Road Course Measurement Bulletin Board    Forums    Course Measurement    Electronic Measurement    Generic Reply to GPS Accuracy Doubters
 Go New Find Notify Tools Reply
 Generic Reply to GPS Accuracy Doubters

 posted 30 August 2010 14:32
Yesterday Joan and I had a visit from Doug Thurston, who used to direct the Columbus Marathon before Joan took over the job. While talking of old times over dinner, Doug mentioned that he has to deal with many inquiries from GPS-equipped runners who are convinced that the course was inaccurate. He worked up what he calls a “generic reply.” It follows.

GPS Readings on Certified Rock ‘n’ Roll Marathon or Half Marathon Courses

Thank you for your comment and question about the length of the Rock ‘n’ Roll race you recently completed. Because we often receive questions about the course length compared to GPS readings, we have developed this standard response.

The course you ran was certified as accurately measured by USA Track and Field, the governing body for road running in this country. Courses are not measured by GPS, tape measure, surveying equipment, car, or by walking with a measuring wheel. All road race courses are measured by a standard bicycle fitted with a calibrated mechanical counter on the front wheel.

GPS readings on the course are almost always slightly longer than the distance measured with this mechanical counter. There are several reasons for these differences:

1) Courses are measured by the shortest-possible route available to runners on race day. This means that the measuring bicycle is ridden within a foot of the curb and turns are “straightened out” as much as possible. This method is used to ensure that no one runs shorter than the stated distance. With other runners on the course, however, this shortest route can be difficult to run exactly as measured.

2) All certified courses include a 1/10th of 1 percent (one-thousandth percent) "short course prevention factor." This is a small extra cushion to again make sure no one runs less than the stated distance. So, in other words, a certified 10 km (10,000 meters) road race is actually measured as 10,010 meters. For a full marathon, that means 42.1925 extra meters, or a little more than 135 feet. This extra cushion is spread out throughout the course, not simply added to the start and finish, and is present in ALL certified courses.

3) And, lastly, consumer GPS devices are not 100% accurate. We have found they are typically 1-to-2% off. GPS accuracy is affected by elevation, the number of turns, tree coverage, tall buildings, bridges and overpasses, and the quality and quantity of satellite reception. In a half marathon, a 1% difference is more than a tenth of a mile. So, readings of 13.25 miles or more for a 13.10938-mile certified half-marathon and 26.50 miles or more for a certified 26.21876-mile marathon are not unusual and are to be expected.

I realize this may be a long answer to a simple question, but we take course distance very seriously and all of our races are certified by experienced measurers, many with decades of experience. I am internationally-accredited and I have measured many of our courses (San Diego, San Antonio, Las Vegas, Seattle, and New Orleans, for example). When I measure a course with my mechanical counter, I also have a Garmin Edge 705 mounted on my handlebars for reference, elevation, and computer mapping. My Garmin reading is always longer than the distance measured with my mechanical counter. For example, I recently measured a half marathon, which is 13.109876 miles, and my GPS read 13.30 miles.

The GPS is a useful and informative training device. I use a Garmin when training and racing. I have come to understand and accept that the information it provides me about distance, pace-per-mile, average pace, elevation, etc., though very close, is not 100% accurate. I do trust the stopwatch function, of course, but I know the distance feature will almost always give me a reading that is “too long” when I race on a certified course. (Consequently, the "pace per mile" or "average pace" functions are too fast.)

If you want to learn more about how courses are measured, visit the “Course Certification” area in the “Products and Services” section of the USA Track and Field website (http://www.usatf.org/events/courses/certification/). If you have any questions, contact me directly at dthurston@competitorgroup.com or by phone at 858-228-3763.

Thanks for your participation and best wishes for your training and racing.

Doug Thurston
Director, Operations
The Competitor Group
Organizers of the Rock 'N' Roll Marathon Series
9477 Waples St., Suite 150, San Diego, CA 92121
dthurston@competitorgroup.com
www.competitor.com
Office (858) 228-3763

 Posts: 1756 | Location: Columbus, Ohio, USA | Registered: 23 October 2004 IP

 posted 31 August 2010 07:25 Hide Post
#1 and #2 are great answers that have been shown by testing to be true.

