Foot Locker admits to adding 70 meters to South Regional course

Math Man wrote:

You mean 8 FEET per mile?

No I don't, good catch. It should be .000126. Forgot to multiply 5280*12. That's what I get for using my phone instead of a spreadsheet.

Mr. Obvious wrote:

Thread Follower wrote:

Agreed, but how do these guys pushing for the steel tape method keep that steel tape stretched and flat to the ground on courses that have undulating terrain and a bunch of turns. Can a steel tape be tensioned at half, or 1/10 of it's length?

I believe a chain can be, due to the multiple links. Not sure about a tape

I would think any of these measuring methods would be a lot more accurate on a flat, level surface than something undulating on hills.

That is probably another reason why there was not discussion on pace and length back in the day. We didn't have many flat, fast courses. It was understood that XC included hills.

Have you ever seen anyone measure a course with a chain? How would that be stretched? What length could be used?

The accuracy of the Jones counter lies in the calibration method which involves taking calibration measurements before and after the course measurement. That method also assumes that the Bike/Jones counter will maintain the same precision throughout the length of the course. This is a safe assumption for road races where the paved surface is uniform from start to finish.

The problem with XC courses is that most are on grass, gravel, dirt, etc. Most XC surfaces are NOT uniform throughout, and you will get different readings from thick grass, thin grass, and gravel. This makes calibration

of the Jones counter tricky and unreliable for XC courses.

The evidence that I have are the measurements made during the "Illinois measuring marathon":

2.9758 miles, steel tape measured by On Course Rating Systems

2,9717 miles, calibrated Jones counter measured by a USATF course certifier

2.9568 miles, un-calibrated Home Depot wheel, measured by DiscoGary

2.9502 miles, un-calibrated "expensive" wheel, measures by DiscoGary

All wheeled measurements followed the string of flags left behind by the steel tape measurement crew so the path was very well defined.

The Jones counter came up 20 feet short, which is pretty good, but it isn't the 8" per mile accuracy that you are used to on the road, and wouldn't be good enough to certify a record performance.

Eliminating all but the relevant portions...DiscoGary did you consider calibrating the measuring wheel at all? Be interesting to know what the did for the accuracy.

Keith Stone wrote:

Math Man wrote:

You mean 8 FEET per mile?

No I don't, good catch. It should be .000126. Forgot to multiply 5280*12. That's what I get for using my phone instead of a spreadsheet.

Are you sure, or have you been off by 8 feet a mile all these years and didn't know the difference?

Not trying to be a jerk, I think being within 8' a mile would be amazing especially on a cross country course. Your claim of 8" a mile seems hard to believe with turns and tangents and duplicating such a ride seems extremely difficult especially when there is that much play within ONE wheel revolution.

For anyone who cares, here's how I would maintain a XC if I couldn't embed markers in the ground.

1. Setup the course for the desired distance, but leave the finish line undefined.

2. Teach a couple young guys how to measure with a steel tape.

3. Have them measure the course out to the desired distance, and then place the finish line at that point.

4. If you want to double check, then have them measure again from finish to start.

5. Next year, do the same thing.

This way even if you have to change the course path from year to year, you can at least claim that the general terrain remained the same, and that the distance did not change. Those are the two most important factors in maintaining a XC course.

I would argue that the 20' difference over an almost three mile course could very easily be user error(slight variance in actual course measured). The ONLY way to do a TRUE comparison is to set up and leave ALL of the steel tape segments on the measured course, THEN use the bike, and wheel and TRY and measure ON the actual steel tape segments. Tell me THEN what you come up with from measurements with properly calibrated devices and we have a REAL study. That is science.

Mr. Obvious wrote:

Eliminating all but the relevant portions...DiscoGary did you consider calibrating the measuring wheel at all? Be interesting to know what the did for the accuracy.

We didn't report the wheel calibrated results because we wanted numbers which would represent the kind of historical measurement that most people would get with the wheel.

