12345

Faulty GPS or Faulty Race Distance? (Read 880 times)

xor


    Yes, if your Garmin reads exactly 26.2 miles at the end of a marathon, the course was probably short.

     

    Unless it was a trail marathon with some switchbacks and/or a couple steep elevation changes.  Garmins tend to misunderestimate those.

     

    If you can dodge a wrench, you can dodge a ball.

     


    Got Hills?

      Faulty GPS or Faulty Race Distance or Fawlty Towers?  (It's John Cleese's birthday today!)

       

      Sometimes it would be very difficult to run to the shortest possible route, as the certified course should have been measured.  If there's an "S" - Curve section of busy road that isn't closed to traffic during the race, the measurer may have gone out on a bike at 5:00 on a Sunday morning to avoid traffic and measure the shortest possible route - spending a lot of time in the middle of the road.  But during the race - maybe at 9:00 or 10:00 in the morning, there may be cars and trucks speeding by, so common sense often (but not always) keeps the runners to the shoulders - and a longer route. 

       

      The course can be measured to stay on the shoulder, but then the race has to take actions to ensure runners don't take a shorter route, going into the middle of the road, if there was a lull in traffic.  They would maybe need to put cones up and/or have course monitors making sure runners were'nt cutting the course short.

       

      And the GPS is a good tool for getting an approximate distance or pace, but you just have to realize its limitations and realize it can misplace your position wildly occassionaly, adding or subtracting a big chunk of your mileage.

      "Not to touch the Earth, not to see the Sun, nothing left to do but run, run, run..."


      Imminent Catastrophe

         

         

        And the GPS is a good tool for getting an approximate distance or pace, but you just have to realize its limitations and realize it can misplace your position wildly occassionaly, adding or subtracting a big chunk of your mileage.

         

         

        This happens a lot in aviation.

        "Able to function despite imminent catastrophe"

         "To obtain the air that angels breathe you must come to Tahoe"--Mark Twain

        "The most common question from potential entrants is 'I do not know if I can do this' to which I usually answer, 'that's the whole point'.--Paul Charteris, Tarawera Ultramarathon RD.

         

        √ Tahoe Rim Trail 100M 20/21 July 2013

        Boston Marathon 21 April 2014

        Tahoe Rim Trail 100M 19/20 July 2014

        TJN


        S Army Kettle run...

          Hmmm....

           

          So ...That guy in my office that keeps asking me how long of a marathon I ran this time  ...  actually knows what he's talking about?  

          Tim 


          Black-Toe-Nailed

            Faulty GPS.

             

            GPS use to shorten the distance triangulating curves: They just calculate a curve as a straight lines, they also do the same when they loose signal (even for a short time) and to make matters worse they have a standard deviation of several meters.

             

            Races are not only measured with GPS but also with podometers; basically a wheel with a mesureament tool attached, as these do not depend on external signals they are pretty accurate.

            --

            "If one can stick to the training throughout the many long years,
            then will power is no longer a problem. It's raining? That doesn't matter.
            I am tired? That's besides the point. It's simply that I just have to."

            Emil Zatopek

               

              Races are not only measured with GPS but also with podometers; basically a wheel with a mesureament tool attached, as these do not depend on external signals they are pretty accurate.

               

              Many are measured with a Jones counter:

               

              http://en.wikipedia.org/wiki/Jones_Counter

               

              MTA: which fits your definition of "a wheel with a measurement tool attached", but it's interesting that the wheel is a bicycle wheel..

                Many are measured with a Jones counter:

                 

                http://en.wikipedia.org/wiki/Jones_Counter

                 

                MTA: which fits your definition of "a wheel with a measurement tool attached", but it's interesting that the wheel is a bicycle wheel..

                 

                OK, I'm pretty much of a weirdo when it comes to measuring my courses (it's gotten progressively worse as I've gotten more into racing over the years) but I actually own a Jones counter.  I don't certify race courses, I bought it because as an engineer I like to measure stuff.  I use it to map out splits on the bike paths and roads I use to run my speed work.  I've found that, even with a Jones counter, you have to be exceedingly attentive to even the smallest of details in order to be able to get the USATF to certify a course.  I believe the Jones Counter is currently the only device the USATF will accept as a measurement tool to certify a course.  

