A New Way to Measure Change of Direction?

Tyreek Hill is an amazing athlete, perhaps one of the greatest to ever grace the NFL. His breakaway speed is undeniable—he’s called “the Cheetah” for good reason—and to top it off, he is shifty as all get-out.

When Hill has the ball, he might go left, right, forward, backward, upfield, downfield, up, down, diagonal, spinning, spiraling, rectangle-ing, into the past, back to the future—he may even disappear into a portal and reappear 45 yards downfield. I’m not really sure.

The point is, anything can happen when Hill has the ball, and the reason is because no matter what he does—whether he hits the jets and runs straight or decides to make a cut—it happens at light speed.

That’s what makes a player dangerous: the ability to do everything fast.

Aside from raw speed, shiftiness is one of the skill sets that distinguishes great from average field-sport athletes. Barry Sanders, for instance, was not known for his stellar speed. Although he was certainly fast, he is most famous for his elusiveness. As a result, he’s fourth all-time in NFL rush yards, despite a relatively short career.

In our world as sport scientists and strength and conditioning coaches, we call shiftiness change of direction (CoD, not to be confused with “Call of Duty”). Because changing direction matters, measuring it is of significance. If we can somehow quantify how shifty Tyreek Hill is, and if we could have done so with Barry Sanders, we’d have metrics by which to compare them to other athletes.

Imagine how great it would be to say, “X prospect is just as shifty as Barry Sanders,” with the same certainty in which we can say, “that guy is just as fast as Deion Sanders was coming out of college; they both ran a 4.21.”

Change of direction testing does exist, but it isn’t great. What really matters—how fast an athlete, well, changes direction—isn’t captured very well with the common battery of tests, including the one at the NFL Combine (more on this later).

Thinking about this problem prompted me to start with a simple question: What is change of direction? That led to an idea on how to better measure it.

First, though, a quick journey through definitions.

Change of Direction vs. Agility

I would be remiss to not mention the difference between change of direction and agility. But, to be honest, many people make a bigger deal out of this than need be. The different operational definitions are helpful to the advanced practitioner, but to the everyday person, the terms are interchangeable.

Agility, by definition, requires a response to a stimulus. Imagine a running back deciding which way to cut as he accepts the handoff in a zone-blocking scheme. He reads the field, sees the movement of the other players (the stimulus), and then reacts and decides which way to cut. There is a change of direction involved (the cut), but it is in response to something. It is a decision.

Unlike agility, change of direction does not include a response to a stimulus.

In a practical sense, change of direction refers to testing, like the 5-0-5 and the 5-10-5, while agility refers to more game-like instances in which reactiveness and decision-making precede movement.

Thus, change of direction is a component of agility, whereas agility includes the cognitive decision-making component that change of direction does not. Agility is the complete skill; change of direction is the physical ability portion.

The Problem with Change of Direction Testing

Dr. Sophia Nimphius of Edith Cowan University has brilliantly highlighted that most change of direction testing is really just speed testing. Her research demonstrated that faster athletes have better scores because they get from Point A to Point B faster, not necessarily because they change direction faster.

She and her team evaluated the 5-0-5 test.

Let’s say Athlete A runs the initial 15-meter segment in 1.5 seconds. The actual change of direction time (more to come on defining this later) is 1.0 seconds, and the final 5-meter sprint is .8 seconds. The total time of the test is thus 3.3 seconds.

Say Athlete B runs the initial 15-meter segment in 1.8 seconds, changes direction in 1.0 seconds, and runs the final 5 meters in 0.9 seconds, for a total time of 3.7 seconds.

Athlete A has a much better time, yet the two have identical change of direction performances. The difference is one athlete runs faster than the other.

You can see how this presents problems and does not accurately capture change of direction capabilities.

Dr. Nimphius suggested using the Change of Direction Deficit to more accurately measure and report change of direction ability. The Change of Direction Deficit is simply 5-0-5 time minus 15-meter sprint time. This measure does capture change of direction ability better than simple 5-0-5 time, providing practitioners with a more valid, widely available approach. However, it still does not quite seem to capture the act of changing direction itself.

Forster et al. recently published a paper in which they put a timing gate 1 meter from the change of direction line in the pro-agility test. The athletes sprint past the gate, touch the line, and sprint back through the gate. This does seem to capture change of direction ability fairly well, and it may be a measure worth considering in future research. Even still, this is a proxy measurement for the act of changing direction, not a direct measurement of it.

These two improved methods aside, the problem with traditional change of direction testing is it doesn’t actually measure change of direction very well.

What Is Change of Direction, Anyway?

This begs the question: What is change of direction?

I think of change of direction as the rate of deceleration and reacceleration of the center of mass (CoM) in a different direction. It is essentially the management and redirection of momentum.

At first thought, a tempting way to measure change of direction performance is analysis of the ground contact time during the CoD step itself. On second thought, however, the pitfalls of doing such are enormous. GCT in this sense doesn’t tell us anything other than how long the foot was on the ground. What we need to know is how quickly the center of mass changed direction.

