The Run Down - The RunVermont Blog

FAQ #25 (2013): Results, and why it takes so long to make them official

by leandre on May 25, 2013 · 7 comments

in 2013 Marathon Updates,FAQ's

Welcome to Marathon May! Every day from now through the Marathon (and perhaps even a few days beyond) we’ll share a new race-related FAQ here on the blog, our Facebook page and via Twitter feed. Got a question you think we should cover? Post it here, on FB or on Twitter and we’ll get right back to you.

Q: Race Results: How come it takes so long for official results to come out? What are you guys doing?

We wrote this post for the first time last year in anticipation of the many questions we get about why results aren’t considered “official” immediately after the race.  Our technical director, Joe, revised the  post for this year.

One of the challenges of producing a race of this size is in making sure our results are correct. At the finish line, off our primary timing system, we produce results that you can look up at the MVP Health Care Results Tent in the north end of Waterfront Park. With over 4000 runners crossing the finish line, it’s nearly inevitable that something goes wrong for someone-a tag is defective or has become damaged during the race and fails to read, a relay team last leg runner does not wear the Last Leg bib, a runner covers their tag with a water bottle or wears it on their back-the list is long on why the system can fail and every system has its strengths and weaknesses. To protect our finish line operations we run multiple backup systems, some data based and some visual (both video and still pictures).

So then what?

Monday morning after the race our staff starts to review the results for anomalies-times that just don’t look right or runners who were missed on the primary timing system. Every year we get emails from runners who may have a problem with the results; either they are missing in the overall results, or maybe a split is missing or wrong. On the relay side we hear from teams who are listed in the incorrect division. While we want to get results up immediately, it’s most important that we get the results 100% correct, even if that takes a few days. At some parts of the race it’s not too hard to catch everyone on primary and backup timing systems, but at other times the density at the finish line makes it impossible. At our peak finish times, 11:30-12:45, we average greater than 1 finisher every 2 seconds, for the entire 75 minute range! In fact, most years around 60% of our finishers come in during the hour of 11:30-12:30. That makes for an intense hour for our timing company and finish line staff.

What kind of “anomalies” do you see-and how do you find them?

In order to verify that all runners listed as finishers completed the entire course, we have a few systems in place. Our first step is to scan the results to look for finishers who are missing one or more of the intermediate split times. Simply missing one or more splits does not automatically mean we classify a runner as a DNF, but it does mean that we look closer into that runner’s data. This year we’ll publish split data at the start, mile 10, mile 13.1, mile 20, and the finish. We will have a mat near 10k on the out-and-back highway, and we will also have at least one other unpublished mat. If a runner had in mind to cut the course they might cross all the published mats, but it’s not likely that they’d also find the unpublished mats.

One of the things we see every year are a few runners who have splits at the start, 10, 13.1, and the finish, but they are missing the 20 mile split. The nature of our course is that a runner who is not ready for the full 26.2 can run the first half, and then drop out at mile 15 and enjoy the post-race party. We also see some runners who get past mile 15, the biggest hill on the course, but they are too fatigued and drop out a few miles later and walk back to Waterfront Park. If someone is missing only the 20 mile split, it’s pretty easy to tell by comparing their halfway split to their finish time if they did not complete the race. What happens a lot of times is the runner will walk through the finish line to turn in their Chip (in the old days), or they’ll walk next to the finish line and the mats will pick them up as a finisher. We’ve gotten better at controlling this over the years, but inevitably a few runners are recorded as finishers because of this who should not be. A lot of times these runners will contact us before we’ve even had a chance to review the results on Monday.

We also look for runners with huge negative splits. A big negative split is not in itself proof that something is amiss, but it is worth checking out. A few actual examples that we have seen:

  • A runner is 1:40 at 10 miles (10:00/mile pace), 2:00 at 13.1 (6:30/mile for this 5k section), then finishes in just over 3:30 (7:00 pace the last half). We contacted the runner, and were told that he ran a friend through 10 miles, then picked it up to his normal pace the rest of the way. The chip data for him and his friend are the same at 10 miles. So in this case the huge negative split makes sense.
  • A runner has no start split, is 1:30 at 10 miles, 2:16 at 13.1, 4:00 at 20 miles, then finishes in just over 5:30. No start split was a tip off that this runner may have started prior to our 8:03am official start. The fact that the 10-13.1, 13.1-20, and 20-26.2 splits were all right on 15:00/mile, while the start-10 split was 9:00/mile, was strong evidence supporting our theory. When we contacted the runner he admitted to starting early.
  • A runner is 1:51 at 13.1 miles, then runs a 2nd half split of 1:23 for a total of 3:14. He is missing the 20 mile split and does not appear in video shot at the 22 mile mark or on the unpublished chip mat data at 21.5 miles. The runner has half marathon times from other races in the 1:45-1:55 range. It is the ruling of the judges that this is a bogus finish time and the runner did not run the entire course-the runner is removed from the results.
  • A marathon entrant (not relay division!) posts a 3:20 finish time with a 13.1 split of 1:45. When reviewing the photos under this bib number the 9 and 13.1 mile photos are of a female and the 26.2 mile photo is of a male. Both runners are wearing the bib on a race belt. When we contact the entrant to determine what happened, we are told they decided to race as a relay instead of the entrant running the whole marathon and they exchanged just before the 15 mile mark. This is of course not permitted. The finish result is removed from our official results and the runners are banned from future participation in VCM.

