Rochelle Yang started marathon training two years ago, working her runs into her hectic schedule as a full-time pharmacy student at the University of Iowa. She trained year-round, putting up with snow, sleet, and regular sub-freezing temperatures. Sometimes, it was hard fitting in runs into her day. After a 12-hour day of classes, studying, and working a part-time job, she'd get home at 7 PM, have a quick snack, and then bang out somewhere between 7 and 13 miles, sometimes on less than 5 hours of sleep.
"I'll be honest," Yang wrote to me via email, "there were many days I hated running" (although she did add a smiley to the end of that sentence).
She dreamed of running in the Boston Marathon someday, but didn't think she'd hit the qualifying time in the near future. Her goal for the first marathon she ran, in June 2015, was simply to finish without stopping or walking. She finished in 3 hours 50 minutes, only 15 minutes shy of hitting the Boston qualifier time, commonly referred to as "BQ." For the next four months, she trained hard trying to shave those 15 minutes off her time.
At the IMT Des Moines Marathon in October 2015, Yang pushed herself, particularly in the last four miles, to hit BQ, which is 3 hours 35 minutes for her age group. "I remember thinking I would be okay with everything in my life going wrong for the next year if I could just beat the 3:35 pacer to the finish."
Yang crossed the finish line, exhausted and numb. Once she regained a modicum of wind, Yang called her parents, fighting to catch her breath and holding back tears at the same time, which, she remembers, made her sound like she was injured. Her parents asked if they needed to call an ambulance. Finally, she got the words out so they could understand.
"Dad, I just qualified for Boston!" She ran 3:32:51, or BQ-2:09.
Yang wasn't automatically invited to run in Boston even thought she ran a fast enough time to qualify. Every year, more runners apply for Boston with BQ times than the race has slots. So they take applications on a rolling basis, accepting the fastest runners first and working their way backwards. Usually, all the slots are filled somewhere in the final group—between BQ-5 and BQ—but nobody knows exactly where the cutoff will be until the announcements go out. For the application period that ended just before Yang ran in Des Moines, the cutoff was BQ-2:28, 19 seconds faster than Yang's time. It would be close.
Yang put in so much work to BQ that it never even crossed her mind that some people don't put in the work. Some people cheat.
Derek Murphy started Marathon Investigation as "kind of an afterthought." He followed the stories of notorious alleged marathon cheaters like Mike Rossi and Kip Litton on the LetsRun.com forums, but didn't post much. Still, he realized the reason they aroused suspicion and got caught was because they publicized their fabricated accomplishments, attracting attention. After following Rossi's case closely, Murphy wondered, "How many more people are like this guy but we don't know about because they don't put themselves out there?"
A business analyst in Cincinnati with a wife and two kids, Murphy downloads race results and looks for anomalies. For the Boston Marathon, bib numbers are assigned in sequential order, so the earlier you get accepted—and, therefore, the faster time you have—the lower the bib number. Based on this, Murphy can roughly predict what time that runner ought to run in Boston if they qualified honestly. If a runner is far off his or her predicted time, Murphy flags it.
From that list, Murphy looks at the mat times, which, using a digital chip in the race bibs, logs a runner's splits every time they cross a mat at certain mile markers. From here, Murphy can spot issues quickly. If Murphy finds several missed mats in conjunction with impossible splits, such as a runner increasing their pace significantly over the second half of the race, they probably cut the course.
All in all, it takes Murphy about 30 minutes from the time he gets the data set to when he has a list of people who almost certainly cheated. From there, he researches their qualifying times and races, looking at photos, split times, and communicating with race directors on individual cases.
These methods are not all that different from those employed by LetsRun forum sleuths. Those threads—hotly investigating claimed athletic feats—have been some of LetsRun's most popular, according to the site's co-founder Weldon Johnson. Rossi, for example, gained international fame after the principal at his kid's school sent him a letter about the child's unexcused absences surrounding the 2015 Boston Marathon, in which Rossi ran. Rossi posted a snarky reply to Facebook, touting his personal accomplishments in qualifying for Boston. "They watched their father overcome, injury, bad weather, the death of a loved one and many other obstacles to achieve an important personal goal." The Facebook post went viral and got him immense media attention.
