When wheeling/measuring a course with a surveyor's old school foot-by-foot wheel measuring device (Jones counter?), should I just measure to the exact foot or allow for a few extra feet per quarter mile? (That means 1320 feet per quarte mile) Is there a protocol for this (because I want these marks to be accurate)?
SCPF - Short course prevention factor adds 1/1000 of the distance to be measured.
Marathon courses are therefore measured an extra 42.195 meters.
Each mile is 5.28 ft. long.
Don't confuse the surveyor's wheel with a Jones counter. They are two entirely different devices. Typically a surveyor's wheel will give you a short measurement, unless you are very careful. This is because it is very difficult to keep it in contact with the ground at all times. It tends to skip over small bumps etc. causing it to freewheel, thereby recording extra distance.
The Jones counter is the only approved device for course measurement.
Use a steel measuring tape with the proper tension. Be sure to correct for temperature.
Tedgo, I also need to measure a course. Where can I find a steel tape measure 10k long?
The Japanese use the "chaining method" - a 50 meter steel tape - yes even for a marathon. They are very good.
You could do the Benji method - measure a course so that the shortest route is the distance to be raced + 5%.
It obviously isn't going to be certified, so...
For cross country courses, I've found the wheel to be about 99% accurate, that's to say, in a stretch of 100m, it was accurate to 99m. This scenario played out over most of our course, enough to say, after measuring 2k, I did the other 2k with the 99/100 rule in mind.
The problems are, as follows:
Measuring too slowly will accumulate EXTRA distance. By MAKING SURE the wheel is constantly in contact with the ground you will go up and down small rises that are not representative of the actual travel. This method is often inaccurate because of small, tiny adjustments you'll make (small zig zags that add distance)
Measuring too quickly will cause the wheel to jump up slightly, and the wheel will decelerate, causing the distance to be short. This is more accurate though, because you can determine a relative rate of failure, and you keep a more appropriate line.
I recently posted a rant against the Short Course Prevention Factor on another website. A SCPF of 1/1000 of the race distance seems OK (e.g. 3/1000 of a mile for a 5K), but I have encountered courses in New Jersey that were over the stated distance by a few hundredths up to a little over a tenth of a mile. Does NJ use a larger SCPF, or are the certifiers here a little over-zealous in ensuring that their courses are "at least" the stated length?
Your last ? first: no, NJ doesn't use a bigger SCPF amount than others. All in the US use the same .1%.
Secondly, how do you know some NJ courses are 'over the stated distance by a few hundredths up to a little over a tenth of a mile'? Did you go out w/a Jones-calibrated bike and remeasure the courses or have a reliable source do so? If not, you're guessing.
Scotth, you sound a bit rankled by my query. When the subject of inaccurate courses comes up around USATF officials, I usually hear the same reaction. The knee-jerk response is that a calibrated Jones counter is the most accurate measuring device on the planet.
Yet, when I "measure" a course with a fairly accurate GPS device, like a Garmin 305, re-measure it, then check with a satellite website, and get the same result to the nearest hunredth of a mile, I can say with some confidence that I have obtained a pretty accurate measurement of the course. When the number I obtain differs significantly from the stated course length (in one case recently by .04 miles and in another by .12 miles, long in both instances), I would be willing to bet that the course is off.
I think my main complaint is that especially here in NJ, certifiers seem to be more concerned with ensuring that a course is "at least" the specified length, than that it is accurate.
OldXCguy wrote:
I would be willing to bet that the course is off.
I think my main complaint is that especially here in NJ, certifiers seem to be more concerned with ensuring that a course is "at least" the specified length, than that it is accurate.
Your sources of accuracy are NOT as good as a Jones counter mounted bike. If your sources were as good, course certifiers would use them. All GPS generated numbers I've seen from certified courses have found the course long. We have some consistency there...consistently long and wrong.
Your final comment is nebulous. All course certifiers measure a course to be 'at least' the nominal distance. You don't say what you think qualifies as 'accurate'.
OldXCguy,
Are you measuring the shortest possible route? Running diagonals from corner to corner can make a big difference if there's enough of them. Did you look up the map of the course on the USATF website to make sure what you're measuring is the same course described on the map? It's possible the race itself is not following the map.
If you have found a certified course to be 0.12 miles long, post the name of the course and the city. I'll look it up, measure it with Google Earth, and let you know if I agree.
