On the warm and sunny morning of Sunday 26th May Jacob Puzey ran a largely solo effort to win the Calgary Marathon 50k race in a time of 3 hours and 7 minutes. During the race Jacob was wearing a tiny set of sensors that were measuring his running gait from the gun to the tape. In this short article I’ll unpack the data for you and give you a rare insight into the running gait of an elite ultra marathon runner working hard in a race.

First here are the high level numbers related to his race. Jacob ran his 3:07 50k at an average pace of 3 minutes and 42 seconds per kilometer. His average step rate was 173 steps per minute, his average stride length was 3.05m, his average power was 398W and overall for the race he took approximately 32,000 steps.

Let’s begin by looking at what was happening during Jacob’s race at his feet and ankles. The shoeprint plot below defines his underfoot foot strike and toe off locations for his left and right feet. His initial foot strike is highlighted in the orange and red data. Meanwhile his toe off locations are shown by the green points. We can see that there is a small but marked asymmetry between his left and right sides with his right foot data being distributed slightly more rearward than his left. However there is great symmetry between the two feet when looking at the data distribution laterally from the inside to the outside of the feet.

Digging in a little deeper to the metrics we can see the difference in foot strike location quantified in numbers with the right side registering a smaller number or more rearward location (0=max heel and 16=max toe). Measured foot pronation was also a little asymmetric, indicating that the left foot was rolling inward more on average than the right. Given his fast running pace his pronation velocity numbers were pretty much as expected and relatively well balanced from left to right.

Using several other gait metrics we can see how Jacob’s legs were working as a whole and get an indirect picture of how his hips were functioning and controlling his leg movement. The results displayed above are all an average for the entire race and all show some small but significant asymmetries between the left and right side.

On average he spent a little more time on the ground on his left side and hence had a lower flight ratio and was less efficient on the left. His impact acceleration was higher on the right whilst his braking acceleration average was lower on the right. And mechanically his right side recorded higher average spring stiffness and vertical ground reaction force rate.

So what does this all mean? Firstly as mentioned above his right side was working more efficiently. More than that it was it was striking the ground harder, with faster loading and more mechanical rigidity. In short his left side just wasn’t working well. Detailed analysis of these numbers on a kilometer by kilometer basis indicates that this was the case for pretty much the entire race. The plot below highlights the difference between his left and right flight ratios from start to finish.

Plot of Jacob’s left and right side flight ratios for the entire race. (Flight ratio is the ratio of time spent in the air versus on the ground. Generally speaking larger flight ratios indicate more efficient running).

After careful analysis of the entire raw dataset I was able to tell Jacob that it appeared that his hips were performing differently and his left hip wasn’t functioning as well as his right from the start of the race. This would have affected his left side efficiency in how he controlled the movement of the whole leg and how he was striking the ground. It was also possible that his pronation numbers were also in part affected by his foot strike.

Here is what Jacob had to say after the race about how he felt his body performed as he was running:

My left knee began bothering me a few months ago. I attributed the pain to tightness in my hips and IT Band which was likely a result of the cold weather. I also felt like it had to do with the hard surfaces and lack of inventory in my go-to Altra Running shoes. I didn’t mention this to anyone leading up to the race. I just tried to keep it all together through the race and then I’d take a break and heal up.

I started race morning with a heat pad on my lower back, hamstrings, and calves. I went through a brief yoga routine and some foam and lacrosse ball rolling to loosen things up in my lower back, glutes, hamstrings, and calves. Then I went through a light warm-up jog with some dynamic stretching – mostly hip mobility and running drills. I was feeling really tight and stiff through the whole process.

The race started with a slight descent in the first few kilometres. Generally, I love any chance to open up my stride and run downhill, but on race day even the slightest descent caused piercing pain to run up my left leg from my feet, through the knee, and into my hips. I tried my best to run as relaxed as possible at every opportunity. The intensity of the pain ebbed and flowed, but it was ever-present.

I was beyond relieved when I crossed the line.

As this race case study shows it is possible to pull apart the gait of a runner in an event and with a highly skilled eye, be able to identify what is and isn’t working well for the runner in terms of their movement pattern. With repeated monitoring over a runner over time it’s even possible to go beyond this and use a baseline dataset to check for negative trends that indicate that the runner needs qualified help.

Huge thanks to Jacob Puzey of Peak Run Performance for his kind participation in this data collection case study. Data sets like this are helping us to constantly refine the technology as well as our methods for interpreting the data, so that we can help more runners out there to stay injury free and running well.

Malc Kent is a professional running gait analyst based in Calgary. He performs both onsite clinical gait analysis and remote gait analysis on runners locally and all over the world every week through his company Runfisx. He has been at the forefront of outdoor and remote analysis methods for the past 8 years and is the official global partner and product expert for RunScribe running technology.