The only test results that have been presented in the past 5 years (current crop of GPS units), that I am aware of, have shown that when there are no tall buildings near the course the distance measurements reported by GPS units are no more than 0.5% in error, and usually a good bit less.

It has been shown in two sets of tests carried out by two different groups that people in races run 1-2% longer than the actual SPR distance of the course. That's really all that needs to be said. And it makes everyone happy because it means that the person reporting his GPS measurement probably DID run 1-2% longer than the stated course distance.

 Posts: 996 | Location: Rochester, MI | Registered: 13 April 2006 IP

 posted 02 September 2010 17:49 Hide Post
Sorry, but I won't concur that #2 is accurate. The SCPF is there to compensate for wobble, or when the SPR is not actually ridden. Therefore, while a 10k is measured to 10,010 meters on the Jones counter, the actual course length will be somewhat less than that.

While I can record the same number of clicks on both rides of a 5k, and occasionally of a 10k measurement, I would bet that if a tape was used, my total course length would be shorter than 10,010 meters.

Nit-picky, maybe, but we should not tell runners that a 10k course is always 10,010 meters, or that a marathon course is always 42 meters long. That is not necessarily correct.

 Posts: 731 | Location: Denver, Colorado | Registered: 09 May 2006 IP

 posted 02 September 2010 23:06 Hide Post
I once had a guy tell me that while warming up he used his GPS to measure the first mile of the course, which I had measured, and it was off. Smiling, I asked him what his GPS said, and he told me 0.62 miles.

 Posts: 996 | Location: Rochester, MI | Registered: 13 April 2006 IP
 posted 02 October 2010 12:15 Hide Post
I would never state that a course I measured was exactly x long (I work at it but I'm not that good!) I do guarantee to the race director that it is at least x kilometers long and is certifiable under our standards.
As an aside, I'm always impressed reading this forum as to the level of accuracy course measurers aspire to and the lengths they will go to achieve it.

 Posts: 10 | Location: Albuquerque, New Mexico | Registered: 24 June 2010 IP

 posted 03 October 2010 07:08 Hide Post
John, we not only go to those lengths, we go 0.001% beyond them

 Posts: 745 | Location: Norwalk, CT | Registered: 24 October 2004 IP

 posted 07 October 2011 10:08 Hide Post
I had RD's call - sounding very worried - after a race, because they had people complaining about their course being long. After talking to way too many of them, trying to explain what I did, I produced a single-page information sheet, "The Course Measurer Says This Is Accurate But My GPS Says..." I began giving this to race directors as part of their cartificate package (certificate, guidance on using the certificate number/map, information sheet, invoice for services).

The sheet, which I've done for 5K, 10K and half-marathon distances, compares RRTC guidance, the answers of a competent course measurer/and the "typical" GPS-using road runner:

- Define "Accurate Course"
CM - USATF/IAAF: "...a course not shorter than the stated distance."
GPS - Varies from person to person

- Define Half-Marathon
CM - 13.10938 miles exactly
GPS - Most say "13.1 miles"

- Number of Unhindered Trips on Course
CM - At least two
GPS - Depends on experience and course knowledge

- Type of Measurement Device
CM - Mechanical (Jones-Oerth)

- Means of Measurement
CM - Counts per Revolution of Bicycle Wheel

- Degree of Accuracy
CM - +/- 3 inches continuously
GPS - 3m-10m radius 95% of the time, 11m-30+m radius 5% of the time

- Device Ensured Accurate?
CM - Calibration must be done at least twice each measurement day
GPS - Accuracy can be displayed but not controlled

- Comparison Between Trips?
CM - Measurements must be within .08% (55 ft) of each other
GPS - Possible, but not before race day

- Course Travel
CM - Shortest possible distance a runner can legally take on course
GPS - Depends on experience and placing in race

- Start/End of Measurement
CM - Exact start & finish points on course
GPS - Usually with gun/horn, rather than at start line

(EDITORIAL COMMENT: I USUALLY TELL GPS-USERS IF THEY DON'T START AND STOP THEIR UNIT AT THE EXACT START/FINISH LINE ALL THEY HAVE IS A REALLY EXPENSIVE RUNNING WATCH ON THEIR WRIST.)

