Expected Longer Recal Tactics

Duane Russell creates a temporary cal course at the race course site. The course is just 2 paint parks about 300m apart (the exact distance doesn't matter). He does his official calibration at home, but then when he gets to the race course he rides this temporary course and records the counts. Then when he finishes the race course measurement he rides the temporary course again, and calculates the percent change in calibration constants from the pre-temporary-cal-course ride. He uses that percent change to estimate what the change in the official constant will be when he gets home and does the real post cal. He uses that to lengthen the course if needed.

Note that you don't need to actually measure this temporary course, and you really need to ride it only 1 or 2 times pre and post cal since you are using it just for an estimate.

Duane can add details to this if I missed something.

I always like to have a cal course close enough so that I can adjust the course, if needed, after the post calibration ride.  It doesn't take long to measure a new calibration course, as long as you have a helper, and then it is available for future measurements in the area.

I would not recommend relying on someone else to adjust the course.

Laying down a new cal course near the course to be measured just isn't always practical, which is how Duane came up with his innovative solution. He often travels long distances from home to measure.

I once posted here about a measurement I did on a hot summer afternoon when I barely finished my second ride as a big T-storm rolled in. The cal course was only about a half mile away, but I had no thought of re-calibrating in a violent storm. When the storm finally passed, I did the post cal in temps that had dropped more than 25° F. The numbers were shockingly different. I would have had to re-measure and move the timing points. Surely most of us have experienced a situation like Oscar's, when the temperature has fallen significantly by the time we arrive at a cal course after measuring.

When I considered that the temperature had varied only a couple of degrees or so during this measurement, before the storm, I decided that I would stick with my pre-cal numbers and discard the post-cal numbers - and not even use the average. My pre-cal was done in the less-hot midday. My two measurements for this 8K were within a few counts. I aver that this course would stand up to the strict protocols of a validation measurement.

I am not bragging about this. I only started getting deltas this close after many years of measuring. I think the relevant point is how much the temperature varies during the measurement. Duane's solution make great sense to me. I would go so far as to say that, unless we believe we have every technical consideration all noodled out and forever settled, we should further analyze Duane's technique. If we can establish appropriate criteria for using his technique, I think we could all but do away with our traditional post-cal requirement. Being a realist as well as a dreamer, I do not expect many of us will agree with this.

Duane Russell's temporary cal course will not give the exact adjustment needed according to established written procedures.  I doubt an adjustment I made based on that would be accepted if the final calibration was longer.   

However, it raises a very interesting possibility:   Suppose you calibrate, drive 50 miles to a course and set up a temporary cal course.  Then you measure till noon, recheck calibration on that temporary course and there are FEWER counts than earlier.  After lunch, you have a flat.  Would the morning measurements be valid using the calibration constant?   There are enough pressure and temperature induced unknowns that you should be at least as well off as under ordinary circumstances.

Oscar

For the sake of argument; What was the difference between the 2 rides?  Since the course was 8K, could it have been measured again followed by a 3rd cal ride in the cooler temperature?  What would be the criteria (regulation) for throwing out the 2nd constant?  Isn't the impact of throwing out the 2nd constant completely separate from whatever circumstances are used to justify that action?  Duane is a final signatory and an experienced measurer, I'm equally experienced (another regulation dealing with level of experience needed here), but not a final signatory.  Would my certifier accept this measurement or the logic behind using a single cal ride?  Again, just for the sake of argument-no position taken or advocated.

I would be in favor of eliminating the requirement to do an official post calibration if Duane's method was used and it showed the calibration constant after the course measurement was smaller than the calibration constant before the course measurement.

If the post-cal using Duane's method showed the cal constant increased during the measurement, then an official post calibration would still be required.

I have never used Duane's method because well over 90% of the time the post-cal constant is going to be smaller than the pre-cal constant. So almost always it's a waste of time. In the very small number of cases when my post cal gives  a larger constant, I can drive back to the course to adjust. Having to do this 1 or 2 times for every 100 courses I measure takes less time overall than doing extra cal rides for every one of those 100 courses. However, if doing those extra cal rides eliminates the need for an official calibration when I get home for 98 or 99 of those measurements, then it's worth it.

I like Mark's idea.

I for one am not a fan of measuring in the dark due to safety considerations, as well as to my ability to sight tangents better in daylight. I have found myself finishing a measurement close to dusk, when the temperature has fallen below the temperature of the pre-cal, but only a handful of times over many years.

Experienced measurers may rarely find large differences between pre- and post- cal numbers, I believe. Usually, I have the same experience as Mark. My post-cal constants are smaller, due to the higher temperature at the end of measuring or by the time I drive to the cal course. In this case, I don't remember ever getting a surprise post-cal constant. There's no need to shorten a measured course in this scenario, IMO.

I have been using 100% nitrogen in my bike tires, which tends to reduce the expansion/contraction a little, though not as much as the solid tires that Duane uses. Still, for measurements that are done in constant or increasing temperatures all the way through, Duane's method seems logical to me.

