Road Course Measurement Bulletin Board    Forums    Course Measurement    General Course Measurement Topics    Using Larger AND Average Constant
 Go New Find Notify Tools Reply
 Using Larger AND Average Constant

 posted 19 August 2008 22:26
I just measured a marathon over the course of several days and five separate measurements. When figuring out the numbers, I worked out the segment distance using both larger and average constant. Then I used the sum of shortest splits to determine the overall length.

My question is this: Is it OK if some of the shortest splits were derived from the larger constant, and some from the average, or is that like mixing apples and oranges? If nothing else, it should result in the shortest possible distance, but is it TOO short?

 Posts: 743 | Location: Norwalk, CT | Registered: 24 October 2004 IP

 posted 19 August 2008 23:09 Hide Post
Jim:

This is similar to my post a month ago. I don't know why you want to use a mixture of constants, but if you are sure your measurements are accurate, I see no harm. However, since consistency is the hobgoblin of small minds, I would chafe at the inconsistent use of the larger constant.

In my case I measured 10 segments, all twice. The 2 rides of each segment were below the required 0.008 (.8%) of each other, in fact they were all better than 0.0005 with some as low as 0.0001. The use of the larger constant resulted in a course 59.5 ft longer than the same course using the average constant. In my case there was a large temperature difference between the times of measuring the constants and I thought (like a runner) that the extra 59.5 feet wasn't needed since the rides were well within the rules for comparison.

Since I received many warnings about the chance of a short course, I caved and used the larger constants.

 Posts: 215 | Location: Connecticut, USA | Registered: 23 November 2004 IP

 posted 20 August 2008 06:19 Hide Post
Jim,

It's hard to tell from your post just when you used average and when you used larger. A 10 km measured using average constant will be shorter than one measured using larger constant, but both will be 10010 or more if measured properly. The more you use larger constant the longer your course will be.

Sum of shorter splits will also tend to lengthen the layout, but I use it a lot because of the difficulty in deciding which is the first ride and which is the second, when I lay out a course over several days of discontinuous measurements.

I mixed these apples and oranges on a recent marathon/half marathon measurement. Both courses were to have the same start and the same finish. A turn-around was to be avoided if possible.

I measured the half marathon first, using larger constant, and established the starts and finishes. A bit of trying some alternate routes was needed before I could get the marathon to come out right. Using larger constant the marathon come out to 42190 meters - 5 meters short. But by average constant it was 42212.

I decided that there was no need to add distance.

Pete Riegel

 Posts: 1756 | Location: Columbus, Ohio, USA | Registered: 23 October 2004 IP

 posted 21 August 2008 06:22 Hide Post
Pete,

I agree that the course will be long by at least the SCP if the average constant is used and the course was measured properly. Its the "measured properly" that I am trying to fix in my head. Two rides within 0.008 of each other is a good sign of "proper measurement" provided the measurer (or both measurers) didn't make the same mistake twice. Assuming honest measurers is there any other indication of "proper measurement"?

 Posts: 215 | Location: Connecticut, USA | Registered: 23 November 2004 IP

 posted 21 August 2008 06:29 Hide Post
There is no assurance of truth in submitted paperwork. The only check we have is the occasional validation.

We ask questions, which if truly understood and truly answered, lead to an assurance that the proper route was measured.

Beyond that, what can be done and still get the work done? Validation history shows that we get it right about 90 percent of the time.

Pete Riegel

 Posts: 1756 | Location: Columbus, Ohio, USA | Registered: 23 October 2004 IP

 posted 21 August 2008 06:39 Hide Post
Thanks Pete, I guess I was looking for a mathematical reassurance of my accuracy.

Pete

 Posts: 215 | Location: Connecticut, USA | Registered: 23 November 2004 IP