- Experience Level
CM - 20-year road racer, 6-year measurer, 27 certified courses
GPS - ?

- Oversight
CM - All work approved by certifier/measurer of over 55 active courses
GPS - ?

This message has been edited. Last edited by: Michael Bowen,

"Typos in this message are most likely the result of a very curious chihuahua."

 Posts: 100 | Location: Pensacola, FL | Registered: 21 December 2004 IP
 posted 16 October 2011 16:59 Hide Post
All of these comments are very helpful to me at the moment--folks who ran the Army 10-miler are writing with the same issue ("GPS says 10.2 miles"). RD George Banker wants to put a generic statement on the website. I hope none of these writers will mind if their points are used (?).

 Posts: 535 | Location: Washington, DC | Registered: 16 November 2004 IP

 posted 16 October 2011 20:07 Hide Post
Bob - I'll send you a copy of the doc if you want...e-mail my "real" work address (michael.bowen@navy.mil) and I'll send it to you (I'll even break it down for other distances if you wish!).

"Typos in this message are most likely the result of a very curious chihuahua."

 Posts: 100 | Location: Pensacola, FL | Registered: 21 December 2004 IP

 posted 17 October 2011 21:29 Hide Post
Michael et al;

This is part of an email from a race director whose marathon course I measured,

"These garmin geeks are driving me crazy! What is your response? I am going
to post an explanation on the website and try to not use bad words."

I used some of your words in my response. This is the 3rd or 4th such question from a race director, not questioning the USATF measurement process or my accuracy, but looking for words that will make the doubt stop. Here's my response;

"USATF has considered using GPS for course certification and decided that it is less accurate than the approved (calibrated bicycle) method which is accurate to + or - about 3-inches of the reading on the counter mounted to the bike. The GPS unit is accurate to about 10-30-ft of the reading on the face of the instrument.

The 0.2 mile increased length noted by someone who doesn't trust the USATF certification process was due to the accuracy noted above and to not running the shortest possible distance, as the course was measured. Interference caused by buildings, trees and clouds also increases the GPS error while not affecting the calibrated bicycle measurement.

The USATF course certification program has been around about 35 years in its current form. It has been tested hundreds of times and found the best method to state with reasonable certainty that the course is at least the advertised distance. The credentials of the [your course] measurer; 25-years measuring, IAAF -B-Measurer, about 200 courses measured, 3 of those had world records set on them and were later length-validated by a separate measurement done by a USATF official. Hope this helps. I can dig up a much more detailed explanation of GPS accuracy, but it puts me to sleep and I'm pretty technical."

Pete Volkmar
Guido Brothers

 Posts: 216 | Location: Connecticut, USA | Registered: 23 November 2004 IP

 posted 18 October 2011 09:27 Hide Post
Great as far as brevity is concerned! I think I would mention about the GPS accuracy: "10-to-30 feet AT A SINGLE POINT..."

Naturally, when you get more than one point (Which GPS' take on the average of every one to two seconds, right?) the accuracy issue is further confounded.

But, for the benefit of the 'garmin geeks' (God, I love that...I need to borrow that for a blog post!) the 10-to-30 should work fine.

A tip of the running cap and the beer mug in your general direction, Pete.

(I was in a conversation with a non-running friend the other night, who understood the GPS/shortest possible distance limitations right off the bat. If a non-runner can understand it...?)

"Typos in this message are most likely the result of a very curious chihuahua."

 Posts: 100 | Location: Pensacola, FL | Registered: 21 December 2004 IP

 posted 12 December 2011 10:53 Hide Post
Here's another great, if somewhat lengthy, explanation. I forwarded the link to all my geeky GPS-wearing runner friends.

In GPS We Trust: Runner threatens law suit over course measurement disagreement.

 Posts: 745 | Location: Norwalk, CT | Registered: 24 October 2004 IP

 posted 27 December 2011 11:59 Hide Post
Several months ago, a GPS watch manufacturer published an "advertorial" in the Washington Running Report in which the author claimed that using its product would provide runners with "precise" measurements of the distances they run. I took exception in a letter to the editor, and WRR then published a brief article on course measurement. The article avoided the GPS issue.