                 

                Boring but True story:  When I first got my JC I decided to measure a loop I run around my block (no traffic, one loop = ~0.60 miles) just to kinda get the lay of the land if you will.  I put the JC on my bike, after first pumping up my front tire which was flat, to about 30PSI, and rode the course 4 times.  Each time I tried my best to take the shortest path on the loop (as a racer would). Two times clockwise and two times counterclockwise.  The number of 'clicks' got progressively higher (enough to make a  significant difference in length).  

                 

                11,483 (CW)
                11,511 (CCW)
                11,536 (CW)
                11,615 (CCW)

                 

                Just using 26" as the fixed diameter of my bike wheel (this is FAR from accurate but will do to illustrate how much of a difference the discrepancy above makes):  The distance of 11,483 and 11,615 clicks is 

                 

                11/260 * 11,483 * PI * 26" / 12 / 5280 = 0.6263 miles     (11/260 is the ratio of the counter on the JC)

                11/260 * 11,615 * PI * 26" / 12 / 5280 = 0.6335 miles

                 

                Diff = 0.0072 miles or about 1.1%

                 

                Not much of a difference in length but absolutely and totally unacceptable for measuring a course.

                 

                My first thought was that the JC was defective as it wasn't a random thing - each loop got longer.  My next thought was maybe the tires were heating up and expanding but that would mean that I would get progressively LESS clicks as they expanded.

                Turns out that the next morning my front tire was flat.  There was a slow leak in the tube that I couldn't perceive in the timeframe of the test but it made a difference to the JC.  I did the experiment again the next day after fixing the leak and pumping the tires to 70PSI, much more consistent:

                 

                11,133 (CW)
                11,133 (CCW)
                11,127 (CW)
                11,135 (CCW)

                 

                8 clicks difference in length or 0.07% which is not great but a factor of 20 better than before - I need to get better at riding the same path each time.  This kind of error would translate into +/- 49.8 feet over a 26.2 mile course, assuming I ride the entire marathon course on the shortest possible path. The Short Course Prevention factor is there to take care of this kind of residual error.

                 

                The above is one of the reasons why, to certify a course, it is required  to ride a known 'calibration' course before AND after you measure the actual race course and ride both the race course AND the calibration course several times.  So many sources of error.  And all of this can be for naught if the RD or volunteers don't follow your course markings correctly.

                 

                GPS?  Fuggedaboudit!  

                 

                PS - hope my math above is right!


                TRIing to beat the heat!

                  If it makes anybody feel better, Garmin 310XT said that the first two marathons I ran (Philly 2011 & Wisconsin 2012) were both 26.57 miles.  I just ran Philly 2012 and tried to pay better attention to tangents and ran... wait for it... 26.6 miles!  I would love for a downward trend to begin.  Even 26.5 would be nice at this point.

                  2012 Goals

                  Sub-1:42 for half marathon √ (1:41 at Disney, Jan '12)

                  Sub-22 for 5k √ (21:51 in Sept '12)

                  BQ for marathon- FAIL


                  Joggaholic

                    I happened to need to do a full charge-and-discharge cycle for my 405 recently. I charged it fully, turned on the gps, left it on the window sill, and headed to bed. The next morning I recharged it and uploaded the workout.

                     

                     

                    I think Gremlins were in my house.


                    Feeling the growl again

                      If it makes anybody feel better, Garmin 310XT said that the first two marathons I ran (Philly 2011 & Wisconsin 2012) were both 26.57 miles.  I just ran Philly 2012 and tried to pay better attention to tangents and ran... wait for it... 26.6 miles!  I would love for a downward trend to begin.  Even 26.5 would be nice at this point.

                       

                      You're talking like 1.2% discrepancy from the certified race distance, which is well within the advertised margin or error for the unit.  Why would you expect any closer?

                      "If you want to be a bad a$s, then do what a bad a$s does.  There's your pep talk for today.  Go Run." -- Slo_Hand

                       


                      HobbyJogger & HobbyRacer

                        Either Wing is leaving too many windows open, or has a bunch of broken windows to fix.

                         

                        Or else Wing's watch figured out how to unlatch the windows.

                        It's a 5k. It hurt like hell...then I tried to pick it up. The end.

                        12345