I would appreciate hearing from more measurers, FS's, and certifiers about Duane's method as restricted by Mark's requirement. It seems to me it could result in more accurate measurements, as well as reducing unnecessary post-calibrations and unneeded shortening - and lengthening - of courses.

Have any certifiers chimed in on this?   Using  before and after "temp-cals" combined with other rationale (temperature change, travel time, course measurements difference) seems sensible as well as abusable.  Maybe the discussion should involve limits such as % difference between course measurements (0.0004 in lieu of 0.0008?) and a detailed explanation of the other circumstances that made an official post-cal impractical.

We have a cal course in Fairfield, CT.  It is 17-mi from Stamford, CT  After 2 measurements of courses in Stamford, we decided to layout a new cal course in Stamford since the 17-mi travel time anytime after noon is at least an hour.  There are 2 of us, making it practical to layout cal courses as needed.  We have 32 active cal courses in CT.  Since the combined age of the Guido Brothers is over 150-yr, and since it's always a goal to get, find, develop or recruit new measurers, maybe a relaxation of the post cal process would help.  I'm betting that laying out cal courses is an important reason why more people don't measure their own race courses.

IMO, establishing criteria for measurements needing no post-cal makes sense to me. The purpose of the post-cal is to bracket the extent of tire size and counts as road and air temperatures vary during measurement, and then to make adjustments to the course to eliminate the possibility of a short course. If the ambient temperature during measurement stays the same or increases until measurement is over, it seems to me a simple "Russell Check" to verify that the counter constant has not increased should suffice.

Unlike Mark, I often find higher post-cal numbers, due to the time of day I tend to start measuring - rarely in the early morning. If the temperature increases, then decreases by the end of the measurement ride, or fluctuates during a long measurement, such as a marathon measurement, a cooler post-cal could result in a third entire re-measurement. A situation to be avoided. In this instance, when temperature fluctuations during measurement (sun going in and out of heavy clouds, a passing thunderstorm) or a cooler (e.g. late day) post-cal are expected, clearly our pre-cal numbers aren't going to be as accurate as we would like. In this instance, using pre-cal or even an average constant could in theory result in an inaccurate, even a short course. Or, even if the overall measurement is accurate, some mile or kilometer splits could be less accurate than we might prefer. Who here hasn't run a race in which well-placed mile markers resulted in some of your split times being unexpectedly higher or lower?

Since we know the SCCF is in place to accommodate the probability of more "wobble" during measurement than during calibration, and of our necessity to swerve around potholes or other small obstacles on the course tangent while riding, would it make sense to increase the SCCF slightly to anticipate such temperature/constant fluctuations?

It is interesting to me that, calibrating over a wide span of temperatures results in such a small change for me in mile constant counts. In my experience, a range of ~ 17,985 in warm weather to ~ 18,030 in cold weather is about as much as I have seen. Maybe using 100% nitrogen in my tires helps contain the variation. I would guess that Duane, using his solid rubber tires, experiences even less. I would like to hear from other measurers what the range of their constants has been, given the same tire. My reasoning is that increasing the SCCF by some small increment could account for temperature fluctuations that could otherwise materially shorten a course. 

Consider the case of a marathon measurement: if the post-measurement temperature is lower than the pre-cal temperature by more than a few degrees, the inaccuracy mave well have accumulated over the day. Maybe, say, mile 10 is only 2 feet short of where a theoretical "perfect" constant-temperature would place it. But, as the temperature fluctuates, or decreases, subsequent miles are off, too. Over the next 16 miles, the error adds up to more than .08%. So, you have to measure the entire course again. Here, the same or a similar temperature variation could occur. All we have then done is repeat the "erroneous" measurement. In the instances when a post-cal (after the second ride) reveals a higher number of counts, we then move the Start, Finish, and/or a turnaround, which then makes at least one mile split - usually the last mile - a good bit longer, for a marathon, than any other. I know I have run many races on certified courses in which the last mile was a good bit longer than the rest. I am taking into account the possibility of fatigue and terrain in the last mile in making this assertion.

I would like to explore the non-trivial feasibility of changing the SCCF. We have established no official criteria for "too long". Historically, some validation measurements have revealed courses with a little "extra" distance - beyond the SCCF, of course. I do not have sufficient data to back up any specific recommendation for how much to change the SCCF to ensure we subsume temperature variations or decreases during a measurement ride.

I would like to hear thoughts from anyone on this board about this opinion. If, for instance, we were to double the SCCF, we would then have slightly more than 10 feet per mile to be "eaten up" by the full complement of wobbles, potholes, and cooling or fluctuating temperatures. Measure with a .002% SCCF, then perform and pass a "Russell Check", and voila - no post-cal or further course lengthening.

This would be a proverbial "good thing".

Some thoughts:

I didn't realize that the "Russel check" was considered as a replacement for the post cal ride in all cases.  If the cal course is near-by the race course, a post cal ride is almost always warranted.  Maybe the very occasional rain or step change of temperature would impact the near-by post cal ride, but by how much?  Therefore, eliminating the post cal ride for conveniently near-by cal courses is not really needed.