BUT.... I have the original measurement spreadsheet.... and I have the calibration measurements we made for the Home Depot Stanley 77-194 wheel... on both the soft grassy sections -3.4 meters per 300 meters, and the flat dirt firm sections -5.7/300... SO .... averaging those out to -4.5/300, and using that average to calibrate the final result we get:

300/295.5 * un-calibrated distance = 3.0018 miles for the wheel

or 0.0260 miles longer than the gold standard, or about 137 feet long

Not good enough for course record purposes. Also note that the wheel slipped more on the dirt than on the grass. The wheel behaves in unexpected ways which is what the On Course people warned us about. The difference in calibration numbers between the two parts of the course is enough to disqualify the wheel from being a useful course measuring tool. There's just too much uncertainty.

The "expensive" wheel was a Meterman Pro200. I do not have calibration data for that wheel.

Math Man wrote:

Are you sure, or have you been off by 8 feet a mile all these years and didn't know the difference?

Not trying to be a jerk, I think being within 8' a mile would be amazing especially on a cross country course. Your claim of 8" a mile seems hard to believe with turns and tangents and duplicating such a ride seems extremely difficult especially when there is that much play within ONE wheel revolution.

If you're not trying to be a jerk that's certainly a borderline jerk comment.

You're clearly not familiar with how the measurement process works because it's not like I'm the only guy that checks the numbers. Yes I'm sure, and NO there's not that much play in one wheel revolution. You'd have to stop and back up to introduce "play" because it's not like any functioning counter is skipping a couple clicks each revolution.

The 8" number comes from the fact that a "click" is roughly 4" depending on the device and bike and a quick survey of several course paperwork spreadsheets showed a click differential of 6 mostly on 5K courses that 2 clicks comes out to roughly 2 clicks per mile. 8' per mile would be outside of RRTC/USATF guidelines. When I look at the varience on a 1000ft or 1/2 mile calibration course the counts are often identical, and if they were off by anywhere near what you're claiming any measurer worth their salt would recalibrate and re-ride.

I could go off an re-calculate and publish the numbers like DiscoGary (good stuff BTW), but it seems a waste of time. You'll notice his numbers ON A XC COURSE show a variance from a tape measurement of 20'. That's comparing two measuring methods, what he doesn't show what happened when a skilled measurer road the course twice and compared THOSE measurements, and that's what I'm speaking of.

In any case I don't claim to have done it on a XC course, but I have done it on more occasions than I can count on roads and mixed terrain. What I am saying is that if I can repeatedly get that close when I've done only 40-50 courses over the years someone that does 20-25 a year (like skilled measurers often do) could do that on an XC course. Also, I don't believe that a wheel is inherently more accurate that a calibrated bike. Both a wheel and a bike would be subject to surface variations

Math Man wrote:

I would argue that the 20' difference over an almost three mile course could very easily be user error(slight variance in actual course measured). The ONLY way to do a TRUE comparison is to set up and leave ALL of the steel tape segments on the measured course, THEN use the bike, and wheel and TRY and measure ON the actual steel tape segments. Tell me THEN what you come up with from measurements with properly calibrated devices and we have a REAL study. That is science.

I already said that the wheeled measurement were taken by following the flags put down by the tape measurements. Everyone keeps making the mistake of assuming that minor deviations from a straight line greatly increases the distance traveled.

Let's assume that we performed the wheeled measurements while completely drunk and veered 2 meters off to the side between each flag 50 meters apart. The total additional distance would be:

root of 25^2 + 2^2 = 25.0799 meters

25.0799/25 * 2.9758 = 2.9853 miles, or 50 feet long

Any weaving should have ADDED to the measurement, but all the wheeled measurements came up short, so any length added to the measurement due to weaving was more than offset by the slipping wheels.

A better assumption is that we were well within 0.5 meters of the tape line:

2.9763 miles, which would be 3 feet long. Not worth worrying about.

Just in case no one has seen this or posted it. This explains it, minus all of the distance-measuring technicalities

http://nc.milesplit.com/articles/167373/meet-the-new-course-same-as-the-old-course-mcalpine-gets-an-adjustment

People don't know about the process of calibrating the Jones counter for road races. They would feel a lot better about certified distances if they knew. You should go through it. USATF Certified means something.

