Data collection: What’s it good for?

At Acumen, we regularly collect data or information on clients in different variables.

For example, we may periodically assess body fat mass in someone whose primary goal is weight loss or body composition. Similarly, we might measure types of jumping performance in an athlete where jumping is critical to their sport (e.g., volleyball, basketball). Acumen Performance has also begun to organize and run testing combines to collect large amounts of musculoskeletal (e.g., range of motion, postural) and performance (strength, power, endurance) data on athletes of the same/similar sports (e.g., rodeo athletes). While these combines are fun and allow an excellent opportunity for us to get to know our clients, the question remains, what do we DO with the data, and how does it aid our decision-making process to HELP the athletes?

         While digging into specific statistical tests and outputs is well beyond the scope of this blog, it is essential to note that nearly all the testing methods utilized at Acumen have validity and reliability data, typically published in peer-reviewed scientific journals. These tests include:

  • Specific isometric (static) strength tests performed via a strain gauge, or the ‘ForceFrame’
  • A variety of vertical jump measures with the ‘PUSH’ accelerometer or flight-time mats
  • Power tests on the Kieser pneumatic cable system.

We take great care in writing protocols so the tests are performed the same if a different coach is required to collect data in the future. Additionally, we are sure that our testers have substantial experience and perhaps certification anytime we use tests that have a high potential for human error (e.g., skinfold calipers). All of the above ensures that we are confident in our data, and any ‘improvements’ seen are likely accurate, not from error nor regular daily shifts in human performance.

         Once logical tests are chosen (little need to test vertical jump in curlers etc.), and testing has been completed, it is time to dig into the data. One of the first things I check for is correlations between the different tests we have chosen to use (Figure 1). By checking the between-test correlations, we can see if there is a highly correlated measurement with all or most of the other variables. While there is nothing obvious in the figure below, we recently tested several rodeo athletes and found that grip strength measured with a fixed versus dynamic arm were nearly perfectly correlated (with very narrow confidence intervals) with each other (Figure 2), with nearly no difference in the rank-order of each athlete. Therefore, we may consider removing one of the grip tests from future testing protocols.

Figure 1. Correlational matrix of hockey performance testing. 

Figure 2. Pearson’s correlation between static and dynamic grip strength. Each dot represents a single athlete. Dashed lines denote 95% confidence intervals.

Another often underestimated use of data is to improve buy-in from clients and athletes. As previously mentioned, we are often left with numbers that we have little outside information (journal articles, publicly available testing combine results, etc.) to compare. In this case, our strength and conditioning specialists can use their best judgment, based on the personalities of the athlete(s), and let the athletes know if they fall in the bottom, middle, or top ranks of their team (Figure 3), or similar population; this strategy is almost always successful as the athletes will work extra hard to improve their position, and/or fend off those chasing them.

Figure 3. Blinded inter-athlete comparison spreadsheet. Cells are conditionally formatted where white = 50% percentile, and green and red = max and min values, respectively.

In the case a coach does not believe direct comparison will be valuable, the data can still be used to track individual progress over time, allowing the athlete to see the fruits (or lack thereof) of their labour (Figure 4). Coaches can also use this information to re-evaluate their programs from a general or individual level. For example, the group or a team is making acceptable progress, but one or two athletes are regularly lagging. The coach can address the training programs or other variables (diet, stress, etc.). Alternatively, suppose progress is slow, or a team is generally regressing. In that case, the coach is alerted, and the training program, sports schedule, or numerous other variables (acute or chronic fatigue) may need to be examined for the team as a whole.

Figure 4. Example testing progression over time. Each dot represents a single athlete.

As always, and regardless of the topic, Acumen is constantly evolving, learning, and bettering our practice. Collecting, analyzing, and utilizing data is only part of the process, but as we continue to build and fine-tune our systems, all data collected will be used towards the ultimate goal of helping every client and athlete, regardless of their goals!