But LetsRun's forums uncovered what they believed to be conclusive evidence that Rossi cut the course in his qualifying race, and that he should have never been allowed to run in Boston. Rossi has denied that he cheated in the race, and Lehigh Valley Marathon race officials did not disqualify him after the fact despite LetsRun's evidence.
More recently, a man named Robert Young tried set the record for fastest time running across America, but fell short amidst suspicions he was fabricating his efforts.
In the Rossi and Young situations, Johnson said, the threads were likely so popular because both runners insisted they were innocent. Young's feat was under suspicion in real time, so there were always new developments for the message board to discuss. Most of the other cheating threads tend to be shorter-lived. "But," Johnson added, "people are amazed at the sleuthing ability of some of our posters." Indeed, these threads involve hundreds if not thousands of intense runners with finely-tuned bullshit detectors, all doing their own smaller version of Murphy's marathon investigations.
Sometimes, commenters on Murphy's site or in the LetsRun forums wonder why someone would cheat in a marathon, the ultimate individual sport. But that's just one side of the question. After all, if running is such an individual sport, why would anyone spend so much time trying to catch them?
"For serious runners, race day is sacred," Matt Taylor, Founder and CEO of running apparel company Tracksmith and former head of global marketing for running at Puma told me over the phone. This is in contrast to what he called the "health and wellness movement" over the last decade or so, which encourages people to be active, raise money for charity, and jog in goofy costumes where everyone gets a finisher's medal. "I think that's rubbed serious runners a bit the wrong way because it's something we've always held as sacred."
So is the sleuthing and outing of cheaters a way to re-claim their territory? Taylor thought it was possible, but allowed that it's only part of the explanation. Particularly in the Rossi and Young cases, these guys were publicly bragging about their accomplishments, using social media to promote themselves. (He also emphasized that he doesn't much care what other people do and doesn't "want to be lumped in with people losing their shit over this.")
Social media and modern technology function as a double-edged sword. The tweets and Facebook posts people put up of finisher's medals they might not have earned taunt serious runners to out them. And the message boards, digital time mats, and race photos make it possible for it to happen in a matter of minutes.
But the main reason marathon cheats likely incite so much outrage in the running community is because cheating violates the very essence of the sport down to its philosophical core. As LetsRun's co-founder Johnson explained, "Running in its purest form is not a battle for external recognition but a battle within ourselves. Cheating changes this and is such an affront to the sport, that people spend a lot of time combating this. Many hard-core runners can't fathom why a runner would cheat in a race, so they go to great lengths to expose them."
This is, more or less, why Murphy spends many of his nights sorting through spreadsheets and looking at marathon photos of people he's never met. He also hopes it serves as a deterrent for anyone else who might course-cut their way to Boston. But there's an even simpler reason why he does it.
Murphy has run in 10 marathons himself, but his last one was eight years ago. Then, he had kids. He doesn't have the time or energy for training anymore. But, he still cares about running and considers himself a part of the community. Catching cheaters is his way of contributing now that he no longer participates. Also, spreadsheets and data analysis are his thing. He's good at it.
Murphy first wrote about Jeff and Sheri Donnelly in April. According to Murphy's research, Jeff ran the Boston Marathon four times: 2011, 2012, 2014, and 2015. Jeff qualified for 2015 with a time of 3:27:49 at the Desert Classic in Arizona, but ran a 4:28:04 in Boston, a full hour slower. Or, to put it another way, he qualified with a 7:55-per-mile pace but ran Boston at a 10:13 pace.
The Desert Classic didn't have mats, but Murphy looked up photos from the race, which occurred at the same time as a half-marathon. Both the half and full marathons were out-and-back loops. Jeff was photographed racing next to a 2 hour 30 minute half-marathoner towards the middle of the race, which doesn't make any sense. Jeff clocked a full marathon in only an additional hour.
His wife, Sheri, ran in the same race. She signed up for the full marathon, but the organizers dropped her to the half-marathoners because her time—identical to her husband's—would have meant she was the first female to finish, but they knew she wasn't.
In late 2015, Jeff and Sheri ran the Surfer's Point Marathon in Santa Barbara, CA, a two-loop marathon rife with course-cutting opportunities. Murphy found that Jeff wore a blue shirt over a green shirt for part of the race, but switched them at some point; a typical move for course-cutters in order not to be recognized by other runners.
