comparisons to a criterion force plate. It was noted that the Just Jump System displayed a CV% of 4.7% in comparison to the 3.7% displayed by the force plate. It was also suggested that the just jump system showed an overestimation of Jump Height, and that a correction equation should be applied to provide better estimates in comparison to that which we would see from a force plate. Again as both methods use slightly different calculations (with a jump mat relying on flight time, which can be overestimated based on a differing position between take off and landing), it is unsurprising to see that both estimates don’t completely agree.
There are also other manufacturer types of contact jump mats that have also been compared to force plate technology, here we covered primarily that of Just Jump (primarily as this is the one device I have most experience with and thus could provide a quick overview of some key research to evaluate the question of accuracy). Nonetheless the jump mat can provide a reliable method of detecting jump height (from the estimation using flight time). Though this may depend on technical factors (such as landing position).
This post has looked to expand beyond the twitter thread discussions around this topic and has hopefully provided additional information to add to the discussion. There are of course plenty of more research papers to provide additional discussion, with this post not intending to be an extensive review of the literature.
The main points to focus on for me are as follows:
Methods to measure Jump Height provide estimates of the true value ± error, recognizing this means that we want to embrace the uncertainty surrounding the measurement of jump height.
The choice of method to measure Jump Height should start with what you are aiming to measure and why in addition to the practicality of the system for your environment (including affordability).
There are multiple studies that provide validity and reliability estimates for each device discussed. Suggesting each device is an accurate means of assessing the construct of jumping. Though it is likely that a Force Plate is capturing the closest estimate of COM displacement for the calculation of Jump Height in comparison to methods like the Vertec and contact mats. (Force plates are however the most expensive system so you’re paying for precision).
It is likely that Just Jump and Vertec will overestimate Jump Height in comparison to Force Plate Technology. With Vertec likely providing the highest estimation. This is most likely due to the method of estimating
Finally we have learnt that as an S&C community, a discussion of a simple poll can spark some really interesting engagement and thought process whereby a number of experts in the biomechanics field provided some great short & potent insights to the questions contained by a simple poll.
Buckthorpe, M., Morris, J., & Folland, J. P. (2012). Validity of vertical jump measurement devices. Journal of sports sciences, 30(1), 63-69.
Cheah, P. Y., Cheong, J. P. G., Razman, R., & Zainal Abidin, N. E. (2017). Comparison of vertical jump height using the force platform and the vertec. In 3rd International Conference on Movement, Health and Exercise: Engineering Olympic Success: From Theory to Practice 3 (pp. 155-158). Springer Singapore.
Cross, R. (1999). Standing, walking, running, and jumping on a force plate. American Journal of Physics, 67(4), 304-309.
Ferreira, L. C., Schilling, B. K., Weiss, L. W., Fry, A. C., & Chiu, L. Z. (2010). Reach height and jump displacement: Implications for standardization of reach determination. The Journal of Strength & Conditioning Research, 24(6), 1596-1601.
Leard, J. S., Cirillo, M. A., Katsnelson, E., Kimiatek, D. A., Miller, T. W., Trebincevic, K., & Garbalosa, J. C. (2007). Validity of two alternative systems for measuring vertical jump height. Journal of Strength and Conditioning Research, 21(4), 1296.
Linthorne, N. P. (2001). Analysis of standing vertical jumps using a force platform. American Journal of Physics, 69(11), 1198-1204.
Major, J. A., Sands, W. A., McNeal, J. R., Paine, D. D., & Kipp, R. (1998). Design, construction, and validation of a portable one-dimensional force platform. The Journal of Strength & Conditioning Research, 12(1), 37-41.
McClelland, Emily L.; Prajapati, Sunil K.; and Weyand, Peter (2022) “Quantifying Getting High Under One’s Own Power – A Comparison of Vertical Jump Height Measurement Methods,” International Journal of Exercise Science: Conference Proceedings: Vol. 2: Iss. 14, Article 53.
McMahon, J. J., Jones, P. A., & Comfort, P. (2016). A correction equation for jump height measured using the just jump system. International journal of sports physiology and performance, 11(4), 555-557.
Mizuguchi, Satoshi, “Net Impulse and Net Impulse Characteristics in Vertical Jumping” (2012). Electronic Theses and Dissertations. Paper 1459. https://dc.etsu.edu/etd/1459
Nuzzo, J. L., Anning, J. H., & Scharfenberg, J. M. (2011). The reliability of three devices used for measuring vertical jump height. The Journal of Strength & Conditioning Research, 25(9), 2580-2590.
Petushek et al (2010) Comparison Of Jump Height Values Derived From A Force Platform And Vertec, https://ojs.ub.uni-konstanz.de/cpa/article/view/4526