Nevertheless, in the D.C. area, it seems to me that there are fewer complaints about certified courses being "long" recently. I am not certain why this may be. My naive hope is that GPS wearers are becoming more knowledgeable about what these devices can and cannot do, and how to benefit from them.

I assume some runners have become aware that civilian-grade GPS devices have a built-in accuracy limitation. Does anyone know what the resolution difference is between Standard Positioning Service (SPS) and Precise Positioning Service (PPS)? Can anyone say what the expected difference in measurement accuracy might be between these two technologies?

 Posts: 427 | Registered: 15 January 2010 IP

 posted 19 January 2012 11:42 Hide Post
quote:
Originally posted by Michael Bowen:
Great as far as brevity is concerned! I think I would mention about the GPS accuracy: "10-to-30 feet AT A SINGLE POINT..."

Naturally, when you get more than one point (Which GPS' take on the average of every one to two seconds, right?) the accuracy issue is further confounded.

But, for the benefit of the 'garmin geeks' (God, I love that...I need to borrow that for a blog post!) the 10-to-30 should work fine.

A tip of the running cap and the beer mug in your general direction, Pete.

(I was in a conversation with a non-running friend the other night, who understood the GPS/shortest possible distance limitations right off the bat. If a non-runner can understand it...?)

Nice, Guido, Michael. Thanks I will use this in a slightly different form. Give me your thoughts on this version:

USATF has considered using GPS for course certification and decided that it is less accurate than the approved (calibrated bicycle) method which is accurate to plus or minus approximately 3 inches of the reading on the counter mounted to the bike. Civilian-grade GPS units are accurate to about 10-30 feet of a single identified point of the reading on the face of the instrument. The “increased length” sometimes noted by GPS device-wearing race participants is due to two factors: (1). the accuracy limitation noted above; and (2) to these participants not running the shortest possible distance, as the course was measured. In addition, interference caused by buildings, trees and clouds increases GPS error while having no effect on the calibrated bicycle measurement. The USATF course certification program has been around about 35 years in its current form. It has been tested hundreds of times by the Road Running Technical Committee of USATF. It has reliably been found to be the best method (other than steel-taping) to state with reasonable certainty that a particular course is at least the advertised distance. The credentials of _______________, your USATF course measurer; 25-years measuring, _______courses certified. Etc…

 Posts: 427 | Registered: 15 January 2010 IP

 posted 23 January 2012 07:06 Hide Post
Lyman, That looks good, I think I'll save it for use when next needed.

 Posts: 216 | Location: Connecticut, USA | Registered: 23 November 2004 IP

 posted 23 January 2012 13:16 Hide Post
Lyman,

I think the generally accepted accuracy of the calibrated bicycle method is +-0.1%. I would use that number rather than the 3-inch number.

The 10-30 foot accuracy for the absolute position of a point is an often-referenced number that has almost nothing to do with the accuracy of GPS in measuring distance. The reason is that almost all of that 10-30 feet is systematic error; it is the same for every point recorded. If two points are offset from their true positions by exactly the same distance and direction, the distance between them is still the same. How do I know most of that 10-30 feet is systematic error? Because if even 3-4 feet of it is random error (different for every point) people would be seeing GPS distances that were in error by 10% or more. I would leave off the 10-30 foot number and just say that people see long values on their GPS because they don't run the SPR.

I think the other parts of your response are good.

There have been several forum threads on the use and accuracy of GPS. You can see them all by going to
Measurement Forum topics
and scrolling down to the GPS listings.

 Posts: 996 | Location: Rochester, MI | Registered: 13 April 2006 IP

 posted 30 January 2012 13:20 Hide Post
Good point, Mark. Perhaps it is better to emphasize the accuracy of the calibrated bike than to shoot holes into the systematic variance placed into the civilian GPS...especially the fact the accuracy of the calibrated bike can be verified.

"Typos in this message are most likely the result of a very curious chihuahua."

 Posts: 100 | Location: Pensacola, FL | Registered: 21 December 2004 IP

 posted 02 February 2012 14:32 Hide Post
This makes good sense. Thank you. I will re-write this statement when I get time.

 Posts: 427 | Registered: 15 January 2010 IP