12345

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

    This article answers all your questions on GPS accuracy....

     

    I ran a Half Marathon a couple of years ago and everyone was complaining about the course because all the GPS were measuring about 1/4 of a mile longer and sometimes even more....

     

    The above article was sent out to everyone and its does a good job of explaining why a race course and GPS might not be totally in synch....

     

    So this article talks about the Jones counter. Those counters basically count the revolutions a bike wheel makes, even multiple ticks per foot of revolution.  The measurement depends on the circumference of the wheel. The wheel circumference will be impacted by tire pressure and by the weight of the person sitting on the bike.

     

    Using a cheaper bike store odometer I did a test to see how much my weight impacted odometer accuracy: I measured a known 0.5 mile course (measured with a rigid surveyors wheel (3 foot circumference).   I inflated my bikes tires as tight as I felt comfortable and then measured the course once while just walking the bike and then again while riding the bike. The difference was not insignificant over just a 1/2 mile.  

     

    I guess my point is I hope those counters are placed on bikes with as rigid tires as possible.  I think they are because the one time I spoke with someone that was asked to be involved in a course certification he told me he had to measure the course, and then someone else with a different counter also measured they course. The two measurements had to be with in a pretty tight tolerance.  

      So this article talks about the Jones counter. Those counters basically count the revolutions a bike wheel makes, even multiple ticks per foot of revolution.  The measurement depends on the circumference of the wheel. The wheel circumference will be impacted by tire pressure and by the weight of the person sitting on the bike.

       

      Using a cheaper bike store odometer I did a test to see how much my weight impacted odometer accuracy: I measured a known 0.5 mile course (measured with a rigid surveyors wheel (3 foot circumference).   I inflated my bikes tires as tight as I felt comfortable and then measured the course once while just walking the bike and then again while riding the bike. The difference was not insignificant over just a 1/2 mile.  

       

      I guess my point is I hope those counters are placed on bikes with as rigid tires as possible.  I think they are because the one time I spoke with someone that was asked to be involved in a course certification he told me he had to measure the course, and then someone else with a different counter also measured they course. The two measurements had to be with in a pretty tight tolerance.  

       

      The procedure for actually certifying a course requires you to mark out a "calibration course" of at least 300m (but they recommend 1000m) with a steel tape and then calibrate the jones counter on it. And you're supposed to ride the calibration course 8 times each time you measure a race course--4 times before and 4 times after the course measurement.

      Runners run.

        Whenever I run a race, I expect the GPS measured distance to be long because I know I won't hit the tangents exactly as measured (although I try REALLY hard).  I ignore the mile markers along the way because they are subject to who drops them at the spot (and I often see paint markings before or after the actual spot).

         

        What gets me is when you see an online map for the race showing the incorrect distance.  Then when you run the race and nail that distance.  Kind of makes you go, Hmmmm...

        And you can quote me as saying I was mis-quoted. Groucho Marx

         

        Rob


        Feeling the growl again

            Anecdotally, we see a lot of cases where the GPS distance is longer than the specified course because the runner wasn't able to take the exact tangents on turns and road crossings, and thus did not run the shortest possible distance. 

           

          It's not just the tangents.  GPS errors are three dimensional in nature; any point calculated has an altitude above/below your exact location, left/right of your exact location, and in front of/behind your exact location.  Because the errors are random, as you run over a course the dimension of in front of'behind more or less cancels out (this error is a primary reason instantaneous speed on a GPS is next to useless at running velocities, however).  If it plots you too far forward one point, the next one it will just short you some in distance covered between but in terms of measuring the length of the course this all cancels out.

           

          The left/right errors, however, do not cancel.  If you run an exact straight line then look at the GPS plot it will not be a straight line, it will have movement -- error -- to left and right of your true path.  All of this extra distance is added onto the route measurement.  This is why if you run a course that does not have a lot of turns you will get better accuracy by turning the "smoothened" function to the highest setting...this algorithm causes the unit to assume some of the left/right is error and average it out of the final result.

          "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

            The procedure for actually certifying a course requires you to mark out a "calibration course" of at least 300m (but they recommend 1000m) with a steel tape and then calibrate the jones counter on it. And you're supposed to ride the calibration course 8 times each time you measure a race course--4 times before and 4 times after the course measurement.

             

            And you have to re-measure if they don't agree enough -- so you want to perform your measurement when the temperature is pretty stable.

