measuring xc courses?

Why even bother? It is a cross country course. Place matters, not time.

? Which line??

1600 relay start in lane #2 should be a 3 turn stagger designed to accomodate the diagnol off of the back breakline to the 200meter start.

On a standard iaaf track with lanes 1.22m (48 inches)& curbing on the inside you get the following staggers for lane 2 from the common start finish:

400 m - 7.038 meters

4 x 400 relay - 10.564 meters.

Your measurement (calibration) would work for using the 400m start in any lane if you where measuring 400 meters.

But if you are using the 400 meter start for the 4 x 400 (1600m relay) than you have a small problem.

It appears that you are trying to do a good job, but I guess I'm not fully understanding.

David Katz wrote:

800 meter calibration course? I don't know of a measurer who uses one over 500 meters since Ted Corbitt back in the 60's. "using two people?". It's tough to do it by yourself!- but it can be done!

I used a surveyed 1/2 mile calibration course for years, up until it expired a few years ago.

As far as laying out a calibration course, it's one hell of a lot faster with two people, but it's possible with one with use of surveying nails every 100 feet (or length of tape).

Measuring an XC course with steel tape would be assine. The number of segments you'd end up with would inevitably lead to measuring errors. Sure it's possible, so is using a ruler. If you're that anal about getting an accurate measurement of an XC course that you don't trust a wheel get yourself a few hundred feet of steel chain or cable and calibrate it for temperature.

Thank you Keith for the kind words (assine).

And yes, I am that anal!

Hey fellas, how many measurers are dancing on the period at the end of this paragraph? Assine? I looked it up and couldn't find it so David, you might be anal (and I mean that in the nicest way) but you aren't assine.

Thanks Scott

time for bed.

everyone have a good night

more measuring on Thursday.

Terre Haute Course ? wrote:

So how come I can get 100m/400m/1600m/etc. when I use my wheel on the track when walking or running to check calibration??

You have to use a straight line for calibrating a device used for measuring running courses.

After all of this, I have a few questions.

What is acceptable accuracy for an off-road cross country course? 40 meters? 1%? 0.1%? 8 seconds? For my training, 1% is overkill, but maybe for cross-country races?

It's cross country after all. If I can accurately measure an 8K course to the millimeter, can I still attempt to compare a flat, manicured golf course, to a mountainous, sandy, rocky, and/or muddy course? With obstacles? On dry days versus sunny days, versus hot days, versus snowy days? Seems like there are a lot of other variables besides race distance.

Looks like the obvious problem with the OCRS study measurements (up to 19 feet/100m!) is the smooth hard rubber surface of the click-wheel. Replacing it with a fat knobby spiked tire, and/or make it heavier, might make it grip better on a variety of surfaces. They need an "off-road" click wheel. Of course poorly calibrated wheels have the obvious problem of error accumulation.

The OCRS paper talks about improving pacing. How can accurate measurements help pacing? After all, I would need a just as accurate realtime measurement of my instant speed (presumably more granular than 1km splits), and an accurate realtime measurement of course difficulty. I think for pacing, we have to fall back to the "old school" proven method of running by feel.

Just some random thoughts.

"Do you still think that if you are measuring a 100m course the error is less than the error in a 10k course that you measure with the same 100m tape?"

I didn't say it was less error just that you cannot take an error of say 1 cm measuring a 100m course and assume it will be 100 cm on a 10k course.

rekrunner wrote:

Looks like the obvious problem with the OCRS study measurements (up to 19 feet/100m!) is the smooth hard rubber surface of the click-wheel. Replacing it with a fat knobby spiked tire, and/or make it heavier, might make it grip better on a variety of surfaces. They need an "off-road" click wheel. Of course poorly calibrated wheels have the obvious problem of error accumulation.

An off road wheel with a fat knobby tire would be continuously changing circumference as the tire wore down, wouldn't it?

As would the hard rubber wheel of the existing click-wheels.

Surely the wheel can be redesigned to increase grip on a variety of surfaces, without introducing excess wear, for improved cross-country course measurements.

Or do you know that the current wheel was optimally designed for wet tall grass, and sand?

RE: measuring xc courses?

Look at the animation of a random walk here:

http://animation.yihui.name/prob:brownian_motion

note how on average, the balls move away from the center where they start. This is the sum of a bunch of random errors.

Notice how most of the directions cancel each other out. If you read what I said, which is most random errors would cancel each other out, I didn't say eliminate the error. Measuring a 100m course is not a large enough sample and you cannot assume the a 5k course with have errors 50 times the 100m course which is what the study assumed.

rekrunner wrote:

After all of this, I have a few questions.

What is acceptable accuracy for an off-road cross country course? 40 meters? 1%? 0.1%? 8 seconds? For my training, 1% is overkill, but maybe for cross-country races?

It's cross country after all. If I can accurately measure an 8K course to the millimeter, can I still attempt to compare a flat, manicured golf course, to a mountainous, sandy, rocky, and/or muddy course? With obstacles? On dry days versus sunny days, versus hot days, versus snowy days? Seems like there are a lot of other variables besides race distance.

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'm not sure why you think this applies to course measurement. You need to have positives and negatives in order to cancel. Relative to a shortest possible path there are only positive deviations. A wheel that wobbles from the shortest possible path can only add distance.

A wheel that is not perfectly calibrated or a tape measure that is short creates systematic errors, not random errors.

That is why I said that the _expected_ measurement is the same as the real distance, but the error increases. We probably don't disagree fundamentally. I actually already described how to predict the error on a 5k measurement.

Assuming a normal distribution with errors, the standard deviation of the measurement of a 5k (50 meaurements) would be the standard deviation of the individual measurement * the square root of 50. So no, not 50 times the error, but more like 7 times.

Bobby1 wrote:

you cannot take an error of say 1 cm measuring a 100m course and assume it will be 100 cm on a 10k course.

You can as long as the error is consistent, otherwise don't use the device. This is why you calibrate the device before and after the measurements.

Mr. Obvious wrote:

An off road wheel with a fat knobby tire would be continuously changing circumference as the tire wore down, wouldn't it?

It's not going to wear down that much in an hour or two. Again, you calibrate the bicycle before and after the measurements of the course, which should account for any difference in pressure in the tire (or wearing down of the rubber).

Bobby1 wrote:

Notice how most of the directions cancel each other out. If you read what I said, which is most random errors would cancel each other out, I didn't say eliminate the error. Measuring a 100m course is not a large enough sample and you cannot assume the a 5k course with have errors 50 times the 100m course which is what the study assumed.

You could just measure the distance from the start line to the finish line, and figure the rest of the errors will cancel each other out. /humor

J.R. wrote:

Mr. Obvious wrote:

An off road wheel with a fat knobby tire would be continuously changing circumference as the tire wore down, wouldn't it?

It's not going to wear down that much in an hour or two. Again, you calibrate the bicycle before and after the measurements of the course, which should account for any difference in pressure in the tire (or wearing down of the rubber).

I'm not a measurer, but I am enjoying the discussion and trying to learn some things.

I understand that bike counters can be recalibrated and agree that the wearing of rubber would be insignificant (certainly much less than the possible difference in tire pressure).

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.