measuring xc courses?

I don't think I was beating it. Anyway, if so, only one beat. It was kind of an open question/response to that OCRS paper, which said that accuracy is becoming more important today, even in cross country.I'm kinda thinking that since terrains and conditions can vary so much, getting within 1% (e.g. 100m over 10K) would still be OK. It's not like they have world records for cross-country distances.I've only measured courses with my footpod, or PC mapping software.

scotth wrote:

There are no established guidelines for 'acceptable accuracy' of any course off-road, dirt, sand or grass. I don't know why you guys keep beating on this topic of non-road measurement. A more interesting contribution to the discussion would be to canvas about 25 folks that've measured off-road courses and see what device they used. Ask them what precautions they took to ensure course accuracy, ie. did they calibrate their device, etc.

I also find it interesting that at Worlds they just list something like 11782m. Is that just some basic wheel measurement, or does the IAAF do a good job of measuring - they just don't bother trying to get the exact full distance?

Mr. Obvious wrote:

I was responding to the guy who suggested putting a big knobby tire on a click wheel. I don't think those can be recalibrated, can they? Maybe they can.

Actually I don't think anyone was considering using a traditional "click" wheel (i.e. one that measures in feet and inches). You can however have a standard bike wheel, say a 16" wheel with knobby tires with a Jones counter on it. It has enough weight to keep it down, wear of the tire is insignificant, and you calibrate the clicks just like you would if it was attached to a bicycle, on a calibrated course laid out by a surveyor or steel tape.

The normal mode of measurement with a Jones counter doesn't result in a feet/inches reading, just a "click" count. For example I would get roughly 4400 clicks on a 1/2 mile course on my mountain bike. You have to do some math each time you measure since you calibrate, get a count, then use that number to compute clicks per mile / kilometer.

Mr. Obvious wrote:

a big knobby tire on a click wheel. I don't think those can be recalibrated, can they? Maybe they can.

Yes. Calibration just means you're checking the number of counts for riding over a set distance. For example if you ride a 1/4 mile steel taped straight 4 times and the reading is 4000 counts, you know that a mile will be 16000 counts, and a 10k will be 99,420 counts. After measuring the course, you ride over the calibration course again to verify the calibration is the same.

Meter Man wrote:

J.R.-

Can you tell us how long it takes you to measure a typical 5k course? How do you locate kilometer and mile marks if you don't look at your click count while riding?

Usually longer than I've planned. :)

It's easier to make a preliminary ride first, to check the terrain and see the best (shortest) route, calibrate the bicycle, ride the course twice, calibrate the bicycle again, and make any adjustments. I would say a couple of hours. Then I go and put the intermediate marks after that, which might be on a different day, again checking the calibration before riding.

Thanks to J.R. and Keith for the answers. I understand better now.

I usually have a bicycle cyclometer going too (calibrated but not accurate), which helps in keeping track of approximate distance.

J.R. wrote:

I usually have a bicycle cyclometer going too (calibrated but not accurate), which helps in keeping track of approximate distance.

Why are you unable to get your "cyclometer" calibrated to be accurate? Have you looked at the cycling products available in the past few years?

Lance wrote:

Why are you unable to get your "cyclometer" calibrated to be accurate? Have you looked at the cycling products available in the past few years?

As explained before, bicycle cyclometers are not accurate.

The smallest resolution of 1/100 of a mile is 17.6 yards, 634 times less accurate than a Jones Counter. you can move the wheel 4 inches and the counter can be off by 7 feet (internally). Besides these things, the cyclometers are not consistent. I'm not interested to look at them as I have a Jones Counter.

This is a bit misleading, you can get readouts of 10m selecting metric. However, this does not indicate the resolution of the distance measurement, just the readout. It is better for an aggregate distance than for individual sub-parts as a result. Also, it will give you the broad picture of where you points are. Riding it first and noting (e.g., toss a marker) at various points for use with more detailed measuring is one often useful approach. Also, if you have several together doing measuring and have calibrated the bike with the Jones counter, the bike can give the partner a heads up approaching the next milestone.

Using a bike preliminarily can be useful when laying out a course where there is traffic that cannot be stopped etc. You can go over once or twice before you final measurement without disrupting traffic etc. and then do your final measurement at an easier time.

Lance wrote:

Why are you unable to get your "cyclometer" calibrated to be accurate? Have you looked at the cycling products available in the past few years?

As J.R. noted cyclometers are just not accurate enough for measuring.

To explain why, note that even a perfectly set bike computer would require calibration each ride. The temperature, weight of the rider, and inflation level of the tires will change the readings. It's possible to do this by changing the circumference number on the computer after riding a calibrated course, then re-riding the course to validate your math was right, but it's a real pain.

The other issue is a bike computer only registers once per revolution of the wheel, meaning one "click" every 80-90 inches, where a Jones counter registers one every 4-5 inches. Clicks can be missed on most bike computers because lack of physical contact between the reader and spoke. This can mess up measurement, particularly if it occurs when you are attempting to calibrate. There is a documented way around this by getting multiple magnets placed on the spokes (http://www.runscore.com/coursemeasurement/Articles/ElecComp.pdf) but you read through that, you'll see Neville's idea is non-trivial to implement.

