Simple Sabermetrics has released a myriad of videos and blogs covering tracking technologies that have made their way into baseball over the last few years. These advances in technology have unlocked new ways for the community to analyze baseball and provide athletes with more sophisticated methods of development.
At the center of all the technology is the data that is collected. Baseball data is collected in many different ways, from pitch flight and batted ball flight to swing path, biomechanics and more. We’ve written about how to use data to inform your decisions and the value of being a data-driven program, as well as published videos that have introduced the basics of each technology. This includes Trackman, Rapsodo Pitching and Hitting, Blast Motion, and more.
So, we’ve covered the basics and importance of the data that these devices track, but we’ve never discussed the value of pairing the data across more than one system. Because each of these tracking devices are unique in their own space, they have limitations. That’s where we must get creative with the tools at our disposal. Pairing one tracking technology with another expands the possibilities to extract the most out of the data being collected.
As we move forward in this new era of technology and analytics, it’s certainly possible that some of these inefficiencies shrink through new inventions and upgraded hardware or software. For example, the new pitch-tracking system at the MLB-level, Hawk-Eye, is able to track data points beyond just ball flight, such as player movement, but the general public does not have access to such rich data. Hawk-Eye has yet to fully infiltrate every MiLB stadium, so it may be years until it makes its way down to amateur baseball.
Let’s consider a few reasons why pairing one system’s data with another provides value beyond belief to your organization.
Combining Swing and Hit Data
Blast Motion and Diamond Kinetics are the two most common bat sensors on the market. If you aren’t familiar with bat sensors, this video here is a great start. These sensors are able to track metrics such as bat speed, attack angle, and on-plane efficiency. In a vacuum, these metrics are valuable to identifying swing deficiencies, but sometimes require additional data to better inform a decision.
Although these two companies have released versions of their product to measure batted ball data, these methods of estimating exit velocity or launch angle are best left to technology specifically designed to track those metrics, such as HitTrax or Rapsodo Hitting. Both of these systems track the flight of the ball after contact has been made.
An athlete hitting with a bat sensor while on a HitTrax or Rapsodo Hitting unit is the most straightforward method of pairing tracking technologies to acquire two times the data points. For example, we can measure an athlete’s average attack angle by part of the strike zone. Below, we can see that this left-handed hitter maintains a consistent attack angle throughout the zone, except the up-and-in area where it decreases by a few degrees.
We know it’s best to take an individualized approach to each hitter’s development plans. By using the paired swing and hit data, we can identify where each athlete is maximizing their strengths, whether that be your desired Blast Motion metric in each area of the strike zone or in which range of a Blast Motion metric the athlete is hitting the ball the hardest.
The heavy lifting on the back-end of this method is deciding exactly how to pair each observation in both datasets. Each piece of technology will generate their own unique ID for each observed swing or batted ball, so we must rely on pairing the instances with timestamps.
There’s no surprise that we’ve seen MiLB hitters begin to wear their bat sensors during games to pair with Trackman or Hawk-Eye batted ball data. While it seems that the pitching community has had better resources to develop athletes over the years, the hitting community can utilize these pairing techniques to advance the methods used to develop their athletes.
Validation, Experiments, and Models
Performing validation tests on the technology you own doesn’t directly impact the development of an athlete, but it provides you the comfort of better understanding the data you collect.
So, what exactly is validation and how do we determine which system we use as a baseline of the truth? Unfortunately, the latter question does not have a definite answer. Each tracking technology has their strengths and weaknesses that we need to keep in mind, but we can use validation methods to find evidence of specific areas of interest.
Validation mainly entails calculating the estimated differences in measurements between systems, but it can also include analyzing the system’s performance under various anticipated operating scenarios. Optical pitch-tracking devices can often suffer from sudden changes in lighting outdoors or indoors. Various systems could capture ball flight or swing paths less frequently for a certain type of athlete (i.e. handedness, height, etc.).
For a full validation analysis, Driveline Baseball published an article for Rapsodo’s 2.0 Pitching unit back in July of 2019. That piece can be found here.
With a group of people and some free time, you can experiment with multiple tracking technologies to perform your own validation. For readers that work for a college program with a stadium pitch-tracking unit and a mobile pitch-tracking device, set up these systems at the same time on the field and throw a handful of pitches. If one of your units is a radar system (e.g. Trackman) and the other is optical (e.g. Rapsodo), what do you notice? Are the movement variables and spin directions different? They should be. There shouldn’t be a tremendous gap between the two system’s other measured variables, but it is an interesting exploration nonetheless.
A paired data set is a rich resource that has use-cases beyond validating each system. Rapsodo measures variables that Trackman cannot, simply due to the nature of the system. With a healthy sample size, you now have a training sample of Trackman data with Rapsodo spin efficiency and gyro degree. Building a model to predict these two measures strictly using the variables that are measured by Trackman becomes a possibility. Pending an accurate model, Trackman no longer needs its partner, Rapsodo, to measure spin efficiency in games. Besides, you’ll find no luck setting up a Rapsodo on-field in the middle of a regular season game.
Qualitative Analysis with High-Speed Video
High-speed video is another form of data that is paired well with quantitative data to gather vastly superior insights. It’s a creative blend of qualitative and quantitative analysis. Without thinking about it, capturing Edgertronic video during Rapsodo bullpen sessions qualifies under our paired data umbrella. The pitch-tracking device measures dozens of data points on the flight of the ball, whereas the camera films the ball release and trajectory at thousands of frames per second.
A pitch-design session is not complete without the pairing of these two devices. The coach or trainer uses both resources together to make better-informed decisions. The slight change of grip or release motion is observed with the video and the subsequent change in pitch shape is detected through the pitch-tracking system. You can see an example of this combination in the tweet from Eric Jagers below.
A qualitative experiment that can be performed with video is validating newer ball flight metrics such as seam orientation, which is measured in Yakkertech’s stadium pitch-tracking unit. There are no other systems that track the seam orientation of the baseball as it is released, so we cannot quantitatively compare the differences in values to another measure. Instead, we can capture high-speed video to estimate the accuracy of these measures.
As these tracking technologies become more commonplace in professional and amateur baseball, we must find new ways to stay ahead of the curve. Many organizations have put themselves past others in the biomechanics realm, but there is a point the industry reaches where most everyone has the same technology. The new competitive advantage belongs to those who find new methods of extracting the most value from their resources, which exponentially expands the number of insights possible to develop athletes.