So, how do we make the results official?

After researching the timing systems based on the emails and phone calls we received from runners who are missing, we can almost always find the evidence the runner wants which backs up their effort on raceday. We work with our timing company to submit our findings, and they also work to review the backup systems. Many times they find runners missing in the results before we even contact them. By the end of the week after the race we usually have 95% of the issues cleared up, at which time our timing company reposts results on Cool Running. As soon as we feel we have all the issues cleared up, which can be as soon as 5 days or as much as 2 weeks after the race, we submit our official results to USATF. And that’s a wrap on another year!

What to do if you finished the race but are missing from the published results?

First off, contact our office as soon as possible. 2-3 days after the race is good, 2 weeks later is not good. The best way to get the ball rolling is to email us so that we can start a paper trail. Phone calls are going to be less effective as an initial contact, but please include a phone number in your email in case we need to speak with you to clarify any items. Information to include that will help our detective work:

  • Bib number
  • Shorts, shirt, shoes, hat…what were you wearing? If you have a picture a friend or family member took of you during the race please attach it to your email.
  • What time was on the clock or your watch when you finished? If the time is from your watch and you started it when you crossed the start line let us know that, and if you have a general idea of how long that was after the horn that is good for us to know too. If you have any midway splits feel free to send them along, even if they’re from different points from our official mats.
  • Did you cross the finish line near anyone else? If you know that person let us know. Even if you can only say it was “tall male, black shorts, blue shirt” that helps.

What can you do to prevent these problems?

The #1 thing is to make sure your bib is clear on the front of your shirt. At any time on the course there may be cameras or timing equipment that needs to able to see or read your bib. Covering your bib with a jacket, sweatshirt, or a FuelBelt means our cameras can’t identify you. Should the electronic timing fail for your bib our cameras are our next step to try to figure out who you are. One thing we learned in 2010 and 2011 is that hydration belts can cause big problems for electronic timing in a bib. The main issue is that the fluids in the bottles prevent the device in the bib from communicating with the timing mats. The secondary issue is that a bib under a hydration belt is easily damaged.

#2 is do not fold or cut your bib. The timing strips in your bib can be damaged if bent. The bibs can handle regular stress from normal use, but they are less reliable if you intentionally bend them. Related to this is try to avoid attaching your bib to a body part that is going to be bending a lot. In 2010 over 50% of the bib errors we experienced were on runners who pinned their bib to their shorts near their hip. The continuous bending motion of your hip flexor as you run can damage the timing sensors.

The bottom line is that our goal for our final official results as delivered to USATF and published online to be 100% correct. In an event of this size that requires post-race research and patience. Any errors you see related to your own results or that you are curious in some else’s result, please feel free to contact us and we’ll check it out for you.

{ 7 comments… read them below or add one }

1 Helen May 25, 2013 at 7:34 am

Interesting read. But, if the fluid in a hydration belt causes the electronic timing to fail, what’s the pouring rain going to do to it? Yikes!

2 Fitz May 25, 2013 at 7:42 am

What rain?! Where’s your optimism Helen? Can’t wait, thanks for all these FAQs guys!

Joe Connelly 3 joe May 25, 2013 at 8:23 am

Helen-

In our experience, pouring rain and profuse sweat can affect the B-tags negatively to a small degree, but the system is getting better every year. The difference with hydration belts is you’re trying to read the tag THROUGH fluid, as opposed to a tag that just happens to be wet. Huge difference.

This is similar to my heart rate monitor. When I’m doing triathlons I don’t get accurate heart rate readings during the swim leg, but on the bike and run even though I’m sweaty the heart rate function works correctly.

Joe Connelly
RunVermont Technical Coordinator

4 Maureen Somerwil May 27, 2013 at 1:32 pm

My time is missing. I tried to send an email but it’s really not clear where it should be sent. I’ve run many, many races and have never heard of such finicky bibs for timing. Maybe you should look into another company.

Joe Connelly 5 joe May 27, 2013 at 3:06 pm

Hi Maureen-

Got your email, hopefully we’ll be able to find your result shortly.

As for “finicky bibs”, although I’m not the timing contractor for VCM I’ve been timing races for 25 years, have worked with all kinds of systems from the most basic manual systems to the most modern chip systems. All systems have pluses and minuses, NONE are perfect. In my experience of racing over 1000 races and working probably 400+/- races I’ve learned that lots of different things can happen to cause any specific individual result to be wrong. I would say that based on what I’ve seen since VCM went to chip timing about a decade ago roughly 75% of the errors are due to things runners do (see the last 3 paragraphs above) and 25% are due to problems within the timing system. But it doesn’t matter which timing system you use, you still have the potential for errors to happen, the mark of a good race is having the backup data available to be able to issue final official results that are accurate.