J.R. wrote:

Lance wrote:

Why are you unable to get your "cyclometer" calibrated to be accurate? Have you looked at the cycling products available in the past few years?

As explained before, bicycle cyclometers are not accurate.

The smallest resolution of 1/100 of a mile is 17.6 yards, 634 times less accurate than a Jones Counter. you can move the wheel 4 inches and the counter can be off by 7 feet (internally). Besides these things, the cyclometers are not consistent. I'm not interested to look at them as I have a Jones Counter.

So let's say I have a "perfectly calibrated" cyclometer and I have it in the metric mode. I'm laying out a 5k road race course. I see the screen reading a 4.98, I slow down and see 4.99, I apply the brakes lightly knowing that in 10 meters the display will change. I stop when seeing 5.00. Wouldn't I be WITHIN 10 meters of 5000m?

Lance wrote:

Wouldn't I be WITHIN 10 meters of 5000m?

Maybe????

But, if so, that would be 394 times less accurate than a Jones Counter.

And how do you measure 1/2 or 1/4 of 10 meters?

What if you travel 4 inches, and it records 10 meters on the cyclometer? To me, that is not accurate. To you it might be close enough, which is fine. As mentioned before, I use one to get a very rough idea of the distance, that is all, but not to measure a course. I also use a map to check out distances, which is surprisingly accurate, but not to get a precise measurement of a course. This depends on how precise you wish to be with the measurement.

Also it is not possible to "perfectly calibrate" a cyclometer. Let's say you're riding over a 200 meter calibration course and one time your calibration comes out to 193 meters, another time 211 meters, 197, 206, 215, 199 etc. What is your calibration? The cyclometers are not consistent in their readings, and are not constructed to be that precise in their measurements.

I think if you use a Jones Counter, then you would easily see this by riding over a calibration course with either device.

Just to make another anal point:

Accuracy measures how close the average measurement will be to the real value.

Precision measures how repeatable a measurement is.

They are not always the same thing and I think folks in this discussion are misunderstanding each other because of this.

Clearly, cyclometers are imprecise, but if you calibrate them enough, one could make them accurate. If they measure to +/- 10m over 5000m and you ride the course 5 times, you'll probably be within +/- 3m.

Jones Counters appear to be precise and accurate.

From the study originally cited, the click wheel is precise, but not accurate.

Steel tape is both precise and accurate.

Now the question is which is cheapest in terms of time and money. Does anyone use a laser site? They would appear to be precise and accurate except for the curves.

dingle wrote:

Just to make another anal point:

Accuracy measures how close the average measurement will be to the real value.

Precision measures how repeatable a measurement is.

They are not always the same thing and I think folks in this discussion are misunderstanding each other because of this.

Clearly, cyclometers are imprecise, but if you calibrate them enough, one could make them accurate. If they measure to +/- 10m over 5000m and you ride the course 5 times, you'll probably be within +/- 3m.

Jones Counters appear to be precise and accurate.

From the study originally cited, the click wheel is precise, but not accurate.

Steel tape is both precise and accurate.

Now the question is which is cheapest in terms of time and money. Does anyone use a laser site? They would appear to be precise and accurate except for the curves.

I would think that the laser would not pick up any rise or run on the route.

It is not just a matter of which method is better, but which is better in person X's hands. Several people on this board thread are among the best as the process. The tape might be 'best' but is so time-consuming and onerous that inattention and error can creep in. It has repeatedly been found that taking a sample of shipments to get summary statistics on shipping is more accurate than using a census because of the errors that result in the process. (I think I had a qualifying exam question requiring a discussion of the issues and the theory involved but that was so long ago...).

With the cyclometer in metric I would not get any readings like that...just the .2. I do get meter by meter with my GPS up until 999m. I also get meter by meter to the 10th of a meter with my wheel and nver have diviations like you mention with my GPS or my wheel. Isn't my wheel at 10cm increments more accurate than your Jones Counter at 4" per click?

J.R. wrote:

Also it is not possible to "perfectly calibrate" a cyclometer. Let's say you're riding over a 200 meter calibration course and one time your calibration comes out to 193 meters, another time 211 meters, 197, 206, 215, 199 etc. What is your calibration? The cyclometers are not consistent in their readings, and are not constructed to be that precise in their measurements.

Another thing with the cyclometers, and I tried this with several of them, is that the calibration from 0 to 1 mile, for example, 1 to 2 miles, 2 to 3 miles, and 3 to 4 miles were different! I used to have a 4 mile accurately measured straight. :)

So even if the measurements could have been held within a certain error for one mile, the units for the other miles were different. It was a bit frustrating and got beyond my tolerance for messing around with them.

What's everyone's opinion on measuring a course with a Garmin forerunner 205 or 305?