The Stability of the Bicycle, David Jones, 1970.
pdf file article
The Stability of the Bicycle, David Jones, 1970.
pdf file article
J.R. wrote:
A bicycle wheel is more likely to keep in a straight line, because of momentum, because of having a constant weight on the wheel, and because the bicycle helps to keep the wheel straight. This is why using the front wheel on a bicycle is more accurate than pushing a wheel.
Sure you could put marks on any wheel, the point being that this does not change the accuracy of the measurement. When the wheel is off 1/2 a meter, it's off 1/2 a meter. When the Jones counter gets the same exact number of counts each time, then the measurement is the same, there is no deviation at all in the measurements.
Great Thread! I have to believe that there are good measurers that can keep a bike wheel straight and there are good measurers that can keep a push wheel straight. Using the push wheel that measures 100.5m for every 100m is no different than using the bike clicker and getting 1208 clicks for every 100m? Or am I missing something?
Isn't 5025 on the Push Wheel the same as 60400 on the Bike Clicker? Both laying out a 5000m route?
J.R. wrote:
Look up "The Stability of the Bicycle", by David Jones.
Here is a video that shows an example of David Jones' experiment, note the bigger the trail of a bicycle the greater the stability (and harder to turn). This is why a touring bicycle is better for course measurement than a racing bicycle.
I don't have more exact figures from my own testing as it's been awhile, but the former should have a more normal trail of 2 inches, where a racing might be 3/4 inch or less. This makes a racing bike harder to ride, though some have caught on (like Armstrong) and use a greater trail for greater stability.
Try this with a wheel. :)
http://www.youtube.com/watch?v=PXRQdWG9FuMThe stable/unstable bicycle
http://www.dclxvi.org/chunk/tech/trail/
Great Info J.R.
But if you take a Surveyors Wheel and remove the count mechanism, the wheel will stand up right. The wheel is round but the base is flat, or squared off. This helps maintain a flat and straight measurement. Take one of these wheels and they will roll as straight or straighter and longer than any bike when sent ghost ridin'.
Could you ride your bike around a high school track on the curve with your front rim not leaving a lane line? This process is done with a wheel on all track painting projects to measure staggers, exchange zones, etc. I'm sure for your product is excellent, but their are other products that are extremely accurate if you know how to use them.
Math is Easy wrote:
Devise MW = 10cm
Devise JC = 4.5"
10cm < 4.5"
Devise MW is more accurate
My J-R Counter ~= 4550 counts per 440 yds is about 3.48 inches per click, which is less than 10cm (about 3.93 in).
Another Coach wrote:
100.5m CONSISTENTLY proves my point. With the Jones Counter wouldn't you be off several clicks each ride and need to take an average, then recalibrate or check again on the calibration course, then add the SCPF? With that wheel I would think rolling out what you calculate in feet to equal 5025m add 5m for your SCPF and would you really be off by much?
No, when I calibrate my bike on a 300m cal course, the difference between rides is rarely more than 1 count.
The point is that 99%(or more) of the time a wheel is used it is not calibrated. Have you ever heard of a cross country coach calibrating their wheel before they measure a course? I certainly haven't. And you have to calibrate it every time before you use it on the same surface you will be measuring. I also calibrated that wheel on a golf course fairway and got an average closer to 99m.
But you're right. If you do actually calibrate your wheel before you use it to measure a course you'll probably get a pretty accurate measurement. I don't think it's going to be more accurate that using a calibrated bicycle, but regardless, it's going to take a WHOLE lot longer. The calibrated bicycle is considered the best method because it gives the best accuracy for the time it takes to do the measurement.
I have read over 7 pages of this thread and most posters are dancing around the real issues of accuracy.
I think it's a waste of time to compare a surveyor's wheel with the bike method (mounted with a JR counter). Neither device will yield an accurate measurement for a cross country course
Why? (as already stated by several on this thread)-
1. Cross country courses have various surfaces. The wheel or bike would need to be calibrated for each of those different surfaces and corections would need to be made for each segment of the course as the surface changes.
2. Just stand behind anyone pushing a surveyors wheel or riding a bike on anything but a paved surfaces and you will see the undulations in the wheel/bike off the true line of the course. It's tough enough to maintin that line on a paved road, let alone grass, gravel or rocky surface.
3. I have also read here that the bike method produces the most accurate course for the amount of time measured. Yes, it's faster but not accurate.
The bottom line:
1. Cross Country Racing is the purist form of racing. You have a start and a finish. The runners all get to review the course. The gun goes off and it's the first person to the finish line. The accuracy of the course is secondary.
2. Stop trying to compare one cross country course to another - apples and oranges.
3. And if you really care about a course being accurately surveyed, get out there with a team of people, a few dozen stakes, hammer,a good steel tape and spend a few days of HARD work and do it right. The athletes spend a lot more time then that training to run on the course.
