It has been brought to my attention that our Course Measurement Manual still says a calibration course "...must be at least 300 meters in length. Greater accuracy can usually be achieved by using a longer calibration course...". Also, "Most calibration courses are at least 500 meters in length. This is recommended...". It goes on to say that a 500 or 1000 meter cal course is desirable.
There has been discussion in the recent past in which I thought it was agreed that a 300 meter calibration course was an acceptable standard, and that no increase in accuracy was observed in longer courses. If this is the case, I would ask if we can change the manual to reflect the acceptance of a 300 meter calibration course, and not cause people consternation if they can't find a straight, level stretch of road longer than 300 meters.
Also, why recommend longer courses, since they take more time to set out, and ride before and after the measurement of race courses?
Is there a reason we shouldn't edit the manual to reflect what was accepted in the prior discussion?
Length of a Calibration Course
Duane,
I remember that you made this comment at our 2010 Annual Meeting in Virginia Beach and, in fact, I included that comment in our Minutes of the 2010 meeting which can be viewed at thread_1754046387.html I think people agreed with that comment, and it's a shame that we never followed up on it. The issue now is exactly how to change the language in the manual. As a possible rewording, The first two paragraphs of the text currently at http://www.usatf.org/events/co...anual/cal-layout.asp might be reduced to the single paragraph:
How does this look?
Meanwhile, the IAAF Measurement manual which can be downloaded at http://aimsworldrunning.org/course_measurement_p.htm should also be revised, as it includes the language, "It should be at least 300m in length, although a length of 500m is recommended."
I remember that you made this comment at our 2010 Annual Meeting in Virginia Beach and, in fact, I included that comment in our Minutes of the 2010 meeting which can be viewed at thread_1754046387.html I think people agreed with that comment, and it's a shame that we never followed up on it. The issue now is exactly how to change the language in the manual. As a possible rewording, The first two paragraphs of the text currently at http://www.usatf.org/events/co...anual/cal-layout.asp might be reduced to the single paragraph:
quote:Accuracy of the calibration course is vital since any error will be multiplied when it is used for measuring a race course. A calibration course must be on a straight, paved, level, and lightly traveled stretch of road, and must be at least 300 meters in length. Accuracy is also generally best if you can minimize the time required to transport your bike between the calibration course and race course. Therefore, you should consider laying out a calibration course close to the race course to be measured, especially when you must travel a long distance to reach the race course site.
How does this look?
Meanwhile, the IAAF Measurement manual which can be downloaded at http://aimsworldrunning.org/course_measurement_p.htm should also be revised, as it includes the language, "It should be at least 300m in length, although a length of 500m is recommended."
I like the revised text, but I’d suggest eliminating the word “level” from the requirement.
Most calibration courses will wind up on a reasonably level road anyway. Also, how level is “level?”
There are often good reasons for choosing a non-level calibration course, such as safety and availability. The calibration course used for the measurement of the 1996 Atlanta Olympic Marathon was not level. It had a hump in the middle. It was chosen because there was nowhere in the start-finish area that had a level road.
I think, but am not certain, that two-way riding of the calibration course eliminates most of the potential inaccuracy that an uphill or downhill calibration course might cause.
Most calibration courses will wind up on a reasonably level road anyway. Also, how level is “level?”
There are often good reasons for choosing a non-level calibration course, such as safety and availability. The calibration course used for the measurement of the 1996 Atlanta Olympic Marathon was not level. It had a hump in the middle. It was chosen because there was nowhere in the start-finish area that had a level road.
I think, but am not certain, that two-way riding of the calibration course eliminates most of the potential inaccuracy that an uphill or downhill calibration course might cause.
I agree with Pete here. I will forward this information to Bernie Conway and David Katz for the IAAF.
Maybe, change the text from level to reasonably level. I think the intent was to prevent creating calibration courses on hills.
Thank you. -- Justin
Thank you. -- Justin
I have asked Bernie Conway's what he feels:
Bernie’s comment: I agree that removing the word "level" from the description of a calibration course should be removed but that a statement similar to the following "the measurer should make an effort to measure the calibration course on a level street but if that isn't possible then he/she is to calibrate his/her bicycle by riding in both directions".
Bernie’s comment: I agree that removing the word "level" from the description of a calibration course should be removed but that a statement similar to the following "the measurer should make an effort to measure the calibration course on a level street but if that isn't possible then he/she is to calibrate his/her bicycle by riding in both directions".
I second Justin's suggestion of "reasonably level". That gets the point across.
I don't like Bernie's suggestion, as that seems to insinuate that riding in one direction for the four rides is an acceptable method. I don't think we should confuse the issue by seeming to accept four one-direction rides. This may be considered if someone sets up on a street that has traffic on it, and, instead of riding against traffic for two of the rides, figures they can ride back to the start on the opposite side. Yes, it makes the total calibration riding longer, but that defeats the intent of (as Pete points out) riding both ways to cancel-out the differences in elevation and/or wind.
I don't like Bernie's suggestion, as that seems to insinuate that riding in one direction for the four rides is an acceptable method. I don't think we should confuse the issue by seeming to accept four one-direction rides. This may be considered if someone sets up on a street that has traffic on it, and, instead of riding against traffic for two of the rides, figures they can ride back to the start on the opposite side. Yes, it makes the total calibration riding longer, but that defeats the intent of (as Pete points out) riding both ways to cancel-out the differences in elevation and/or wind.
The language has been updated in our online Course Measurement Manual. In the Laying Out a Calibration Course chapter at http://www.usatf.org/events/co...anual/cal-layout.asp the new language is what I proposed earlier in this thread, except that "level" has been changed to "reasonably level." It was also necessary to add the word "reasonably" in the Statement of Requirements chapter at http://www.usatf.org/events/co...ual/requirements.asp The new language is also in the PDF version that you can download from http://www.usatf.org/events/co...on/manual/manual.pdf
IAAF has also agreed to update the language in their manual, when it will be revised sometime later this year.
IAAF has also agreed to update the language in their manual, when it will be revised sometime later this year.
Thanks, Bob! Much less ambiguity.
On another note here, the IAAF has adopted the same language.
I would like to add to this thread with a situation which arises once in a while. If a race is being held on a unpaved surface, more accoracy can be obtained if the calibration course has the same surface. I recently had a message fron a new measurer who is planning a race totally on dirt roads. I suggested that he lay out a calibration course on a dirt road.
I'm not sure that this has to be addressed in the rule .... but.
I'm not sure that this has to be addressed in the rule .... but.
Paul, I think this is a very valid point. I do think it should be mentioned in the guidelines, that if the course is to be primarily on an unpaved surface, the calibration should be done on the dominant surface, as this increases the accuracy of the measurement.
Folks who are following this thread may also want to look at another thread under general measurement issues, called "threshold for unpaved portions". I posted some calibration numbers for an out-back 20 km I recently measured on the C&O Canal towpath. I calibrated on road and on canal surface. The numbers convince me that, as Paul and Duane are saying, it is preferable to use a calibration course with a similar surface as the course, BUT using a road calibration would still be "safe" (not short) without making the runners run way too far. Someone else might interpret those numbers differently though.
However I would be wary of advising someone to use a "rough surface" for calibration if a significant part (I don't know, more than 20%??) of the course happened to be a smoother surface
However I would be wary of advising someone to use a "rough surface" for calibration if a significant part (I don't know, more than 20%??) of the course happened to be a smoother surface