For far-away cal courses and the very occasional rain or step change of temperature, the "Russel check" could be used in addition to the post cal ride to convince the certifier that the course is accurate without the adjustment due to a significantly larger and probably not reasonably applicable post cal constant.

Additionally, for courses shorter than a half marathon, a third ride isn't really that time consuming, maybe an extra hour or so.

Now we're talking about marathons and half marathons.  Wouldn't it be better to have a discussion with the certifier  for those special cases where the cal course is far away and there is the possibility of significant temperature change.

This seems preferable to changing the short course prevention factor for the purpose of eliminating the post cal ride.

GB: how are you taking a recommendation of the Russell Check for, as you say, "...a replacement for the post cal ride in all cases"? I do not see where anyone is proposing this. The issue we are addressing is the not uncommon situation in which performing a post-cal presents one or all of three distinct problems:

1. The cal course is a long distance from the course being measured
2. Post-cal readings, as Paul describes in his original post on this thread, may require a long trip back to the course to make adjustments
3. Post-cal readings may not accurately reflect the most accurate constant. Post-cals can result in lengthening courses that are already accurate enough to pass a validation without any adjustments.

Those of us who tend to start measuring in the middle of the day, or who measure marathons, encounter the situation described by Paul more frequently than those of us who like to start measuring in the wee hours. I have measured the Baltimore Marathon twice. It took me 12 hours of riding t get through the course all times because of having to wait for traffic at numerous intersections. The last time I did this, the temp was a good bit lower when I finished riding than during pre-cal. I was faced with moving the Finish line far enough that the distance from the 26 mile mark to the adjusted Finish was over a quarter mile. In this one instance, a doubled SCCF would have prevented this particular anomalous "adjustment".

If we all had a magic app that would record the ambient temperature every minute as we ride, then adjust our working constant accordingly, on a minute-by-minute basis as we measure, and likewise adjust our measuring target numbers on the fly, this discussion would be academic. 

Could somebody get to work on developing this app? I would buy it.

Your penultimate paragraph "....Measure with a .002% SCCF, then perform and pass a "Russell Check", and voila - no post-cal or further course lengthening."

I thought you wanted to eliminate post cal by using a 0.0002 short course prevention factor.

Yes, and in my "pre-penultimate" paragraph:

"I would like to explore the non-trivial feasibility of changing the SCCF. We have established no official criteria for "too long". Historically, some validation measurements have revealed courses with a little "extra" distance - beyond the SCCF, of course. I do not have sufficient data to back up any specific recommendation for how much to change the SCCF to ensure we subsume temperature variations or decreases during a measurement ride.

Make sense? I am advocating a serious look at this.

OK, that post was my thoughts.

A situation that makes me worry about adopting the Russell check as a way to avoid having to do a post-race calibration is the following:

Measurer does a pre-cal at home when it is 70 degrees. During his drive to the course a storm moves in and the temp drops 10 degrees. He gets to the course, does his Russell pre-cal, measures the course, does his Russell post-cal. His Russell check tells him his cal constant didn't change during his course measurement, so he doesn't do a real post calibration. Result: short course.

Maybe what's needed to avoid the long drive between cal course and race course (that seems to be the real issue) is some ideas about laying out a  cal course using one person.  Or, maybe using a temporary, one-time, shorter cal course, laid out  and documented using the existing rules.

The biggest issue for me about laying out a cal course on site is when it has to be done. I like to arrive at the course site as early as possible, usually just around first morning light, so traffic is as light as possible. If I have to spend the first 60-90 minutes finding a good location for a cal course, measuring a cal course, and then calibrating my bike, by the time I get to my second measurement the traffic has increased significantly. Regardless of what people say, it takes at least 60min to do all of that, more if things don't go perfectly.

The only accuracy advantage of doing the pre-cal onsite is avoiding the change in cal constant in the time it takes to drive to the course. If you are arriving at the race course at first morning light, the temperature usually changes very little in the two hours or so before that.

I usually calibrate so early in the morning (3 am to 5 am depending on how far I have to drive), that the overnight temp has not finished dropping.  While the regs may not outlaw that, at least the first ride is likely going to be done colder than at calibration.  Just saying we already have a problem pretty much like the "10 degree drop" situation.  And anyway, there could be more of the same temperature fluctuation issue on the drive back to cal course. 

This is practically the reverse situation I encounter, since I measure primarily in daylight hours, often finishing when the temperature is lower than the average while I am measuring. IMHO, both of these situations could be addressed by adopting a SCCF that would "cover" the potential inaccuracies in measurement caused by post-cals in lower temperatures.

It seems to me we would need to accumulate some data on how much measurements are skewed or potentially changed in many of these situations, then graph the results for several measurers. Obviously, different tire types and sizes will produce a range of results.  We would then bracket the approximate differences in measurements when post-cal temps are lower than pre-cals. 

If, for instance, a post-cal measurement is say, 15 counts higher - 18,000, for example, than a pre-cal average of 17,985, we are contemplating an inaccuracy of less than 4.5 feet per mile. If we had used a SCCF multiple of 1.002 instead of the established 1.001, we would have measured a course needing no adjustment.