DiscoGary wrote:

People don't know about the process of calibrating the Jones counter for road races. They would feel a lot better about certified distances if they knew. You should go through it. USATF Certified means something.

I agree most people don't understand the process, and more (particularly elite and sub-elite) people should. Wandering all over the road on a road race is like running track in lane 3. No serious runner would do the latter, but I see serious runners do the former all the time.

I don't understand the "You should go through it" comment. As I noted I've gone through it 40-50 times myself plus numerous times helping others. Perhaps you meant "everyone should go through through it", and on that I would agree.

I remember years ago before I remeasured a local half marathon I did a casual ride of the course first just to get a feel for the tangents. I didn't track the splits but I wrote down the start/finish counts. After I rode the course "for real" twice I compared the no-tangent ride to the tangent rides and the overall difference was nearly 40 meters. May be only a few seconds, but I mentioned that to someone that missed his OT marathon time by 8 seconds and I could see him doing the math in his head.

brianru wrote:

Just in case no one has seen this or posted it. This explains it, minus all of the distance-measuring technicalities

http://nc.milesplit.com/articles/167373/meet-the-new-course-same-as-the-old-course-mcalpine-gets-an-adjustment

They used a calibrated wheel to do their measurements. They are not getting the accuracy they think they are getting.

I meant that you should explain the road race measurement process right here in this thread.

People can imagine a dozen ways how it could go wrong, so they won't believe anything until they understand the process.

DiscoGary wrote:

I meant that you should explain the road race measurement process right here in this thread.

People can imagine a dozen ways how it could go wrong, so they won't believe anything until they understand the process.

Ahhh OK. I wrote something up some years ago when I was training some people, I'll see if I can find that. If not I'll put together a synopsis when I have some time. Not today though, there's a box of Christmas lights my wife dug out with my name on it.

In the meantime I would encourage anyone interested in geeking out to look at:

The list of names there is some guys that really know their shit. I learned how do it right when a couple of those guys taught a seminar in course measurement at an RRCA conference over two decades ago. I learned the dozen ways it could go wrong through personal experience.

Thanks for the response - but how about: How many stakes are used to create a smooth line to measure curves with a tensioned "cable"?? Maybe DiscoGary could pipe in on this as well?

Using a bike to measure distance is easy and 'very' accurate. The basic principle behind a bike odometer is a magnet mounted on the spokes with a sensor on the frame which measures how many times the wheel goes around as in how many times the magnet contacts the sensor. You calibrate it by moving it up or down the spokes, which translates to distance. In the old days, this was purely mechanical, nowadays, bike odometer are electronic and use the same principles to determine speed as well. Most bike shops have a computer that will calibrate your bike/odometer. You can also calibrate your bike by using a football field and/or with basic geometry. Beyond the math, the beauty of a bike is that the wheels contact the ground with constant pressure and they are not capable of making sharp turns, just like a runner. Bikes are so accurate, that they have been used to measure and/or verify big-city marathon courses.

I have also used the old Nike+ for the ipod nano and the Livestrong

pedometers for distance measurements. They are fairly accurate if you put in a lot of time to get the perfect calibration. I am by no means an expert, but when you have designed XC courses for AAU and USATF youth competitions, you get pretty good at measuring. You are sometimes laying out courses where one did not previously exist and because different age groups run different distances, as a result, you end up designing several course for one meet.

With all of that said, I reiterate that the original McAlpine course was not short by an appreciable amount and mountain bikes are excellent tools for measuring a XC courses. Btw, to truly understand the fact times at McAlpine, you need to come see a course with only 2 appreciable turns and several 400m straightaways on compacted dirt.

Keith Stone wrote:

In any case a good biker is going to hit both measurements within 1 or 2 clicks of the Jones counter on a 5K, which is a variance of 8" either way, certainly not "precise" but I'd wager it's well beyond the abilities of most walking wheel users on an XC course.

For sure, electronic survey equipment with tape/cable/chain for the curvy parts could be precise but let's face it, .008 is good enough for world records so it should be good enough for cross country.

based on what you just posted and your ".008" figure, even the furthest off measurement in the Illinois State Meet measurement would be "good enough for world records"...are you sure about this figure?