Our current timing contractor is one of the largest companies in the business in New England. We could look into using another company, but it would be impossible to find one with any higher level of competence and professionalism.

Joe Connelly
Technical Coordinator

6 Nicholas Ferron June 14, 2013 at 8:15 am

I am very happy that VCM is taking efforts to detect cheating and I’m happy to wait as long as necessary to get true and honest results. I believe that the VCM is especially susceptible to cheating, particularly this year, for a couple of reasons:

1. This year, a particularly large number of runners want to qualify for Boston 2014. The consensus is that the race will fill up even before it opens to the standard “B.Q.” crowd and most believe that they will have to run B.Q. minus 5, or even B.Q. minus 10, to have a shot at registration. As VCM is one of the closest full marathons to Boston and one of the most popular Boston qualifiers, I don’t think it’s overly pessimistic to think that a few runners would cut corners to increase their chances at being part of a potentially historic 2014 Boston Marathon.

2. The course has several out-and-backs, all of which end within a few miles of the finish line. It is possible, therefore, that a few less scrupulous runners could think they could beat the system by running one or two loops, wait a while, and then sneak back in and cross the finish line.

3. The weather was awful — maybe awful enough to provide justification for some people to cheat. Less than ideal weather conditions could provide a built-in excuse for some.

Thank you for all of your efforts and for generally putting on a good race. For my part, I do believe that there are cheaters, and in spotting them, VCM is not only protecting the integrity of its own race, but also the integrity of Boston Marathon and sport at large.

Joe Connelly 7 joe June 14, 2013 at 3:14 pm

Thanks for the kind words Nicholas. There are ways to “beat” almost any race out there if one is motivated to do so. From the Boston Marathon to the Green Mountain Marathon I doubt there is a road race out there who hasn’t had to deal with this situation. (True story, GMM 1989 which I did, the 2nd place finisher was DQ’d for hitching a ride…82 finishers, out-and-back course, what was that guy thinking?!?).

I think there are 2 main things we do at VCM to deter runners who want to cut the course and to catch them if they do; 1) we place Course Monitors at places that we believe would be tempting, for example right around mile 18 runners are within 50 yards of the bikepath (mile 24), we place a CM there who is trained in what to look for, 2) we almost always have at least 1 unannounced chip mat. Last year we added the mats at 10k and 21.5 miles, which are both near the furthest points a runner would have to go to. If a runner’s data record has them missing a mat it’s not an automatic DQ, but it does mean we look at their splits, video, and pictures more closely. Usually it’s just a case of a chip malfunction but other times it’s pretty clear that the runner didn’t run the full course.

One thing that was interesting to me last year, we had an unannounced mat at mile 21.5 basically to back up our mat at mile 20 (if you’re only missing on one of those it’s a chip malfunction, if you’re missing on both but have other chip splits that would be highly suspicious). We had ZERO finishers where the 21.5 mile mat gave us any concern. Based on my experience that was a big surprise. Due to that we did away with the 21.5 mile mat for 2013, but some years it will be back out there.

For those of you who are into data/statistics, this year we removed 3 finishers from the results for suspected course cuts out of the 2625+/- marathon finishers. One I am positive of, a close friend of mine was working the spot where this racer cut. Within 2 minutes of the racer cutting the course (he cut about 4 miles out) my friend called in a description and bib number. The other two cases we found through reviewing the chip data and race pics/video. Other than that I have gone line by line through this year’s marathon results and I’m confident that all who are listed as finishers completed the entire course.

At some point, maybe for April Fools Day next year, I should write up a “how to cut the VCM course” tutorial. After 25 years around this race and running the streets/trails of this city (see my January 28 blog post for a different take on “running Burlington”) if there is a way to cut our course I know it. Local knowledge comes in handy when we’re training our Course Monitors.

Leave a Comment

Previous post:

Next post:

KeyBank Vermont City Marathon & Relay
RunVermont Events
KeyBank Vermont City Marathon & Relay »
Sport & Fitness EXPO »
MVP Health Care YAM Scram »
FirstRun »
Half Marathon Unplugged »
Youth Fun Runs »
RunVermont Supported Events »
Youth Programs
Ready, Set, Run! »
Junior Youth Relay »
YAM Scram »
Fun Runs »
Hall of Fame
Profiles »
Nominate »
About Us
Overview »
RunVermont Staff & Board »
Volunteer Opportunities »
Employment and Internship Opportunities »
News »
RunVermont Services »
Community »
Calendar »
Race Results »
Race Services »
RunVermont Store »
Contact »
Blog »
Become a Sponsor »
 
Vermont City Marathon
RunVermont