Both Jeff and Sheri—who was also busted for course-cutting at the 2015 Big Wildlife Run in Anchorage, AK—ran BQ times at Surfer's Point and, according to Murphy's research, registered for Boston 2017, the same race Yang worked so hard to qualify for. Several calls to the Donnelly home were not returned.
Murphy also suspects Sheri ran the 2014 Boston Marathon with her husband despite the fact that she didn't qualify by using his bib to forge her own, changing his number #5895 to #5855. As Murphy points out, bib #5855 for the 2014 Boston Marathon was a 42 year old male who was photographed throughout the race.
When asked about Murphy's efforts specific to the Boston Marathon, a spokesman for the Boston Athletic Association, or BAA, told VICE Sports: "We rely on the race organizers and timing systems they employ to produce true and accurate results, and we also rely on the honesty and integrity of 99.99 percent of competitors who compete fairly and strive for their own personal records."
Murphy couldn't offer a precise number, but he estimates roughly three to four percent of Boston Marathon runners qualify by cheating. If accurate, this would mean roughly 600 runners out of 30,000 may have lied their way to the marathon.
"For the relatively tiny minority of participants who seek to gain unfair advantages," the BAA said, "there is sometimes no better method of rule enforcement than from witness accounts and reporting of fellow participants who also believe in a clean sport. And when such transgressions are reported, we trust race directors to fully investigate the matter, and adjudicate their race results if necessary."
After putting together his dossier on the Donnellys, Murphy notified the Surfer's Point Marathon race organizer, Bill Escobar, who reviewed the evidence and agreed with Murphy's findings. In September, Escobar disqualified the Donnellys, and notified the BAA that these runners were no longer eligible for Boston.
That same week, Yang didn't sleep well. She had applied for Boston and was anxiously waiting to hear back. She knew her chances were slim and tried her best to temper expectations, knowing -2:09 wouldn't have made the cut for the previous year's Marathon.
On Wednesday, September 28, she got the email. She was the only one in the office, so nobody heard her shout.
She didn't know it then, but the 2017 Boston Marathon cutoff time for the female 18-34 age group was BQ-2:09, precisely Yang's time. She straddled the cutoff line, but managed to fall just inside. Yang may have been the last person accepted for Boston 2017.
That day, another runner, via Reddit, brought Murphy's website to her attention. The Donnellys weren't the only runners disqualified thanks to Murphy's work. Yang concluded that, if it wasn't for him, she might not be in Boston.
Yang wrote Murphy an email, which he then posted to the site. She explained how she was just barely accepted, calling it "the biggest stroke of luck I may ever experience in my life. And I know I would not have gotten in if it weren't for the investigative work you do!
"You help get illegitimate runners kicked out of Boston so those spots can be filled by runners who have earned them. The time that you've put in, the work that you do that maybe feels thankless sometimes--has allowed a young runner to be able to complete her dream and run in her first Boston Marathon. And for that, I cannot thank you enough."
It's hard to say whether this conclusion is true. As Murphy wrote in the post, his work doesn't free up additional slots. A BAA spokesperson confirmed that disqualified runners don't have their spots filled. Instead, fewer people run.
But, Murphy also uncovered several dozen runners in the year leading up to Boston registration who, without his efforts, could have registered with falsified BQ times. Would those runners have pushed the cutoff to BQ-2:08? It's impossible to know.
This ambiguity opens the door to different interpretations, and the way runners perceive this dynamic is indicative of how they perceive running itself. You can believe Murphy helped Yang make the cutoff and that cheaters are an affront to everything the sport stands for. But you can also believe Yang will be running in Boston, not because of a business analyst in Cincinnati, but because of herself. She's running in Boston because, at the IMT Des Moines Marathon, there were 49 turns, six loops, and 175 feet in elevation change. If she had run a single second slower—slowed up to avoid a crowd before pushing through, taken the long angle around a turn, an extra half-beat when grabbing a water at the water station—she wouldn't be going to Boston, Marathon Investigation or not.
Want to read more stories like this from VICE Sports? Subscribe to our daily newsletter.