The Michigan Engineer News Center

Has the Olympics changed how it measures false-starts in track?

A Q&A with a biomechanics expert who has researched reaction times| Medium Read
EnlargeA runner kneels down to prepare to begin the race
IMAGE:  Olympic reaction times have not become faster due to improved training but, instead, likely represent changes in the computer algorithm used by Swiss Timing. Photo courtesy of Getty Images

In 2011, James Ashton-Miller, a Michigan Engineer, helped reveal that Olympic starting-line technology created a different experience for male and female sprinters. It did not accurately detect false starts by women. His latest work provides insights into what may, or may not, have happened since.

Ashton-Miller is the Albert Schultz Collegiate Research Professor & Distinguished Research Scientist in mechanical engineering and biomedical engineering.

A paper you published in 2011 analyzed data compiled during the 2008 Summer Olympics in Beijing from track and field events. Those data indicated that female sprinters had slower reaction times than their male counterparts. What troubled you about your own findings in that paper?


Ashton-Miller:
It happens that, as part of a previous study on the biomechanics of falls, our team measured lower extremity reaction times in a lab. Those data unequivocally showed that for small forces, women reacted faster than men. But if a much larger force was required to register reaction, then it took women longer than men to reach that threshold.

Now, both plates on the starting blocks at the Olympics are equipped with a force sensor. When the athlete adopts their starting position, one foot goes against each plate and, in order to accelerate forward, the athlete has to push off against these instrumented blocks. If the blocks register a certain level of force before 100 milliseconds has elapsed after the starting gun, it is deemed a false start.

Our analysis showing slower reaction times by women in Beijing surprised us, but initially we accepted the finding. Eventually, based on our lab study, we speculated that Omega sister company Swiss Timing (the official Olympic timekeeper) must have used the same force level on the starting blocks for men and women in determining whether a false start had taken place. That would mean that the sex difference in reaction time was likely an artifact.

Once you realized what might be causing the difference in women versus men’s reaction times, what step did you take?


Ashton-Miller:
Prior to publishing that article, we reached out to Omega to ask what force threshold they used on their starting blocks and whether it was the same for men and women. They replied that information was proprietary.

Then, using results from our earlier lab study which measured lower extremity reactions, we calculated that by reducing the force threshold for women at the Olympics by 21 percent, the reaction gap should disappear. We published that suggestion in the 2011 paper hoping that it might spur Omega to make the competition more equitable for women, since with their current system a women could false start by approximately 20 milliseconds and not be caught. That would obviously be unfair to other women in the race.

So what questions did your team tackle next?


Ashton-Miller:
Once we knew about this bias in measuring women’s reaction times, we wanted to see if the results from the 2008 Olympics were consistent with other Olympic Games before and after.

We also wanted to see if Swiss Timing had continued to use the same reaction time calculation in subsequent Olympic games. So I asked graduate student Payam Mirshams Shahshahani to compare reaction data from the ‘04, ‘08, ‘12, ‘16 Summer Games using a different kind of statistical analysis that could take advantage of a repeated measures analysis.

What did you find?


Ashton-Miller:
Our findings showed the same difference – women showing slower reaction times than men – in 2004 and 2008. But in 2012, a year after our first paper appeared, the difference between the sexes disappeared. Women’s reaction times had gotten significantly faster with men’s times remaining consistent.

In our most recent paper, we argue that these Olympic reaction times have not become faster due to improved training but, instead, likely represent changes in the computer algorithm used by Swiss Timing.

We conclude that it is important to set the force thresholds for women lower than men in order for a fair women’s competition. For if a female sprinter knew that the men’s values were being used, she could intentionally false start by up to 20 milliseconds without being flagged by the electronic timing mechanism.

What ramifications does your analysis have outside the world of track and field?


Ashton-Miller:
Acknowledging that, depending on the force involved, there can be a sex difference in reaction time is also important for designing interfaces by which humans interact with machines: Women should not be disadvantaged in implementing time critical decisions because of the need for high force thresholds. This may be particularly important for the elderly whose longer reaction times mean they have less time to mount their response in time critical tasks.

The new study was done in collaboration with Payam Mirshams Shahshahani, a doctoral student in mechanical engineering; David Lipps, a professor of kinesiology and biomedical engineering; and Andrzej Galecki, a research professor in gerontology and biostatistics. It is published in the journal PLOS One, titled, “On the apparent decrease in Olympic sprinter times.”

Our Future, Your Impact
Female scientist in the lab

We use engineering to improve the quality of life – for the common good. Solving societal problems requires leadership and collaboration. Join us. Turn your inspiration into action

A runner kneels down to prepare to begin the race
Portrait of Jim Lynch.

Contact

James Lynch
Research News & Feature Writer

Michigan Engineering
Communications & Marketing

(734) 763-1652

3214 SI-North

Researchers
Hurricane Florence approaching the east coast. Photo courtesy of the NOAA

Hurricane Florence: U-M researchers forecast impacts

More than 2 million people could lose power, and flooding is the major concern for several reasons. | Medium Read