For example, I received a Garmin Forerunner 620 for Christmas. The Garmin 620 is unique among Garmin watches in that it adds in the ability to measure a slew of new variables as part of the new “running dynamics” suite collected by the HRM-Run monitor that pairs with the watch. I’ve only just started playing with the 620 (loving it so far, hopefully will get a first impressions post up soon), but I’m not yet sure what to make of the data it provides me.

Here’s an example data summary from a run from a few days ago:

First off, there’s no way my max heart rate hit 188bpm on that run – it was a slow run with my dog on very icy and wet surfaces – sometimes we need to be aware that these devices make mistakes.

In addition to standard heart rate data, the watch spits out data on Training Effect, Cadence, Vertical Oscillation, and Ground Contact Time.

Very cool stuff, but my questions are:

A) How accurate are these measures?
B) What is the practical significance of this data for a runner like me?
C) How do the data vary with things like speed, terrain, footwear, etc.?
D) What are the normal ranges observed for these data (is there even a “normal”)?
E) If I am way out of the normal range for any of them, is there any compelling benefit to making a change?
D) If I need to make a change, what’s the best way to do so?
F) Is there any risk associated with making a change?

Don’t get me wrong, I’m a data geek, and being able to play with this stuff excites me. One of my goals for the coming weeks is to experiment with the watch and see how the numbers change with various conditions. I also hope to dig into the scientific literature a bit on a few of these variables and see what I can come up with that might be of practical value to everyday runners. But right now they mostly represent a curiosity (maybe with the exception of cadence, which I have a pretty good handle on).

The issue I have with some of the tech coming out is that complex biomechanical information may get overly simplified to the point where it loses meaning, or may actually do harm. To give an example, Techcrunch recently posted and article on the topic of health tech, and in it they feature the new Sensoria Smart Socks. The Sensoria socks contain textile sensors that can measure foot contact location, and they are thus able to give real-time feedback on foot strike (see Sensoria app image at left).

As someone who works in a clinic doing gait analysis, I can see value in a gadget like this (if it works well) for a client a who has a very good reason to consider altering their foot strike and needs something to help keep them on task.

But, my worry is when you see a line like this one in the Techcrunch article: “The Sensoria smart sock correctly diagnosed that I make the runner’s rookie mistake of heel striking, leading to a workout-stopping knee pain.”

I’m not sure I’d call heel striking a “runner’s rookie mistake,” nor do I think we have any strong evidence to say across the board that heel striking is bad or causes knee pain. What’s more, moving from a heel strike to a forefoot strike poses some amount of risk due to the changes in lower extremity loading that occur. Given that runners often tend to be overly enthusiastic (to put it nicely!) with things that promise performance improvement or injury prevention, the opposite result may be just as likely: The Sensoria tells you you’re a heel striker, you actively try to change that quickly because you think heel striking is evil, and you wind up breaking your foot or shredding your calves.

The point here is that if the Sensoria works as indicated, it could be a very cool tool under the right circumstances, but my fear is that in reality it will be a tool used needlessly to get people to change form who would be better off just staying with what works for them.

So, to end this rant, I’ll say that I’m as excited as anyone about the tech on the way. It allows us to measure things (or at least estimate things since accuracy may not be perfect) affordably that could previously only have been measured in a high-tech lab. But, lets not get overly enthusiastic about the stuff we measure until we have a good understanding of what it is telling us and what the practical meaning of such data is.

I have a fairly long history of working with high-speed (i.e., slow-motion) video. As a graduate student, I did my dissertation research in a lab that specialized in locomotor and feeding biomechanics of animals, and to do these types of studies, the ability to film at very high frame rates (200 frames/second and up) was a must. However, the NAC cameras we used in those days priced easily in the 5-digit range (possibly even 6-digits), making the technology far too expensive for the average person wanting to play around with slow-motion video. Technology has come a long way in recent years, and cameras that can record at such high-frame rates have finally dropped into a price range where they are affordable to people not working in grant-funded academic research labs.

Let me start by describing what exactly I mean when I refer to high-speed video. Most home digital video cameras shoot at a frame rate of ~30 frames/second (fps), which means that they essentially are taking 30 still photographs per second, then playing them back in in sequence such that they appear to form a seamless video. It turns out that 30 fps is sufficient enough to prevent us from being able to detect the gaps between images with the naked eye when they are played back. However, if you were to slow down video taken with a regular camera (e.g., in a program like Virtualdub), you would notice a distinct “jumpiness” due to the gaps between images – the video would be far from smooth, and this is why you need a more capable camera when you wish to film fast action and play it back in slow motion.

My series of slow-motion running videos have sparked a lot of commentary on this blog, as well as a decent number of direct e-mails to me from readers (for an example, see this post looking at the effects of shoe type on footstrike). Most of these clips have been shot with a Casio Exilim EX-F1 digital camera. The EX-F1 is an amazing piece of electronics, and in terms of high-speed video capability, it does for about $1000 what historically could be done only by far more expensive cameras. The Casio Exilim EX-F1 does a lot more than just high-speed video (e.g., HD video at 1080i, 60 high-res stills in one second in high-speed burst mode, plus a lot more – you can read the full overview on Casio’s product page for the EX-F1), but that’s it’s primary use for me, it does it very well, and that’s what I’ll focus on here.

The Casio Exilim EX-F1 can shoot high-speed video at 300, 600, or 1200 frames/second. As you increase filming speed, however, you get a loss of resolution as follow: 512 x 384 (300 fps), 432 x 192 (600 fps), 336 x 96 (1200 fps). That being said, the 512 x 384 resolution is more than suitable for my purposes, both personal and academic. The EX-F1 also has a variable 30-300 fps mode that allows you to toggle the movie recording speed between 30 fps and 300 fps on the fly. This would allow you to switch between real-time and slow-motion as the need arises (think capturing a baseball player sliding into a base). Below is a sample slow-motion video of Ryan Hall running in the 2010 Boston Marathon recorded by two of my students at 300 fps on the EX-F1:

Ryan Hall – Boston Marathon 2010 from Runblogger on Vimeo.
Video clip of Ryan Hall running near mile 17 in the 2010 Boston Marathon. Filmed at 300 fps with a Casio Exilim EX-F1 digital camera. Video courtesy of https://runblogger.com.

You need a lot of light to get image clarity/quality with high-speed video cameras, and the EX-F1 is no exception. While shooting outside usually gives good results, I have to use halogen spotlights on subjects in my lab to get crisp images with this camera. In fact, the 1200 fps setting requires extremely well-lit conditions (and I mean very, very bright) to get a usable image. Below is an an indoor video of me running on a treadmill in my lab shot at 300 fps with the EX-F1 (be aware that there is some loss of quality when uploading to Vimeo):

Brooks Launch from Runblogger on Vimeo.
Slow motion video of treadmill running in Brooks Launch – mild heel strike here. Video shot at 300 frames-per-second with a Casio Exilim EX-F1 camera. Courtesy of https://runblogger.com/.

So, my experience so far with the Casio Exilim EX-F1 has been extremely positive, and for a small-college researcher like myself, it’s an affordable option that allows me to still do many of the things I did in graduate school at a large research institution. I’d say that If you have money to burn, or are in an academic or competitive athletic environment (e.g., biomechanics research), the Casio Exilim EX-F1 is a great choice for affordable high-speed video. As time allows, I’ll try to add some additional videos at 600 and 1200 fps for you to view, as well as one in HD.