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

              Whenever I run a race, I expect the GPS measured distance to be long because I know I won't hit the tangents exactly as measured (although I try REALLY hard).  I ignore the mile markers along the way because they are subject to who drops them at the spot (and I often see paint markings before or after the actual spot).

               

              What gets me is when you see an online map for the race showing the incorrect distance.  Then when you run the race and nail that distance.  Kind of makes you go, Hmmmm...

               

              Yes to this.  You cannot count on mile markers being placed in the correct spot, even at certified-course races (shoot, even the start / finish / turnarounds etc might be placed incorrectly on a certified course!).  

              I ran a USATF-certified half marathon on Sunday and ended up with 13.35 miles (don't care, I force the distance to 13.11 miles in my log) but mile 4 and 5 were placed incorrectly and then you turn around and, of course, miles 9 and 10 were off as well.  This race has been run many, many times and the mile markers are actually spray-painted in the road WRONG.  Yet no one ever goes out and fixes them for the next year!  You just have to learn pacing by effort and just use the GPS for funsies.  The GPS is almost useless on this course (Humboldt Redwoods 1/2) as the trees block GPS signals and I'm actually surprised at how well it does in this situation.

               

              4 Interval 1.2 mi 7:32.73 26:29.27 6:18     Marker was off
              5 Interval 0.82 mi 5:18.05 31:47.32 6:28     Short mile from mile 4

               

              9 Interval 0.87 mi 5:34.65 56:44.64 6:25     Same misplaced pair (miles 4/5 and miles 9/10)
              10 Interval 1.14 mi 7:27.68 1:04:12.32 6:33     long mile
              xor


                "my course was long!" gps season started a couple weeks late this year.

                 

                  Are we saying that the title of this thread poses a false dichotomy?

                    What I have learned from all this is that I need to run faster...

                      What I have learned from all this is that I need to run faster...

                       

                      how so? :-)

                        how so? :-)

                         

                        Because the faster you run in a straight line relative to the GPS sampling speed, the more accurate your GPS measured distance will be?

                         

                        Or because back in the "good old days" (before I started running) it was about racing the race and not worrying if the advertised distance was off by a few angstroms in one direction or the other?


                        HobbyJogger & HobbyRacer

                          No, I think it will be more accurate if you run more slowly. Everyone needs to run more slowly. Everyone will get better accuracy, and I will get... well, I won't tell you, but I will get some benefit myself, once everyone starts following my program.

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

                          xor


                            Stop swinging your dichotomy around.

                             

                              Or because back in the "good old days" (before I started running) it was about racing the race and not worrying if the advertised distance was off by a few angstroms in one direction or the other?

                               

                              Big grin

                                What I have learned from all this is that I need to run faster...

                                 

                                This really is a nugget of wisdom.  If you are relying on your GPS for pacing and you have an important goal time that you want to achieve, you better not be relying on running the pace that your GPS says you are running.

                                 

                                9:09 is the pace that you should run for a Half Marathon to break 2 hours.  If you run a 9:09 pace per your GPS when you get to the finish line you will be very disappointed because you aren't going to break 2 hours.

                                 

                                If on average the GPS measures long by 1% then you need to run 1% faster than your GPS tells you to make your goal.  That means you need to run a little faster than what your GPS tells you.  Instead of running 9:09 pace according to your GPS you need to run a 9:04/mile according to your GPS to make it to the finish line in under 2 hours.  9:04 for 13.23 miles = 1:59:57.

                                 

                                Even on a 5K it's going to make about 15 seconds difference which could make or break making a goal time.  I'm trying to shoot for a sub 23:00 5K coming up Saturday.  To do that I need to run a 7:24 pace for 3.1 miles.  On my last 5K my GPS logged it a 3.16 miles.  If I run a 7:24 pace for 3.16 miles I'm not going to break 23:00, I'll be almost 20 seconds short!  To make my goal I need to run at a 7:17 pace according to my GPS.  7:17 for 3.16 miles = 23:01 so I'll have to shave 1 second off there at the end! ;-)

                                 

                                Of course if you are already running all out or have already blown up it won't make a difference!

                                Age: 46 Weight: 205 Height: 6'2" (Goal weight 195)

                                Current PR's:  Mara 3:48:09; HM 1:43:26; 10K 43:59; 5K 21:27

                                12345