Too many people are looking for a quick and simple way to do a difficult job....forget it!
If anyone wants to assist I would be more than happy to go out and measure Van Cortlandt (for free). but be prepared to spend all day or 2 or 3.
P.S. Just got in from measuring a 5 Mile race in Cnetral Park. It was a beautiful evening. Three measurers worked during the night after most runners/bikers had left the park. We all came up within a meter of each other after 5 miles. Our pre-calibration and post calibrations were within 1 count. Life is good.
I was thinking the same thing, at least with respect to the OCRS paper in one of the early posts. The problems with the different surfaces was that the wheel they used slipped too much on some surfaces. I guess using a Jones Counter on a bicycle, in addition to any inherent extra accuracy or improved calibration techniques, would slip less, simply because of the extra weight of the rider applying a greater normal force.
By all means, calibrate a click wheel if you want, but solve the wheel slipping in long wet grass issue first. Should we debate about 4" or 10cm of accuracy, or bias, or precision, or systematic errors, when the click wheel is routinely measuring 80m - 100m short, because it's slipping in wet grass?!?
On a slightly related topic, 5 years ago I bought a Polar S625X HRM, with a footpod for measuring speed and distance.
While I still have a slight personal preference for the footpod over the GPS, (because I often run in heavily wooded areas), I was nevertheless disappointed with accuracy issues, and the need for frequent calibrations for different shoes, running surfaces, and even running speeds.
David-
Thanks for your replies on this topic. It's great to have a seasoned professional share their insight. With that being said could you expand on your responses?
When you mention neither a surveyor's wheel or the JR counter will yield an accurate measurement. Please define accurate. If YOU were using either devise that was properly calibrated and measuring a CC course that was firm with short grass, how much would you suspect to be off the true distance?
1)When you use a steel tape in tall grass do you let the tape lay on the grass or somehow tack it down and stretch it to the proper tension to insure accuracy?
2)Would in not be true that the more experienced the measurer, the more likely the wheel would remain more straight, or on line with what was attempted to be measured. Is it not also true that the side to side deviations are minimized at speeds in excess of 3.5 mph?
3) Again, you mention being accurate. The JR Clickers swear accuracy to one click per km/mile/etc. Is it possible to be that accurate over that distance with any device including your steel tape method?
Bottom Line Questions:
1)I like your view point here but again I think the definition of accuracy is key. Why make this statement then blast traditional measuring procedures?
2)I'm not sure too many people are doing that. I think it would be nice to see actual distance measured on the results page. When I see EVERYONE running big PRs I think short course, not fast course. It would be great if the results listed 7918m Steel Taped by David Katz 10/27/10 and permenant markers were set on courses that allowed such.
3)On a curve, how many stakes are needed, and how much distance can you measure at a time to get the shortest possible route and get the curve to remain smooth equaling the proper distance with a steel tape? What do you do on high school courses on the curves where there is no outside natural boundary? Also, how much pressure do you need on your tape to insure accuracy? What do you do at every high spot and every low spot? Do you stretch the tape over these undulations in the running surface or use something to keep the tape in the low spots?
So when you measured the 5 mile course in Central Park, did you use the JR Counter or a steel tape? Did you follow USATF rules on curves or use what seems very liberal and arbitrary and go with the "sensible line"?
Thanks in advance for your responses and contributions to this topic.
rekrunner wrote:
I was thinking the same thing, at least with respect to the OCRS paper in one of the early posts. The problems with the different surfaces was that the wheel they used slipped too much on some surfaces. I guess using a Jones Counter on a bicycle, in addition to any inherent extra accuracy or improved calibration techniques, would slip less, simply because of the extra weight of the rider applying a greater normal force.
By all means, calibrate a click wheel if you want, but solve the wheel slipping in long wet grass issue first. Should we debate about 4" or 10cm of accuracy, or bias, or precision, or systematic errors, when the click wheel is routinely measuring 80m - 100m short, because it's slipping in wet grass?!?
The big problem with that study was actually the wheel they used. The Rollatape brand has a reputation of being highly inaccurate. Any Click Wheel is already questionable because of the one click per revolution, and has the potential for error with "over spin" the quicker it is pushed. In all surfaces the brakes are hit everytime the clicker makes contact with the counter, which is probably why there is more error in wet surfaces - in addition, proper pressure must be applied. I don't believe handle pressure was ever taken into consideration in that study.
rekrunner wrote:
On a slightly related topic, 5 years ago I bought a Polar S625X HRM, with a footpod for measuring speed and distance.
While I still have a slight personal preference for the footpod over the GPS, (because I often run in heavily wooded areas), I was nevertheless disappointed with accuracy issues, and the need for frequent calibrations for different shoes, running surfaces, and even running speeds.