A New Way to Measure Change of Direction?

I’ll be honest—I did not do a full literature review to see if this has already been done or discussed. Frankly, my goal here is to put this thought into the performance universe, so I’m not going to spend hours digging through research to verify if this has already been proposed. If it has, I certainly think not many know about it.

In any case, below is a strategy coaches can use to quantify change of direction performance:

(Entrance velocity + exit velocity)/GCT

Entrance velocity: CoM velocity at touchdown

Exit velocity: CoM velocity at toe-off

Ground contact time: total time foot is on ground

This formula accounts for the speed at which an athlete enters the cut, which is important, as cutting from a jog is much easier and quicker than from a sprint. Moreover, faster entrance speeds more accurately resemble game-time performance than slower entrance speeds, assuming athletes move fast in games. Athletes are “penalized” with a lower score if they enter the cut slowly.

The formula also accounts for exit velocity, which has obvious sports performance implications.

Dividing by ground contact time, inspired by RSI, adds the time element. Speeds coming into and out of cuts are important, but the rate at which those speeds are created is a critical factor in sport and change of direction performance.

Thus, this formula allows for the analysis of the act of changing direction itself and excludes the noise of sprint times into and out of the direction change.

For simplicity’s sake, assuming this test does not already exist and thus does not have a name, let’s call it the Change of Direction Index (CoDI):

The CoDi in Action

I recognize that the technology necessary to collect the data required is not commonly available. Not everyone has a system capable of measuring center of mass velocity, let alone the ability to sync it with the required tech to determine ground contact time (force plates, a high-speed camera system, or a contact grid). Perhaps radar could work if you could be certain the center of mass is the point being measured, but that seems tough to do in a change of direction task.

I am fortunate to work at a facility that does have the necessary tech, so I played around and collected some data. Our IMU system and a high-speed camera synched beautifully—thanks to the excellent integration software Noraxon makes, it was pretty simple. The athlete completed a 5-0-5 test (two total trials—one turning to the left, one turning to the right), and I noted touchdown and toe-off of the cutting step with the camera.

Thanks to Lindsey, the main tech support engineer at Noraxon (who is always willing to help), I was able to extrapolate center of mass position and velocity. (Read: Lindsey was able to remotely access our computer and run a few fancy algorithms while I watched.)

Video 1. A screen recording of the Noraxon software showing touchdown and toe-off of a 5-0-5 cut step. The line graph on the far left shows CoM velocity, and the skeleton on the right is a digital representation of the IMU data and mirrors the athlete’s body movement.

In any case, below is the data.

Take the data with a grain of salt—it was one athlete doing just one trial per foot. Nonetheless, it is interesting to note the scores and the difference in scores. The data would indicate an asymmetrical performance, with the right foot outperforming the left.

Just as reactive strength index (RSI) can be applied to many types of jumps, so to the CoDI can be applied to several types of direction changes. A 5-0-5, the pro-agility (5-10-5), a simple standardized 45-degree or 90-degree cut with a 10-meter run in, or even actual game-time cuts. I’m sure NFL Next Gen stats can generate angle of cut and ground contact time, and we know they track player speed.

And, as you see in the data above, the CoDI may also be used to measure asymmetry in performance when cutting left versus right.

I invite our community of coaches and critical thinkers to tear this construct apart. Certainly, this is not without flaw. But, at the risk of being taken for presumptuous, I do believe there could be a place for this measure in the world of sports performance.

Lastly, I invite practitioners to apply the CoDI with their athletic populations and share the results. Tag me on Twitter or otherwise make your data public, if you wish.

At the very least, let’s continue to think, push boundaries, reevaluate approaches, search for improved methods of testing and measuring that which we truly desire, and discard that which becomes obsolete.

Since you’re here…
…we have a small favor to ask. More people are reading SimpliFaster than ever, and each week we bring you compelling content from coaches, sport scientists, and physiotherapists who are devoted to building better athletes. Please take a moment to share the articles on social media, engage the authors with questions and comments below, and link to articles when appropriate if you have a blog or participate on forums of related topics. — SF

References

Forster, J.W., Uthoff, A.M., Rumpf, M.C., and Cronin, J.B. “Advancing the pro-agility test to provide better change of direction speed diagnostics.” The Journal of Sport and Exercise Science. 2021;5(2):101-106.

Nimphius, S., Geib, G., Spiteri, T., and Carlisle, D. “Change of direction deficit measurement in Division I American football players.” Journal of Australian Strength and Conditioning. 2013;(S2):115-117.

Nimphius, S., Callaghan, S.J., Spiteri, T., and Lockie, R.G. “Change of direction deficit: A more isolated measure of change of direction performance than total 505 time.” The Journal of Strength and Conditioning Research. 2016;30:3024-3032.

Nimphius, S., Callaghan, S.J., Bezodis, N.E., and Lockie, R.G. “Change of direction and agility tests: Challenging our current measures of performance.” Strength & Conditioning Journal. 2018;40(1):26-38.