Dukerdog, of course I can't be sure that I've cut every tangent as tightly as possible, but I'm certainly not zig-zagging around when I'm racing and/or trying to measure a course. In mapping a course, the tendency is to come up short, unless one plots points for every little turn or curve on a course. The 10K course that I found to be 0.12 miles long, using a Garmin 305 and a mapping site, NOT A JONES COUNTER, was at a race for whom the director is a friend of mine, so I would rather not post it. If you would like to provide your Email address, I'll send you the info.
Soctth, how can you make a statement like, " All GPS generated numbers I've seen from certified courses have found the course long. We have some consistency there...consistently long and wrong." Can you offer any
scientific evidence to support your claim? Please do not try to make the case that I am doing the same, as I am giving specific instances where the measurements are reproducible using different means. Oh, and I would much appreciate it if you would lose the condescending tone.
Dear OldXCguy,
I love new technology - I use as much of it as I can. GPS units are great and give a relatively accurate measurement of a road race course. I have used handheld units many times while also using the accepted methods of mesurement (calibrated Jones device). In every case, the GPS gives a longer reading then the accepted method.
The GPS measurement is always inconsistent with splits. This could be for several reason:
- trees or buildings block the signal
- commercial grade units can't follow the exact path that measured by the bike.
I can reproduce a measurement with the bike/Jones method within 2 meters of a 10k course and could never repeat that accuracy with the gps units available.Infact this past weekend I re-meaured a hilly, winding 5k course for the fourth time and came up within 1 meter of the first 4 measurements.
I must be doing something very wrong for the past 30 years.
I welcome the day when an electronic device can give us the same or better accuracy for the requirements of a road race without spending days to measure (we could always use an EDM)- but I don't think that day has come.
Warmest regards,
Oldmeasuringguy.
OldXCguy - Scott can make the statements he makes because he has (I believe, Scott correct me if I'm wrong) measured hundreds of race courses using the USATF accepted methods. Experience is a great educator in this process. How many courses have you measured by the USATF accepted method? If you truly believe that courses are being measured substantially long by the Jones counter maybe you can do some comparisons with your Garmin 305. The USATF requirements are published, I believe at
, and official course maps with tremendous detail are also available through the USATF website. I would honestly be very interested in your findings.
I can say that I agree with Scott that, at least when it comes to marathons, Garmins and other household GPS devices tend to overstate race course distances. I have no scientific evidence to document this, just have read dozens of race reports (mostly on the runners world and cool running forums) that question various marathon courses as being "long". I cannot recall ever reading 1 race report that said their Garmin showed a short course.
The marathon I work for has gotten comments from runners that their Garmin showed the course to be anywhere from .2-.4 long. I was in on the certification ride so have faith that it was done properly. Plus we have remeasured some intermediate sections and have always found a high degree of accuracy in any subsequent measurements (within 6" +/- of the original mile marks). Off that I'm going to say the Jones counter is more accurate than a Garmin.
In the race you set as an example, the 10k that you measure as .12 miles long, did your Garmin and the mapping site give you the exact same discrepancy? It would seem nearly impossible to reach this level of consistency by these 2 methods but I'll grant that it is not impossible. My math says this is ~634' long. A 10k, measured perfectly according to the USATF rules including the SCPF, should come up about 33' long, if I understand the process correctly. That leaves ~600' unaccounted for. You would have to veer all over the place to add an extra 100' per mile. Is it possible the start or finish was set up in the wrong place? You'd have to check the map published on the USATF site and compare it to the actual spots used to be sure.
OldXCguy, please send me that course info.
runnermark[at]gmail.com
You need to find the fractal dimensional increment first. Then, from a small portion of the course measured with your large "ruler" (wheel) you can determine the real measure of the course based on your measure of the entire course with the wheel.
[quote]d2xccoach wrote:
I have no scientific evidence to document this, just have read dozens of race reports (mostly on the runners world and cool running forums) that question various marathon courses as being "long". I cannot recall ever reading 1 race report that said their Garmin showed a short course.
Of course not. I bet you've never seen any of them complaining about the good weather either. When someone's Garmin says the course is long, they decide the course is long. When someone's Garmin says the course is short, they decide their Garmin is wrong, record their new PR in their logs, and don't tell anyone what their Garmin said.
It may be the case that people wearing Garmins in races usually come up with slightly long readings, simply because they aren't running the SPR, but the data source you mentioned is extremely biased.
why do a lot of you insist on equating miles and meters all the time, you really complicate things.If you're adding
1/1000 to a 5K for SCPF then just say you add 5 meters, why convert it into miles all the time, otherwise you have to correct for the conversion correction as well.