The foodpod looks like quite an interesting device, though quite pricey, and even with that it's only an accessory, so the real cost is... $200?
You bring up great points about different shoes affecting the stride. Also there is a major difference between barefoot running and with shoes. I wondered if the footpod would be accurate for walking, vs running easy vs fast, and apparently it is not.
There is a specific relationship between rate and length of stride, with increasing speed, which varies by person. For me it is approximately a 2/3's relationship. This could be calibrated relatively accurately, were it designed accurately into a device for that purpose. Then there is the difference in footwear, but that could be accounted for too, for example by adjusting the calibration for the different footwear.
So far I think there is no device that comes close, so I've not bothered to look into them much. I think even if there was an accurate stride count device for a reasonable price, then it could be useful, but I've not even found one of those yet.
Way to go Surveyor Steve - way to kill the thread!
Looking for more Info wrote:
Way to go Surveyor Steve - way to kill the thread!
I'm sure David will reply when he gets a chance. Responding to Let's Run Threads is probably not their only task in life.
It is surprising that no others have chimmed in though.
Steel tapes are useful for setting up calibration courses, as these are straight line measurements.
I see no reason why they would be any good for measuring cross country courses though, as such courses are not in straight lines. Apparently some people use them but I have no comment about that.
Also I've not used a surveyor's wheel. Are they completely metal or are the tires made of rubber? Have you given one a push down a long hill to see how far it will go on it's own?
This reminds me of when I pushed a couple of car tires down a 100 meter hill when young, the stables, horses and my dad at the bottom. The tires were really cool zooming down the hill on their own, but the horses freaked out, were jumping around and my dad had a fit. Geesh the trouble someone can get into just for pushing a couple of tires down a hill.
I agree about the steel tapes. Can't imagine spending 2-3 days to make a 5k course 9 meters longer than I would get with my wheel or GPS in 20 minutes.
David Katz wrote:
Just got in from measuring a 5 Mile race in Cnetral Park. It was a beautiful evening. Three measurers worked during the night after most runners/bikers had left the park. We all came up within a meter of each other after 5 miles. Our pre-calibration and post calibrations were within 1 count. Life is good.
That's great. My best friend, and a couple others I knew were in the group (of 17?) who measured the course for the '84 OG marathon. I passed on the opportunity, though it would have been an interesting experience, and a chance to have met some of you posting on this thread. Funny, as that was quite a long time ago, 26 (miles) years!
J.R. wrote:
David Katz wrote:Just got in from measuring a 5 Mile race in Cnetral Park. It was a beautiful evening. Life is good.
That's great. My best friend, and a couple others I knew were in the group (of 17?) who measured the course for the '84 OG marathon. I passed on the opportunity, though it would have been an interesting experience, and a chance to have met some of you posting on this thread.
I was a part of the team that (re)measured the '96 Oly Games marathon course and DK was my roomie. He seemed such a regular guy at the time...I'd no idea he was such a nerd. But he's a nerd in great demand (and who's passion I share & admire).
I lived in LA at the time and ran for Culver City AC and sort of knew the folks doing the marathon course.
J.R. wrote:
David Katz wrote:Just got in from measuring a 5 Mile race in Cnetral Park. It was a beautiful evening. Three measurers worked during the night after most runners/bikers had left the park. We all came up within a meter of each other after 5 miles. Our pre-calibration and post calibrations were within 1 count. Life is good.
That's great. My best friend, and a couple others I knew were in the group (of 17?) who measured the course for the '84 OG marathon. I passed on the opportunity, though it would have been an interesting experience, and a chance to have met some of you posting on this thread. Funny, as that was quite a long time ago, 26 (miles) years!
The additional cost of a footpod is usually around $100-$120.Since it's an accelerometer, it's sensitive to anything that changes acceleration of the foot, like cushioning in shoes, and hard surfaces versus soft surfaces. It also doesn't know if the wind is slowing you down or speeding you up, or if you are slipping in sand.The accelerometer assumes the foot is flat when it strikes the ground, so it is sensitive to changes in foot strike caused by speed changes, e.g. toe-striking while sprinting versus mid-foot or heel-striking during recovery runs. POSE runners may also have trouble getting consistent accuracy.Polar's implementation in the S625X had a specific issue, I believe while attempting to filter out bad samples due to transmission errors, that delayed detection of speed changes by about 10-15 seconds. So each rep in interval training measures short, or the first lap in a race on the track measures short. There are a handful of other "software errors" which further affected accuracy (introducing errors, or dropping significant digits), which I won't go into here, but these are not inherent in the footpod.With some care during calibration, you can minimize errors to around 1-2%, which may not be suitable for competitions, but certainly for logging new routes in uncharted forests.
Having briefly looked at all 8 pages I am still amazed that anyone would want to accurately measure a cross country course. Do you people have any idea what you are saying????????? Learn the history.