mapmyrun vs gps, how much difference?

It'll vary depending on exactly where you run on the course. Inside and outside on the "lanes" and tangents make a difference. Also, the distances on maps usually follow exactly where you draw the lines or down the middle of the course, so unless you run that same exact course you'll get a variation. This gets talked about when people discuss marathon courses being short/long. There was one about AlSal's NYC record a few weeks ago.

http://www.letsrun.com/forum/flat_read.php?thread=4272828

Also, gps is not fool proof. If the signal gets broken up it'll pick back up and draw a straight line from the last point to the newest point. That can add/subtract distance, pace, and time.

And, the "current pace" is not right. It's the pace you WERE running about 30 seconds ago. For tempos it'll help but not guarantee you are on pace all the time.

Just my take and experience. Its a tool that has benefits and drawbacks.

Neither is very accurate, so they will differ. gmap-pedometer.com overestimates my runs by .1 to .2 miles for a 5 mile course. Not sure why.

I tend to use the mapping software on the logging site RunPartner.com, I find this to be more accurate than mapmyrun. Also, RunPartner isn't infested with ads like mapmyrun is, probably the main reason I prefer it, along with the fact at Runpartner the mapping is integrated onto the same site where I log. The RunPartner mapping also has the same nice features as mapmyrun, such as 'autorouting' that when enabled will connect two points assuming you followed the roads, which makes mapping much faster and more convenient.

Still, of course, it is not *perfectly* accurate, and neither is a GPS.

The main reason runs mapped online (both mapmyrun and RunPartner) tend to be a little long, is because you don't actually run in the middle of the street and make 90 degree turns, etc. If you are able to very carefully make it follow your exact route, then yes it will be exactly right, but who has time for that? Better just to keep in mind that the mapped distance is actually about .1 miles further than you ran (for a 5 mile run on mapmyrun or a 10 mile run on RunPartner). Another detail, the more corners there are on the run, the more the mapped distance will be an overestimate (unless, again, you are going through and 'fixing' all of the corners, each of which adds like 5-15 meters).

I've had the opposite experience. I had several runs plotted on Google maps-based sites (mostly gmap-pedometer, some on mapmyrun). After I got a GPS (Garmin 305) and checked they all came up short, usually by around 6-8%. In other words, a 10 mile mapped route might measure 9.3 using GPS.

What's odd is that I can superimpose the routes from the mapping sites with ones generated by the GPS and they look essentially the same; if anything the GPS routes might have more "wiggle", which should make them longer. I'm not talking about routes with "dropouts" that a previous poster mentioned, where the GPS draws a straight line between points where it's lost a signal, which does happen on stretches with lots of tall buildings.

Yeah. I guess you can run a track if you want more exact distances. But, for xc and road races that won't really cut it. Hills, crowds, unkind wind, varying terrain, etc.

If you're really worried about the distances you are hitting, run an extra mile just to be OCD? Or end with a mile of 100m strides 100m jog?

I dunno. There's no way you'll get a perfect measure.

My courses are accurate, as I measure them with a calibrated Jones Counter that I got in the mid 1970's. It was $30 at the time. I think they are quite expensive now, but shouldn't be, because it's a very simple device.

I've been using a map in the last several years for prechecking of courses. The one at the USATF site is my favorite, and I've found it to be quite accurate, maybe because I use it the same way that I'd run (ride & measure) the course.

http://www.usatf.org/routes/map/

Quite a few years ago I tried to calibrate a cateye bicycle cyclometer. This turned out to be impossible, as the cyclometer was inconsistent as well as not accurate. Being used to an error no greater than 1 inch for 1 mile, 20 yards was not acceptable.

That's the rub though, right? You have to run the course exactly the way you map it for it to be "accurate". Since it's really hard to do that it's going to be inaccurate. Even officially mapped courses will come up short depending on exactly how you run it. Inside or outside of curves and tangents on corners.

It is unacceptable for an official course to be short by the distances you mention, but for everyday training I'm not sure anyone can guarantee 1-inch accuracy. I do applaud your dedication to your measurements though.

This is the exact opposite experience from what I have had. The mapmyrun runs are all SHORT. 8.02 on mapmyrun is 8.19 on GPS. 3.34 on mapmyrun is 3.44 on GPS. I have a wheel, but I am not worried enough about it to walk 3 miles. The 8 mile run is a curved road, point to point, I run on the left side the whole time. It is pretty hilly. Thoughts?

I guess the real question is, in general, which if either is usually closer? It sounds like the answer is I should measure my tempos in minutes!

My Garmin 310XT is absurdly accurate - I have unreal run to run consistency on the same courses - readings that don't vary by even .01 from run to run. It is very precise on certified courses.

The only comparison I made from Garmin to mapmyrun.com was on a 6ish mile loop around a lake near my old house and they were surprisingly close. Why not look up a certified course map at USATF and then redraw the map on mapmyrun.com if you want to get a good idea.

appreciate the help wrote:

I mostly use mapmyrun to figure out loops to run, all on roads, and then i never have to wear a watch as I know the loop is 8 or 10 miles. However I would like a 5 mile tempo to be a little more precise. I dont own a GPS, but the only time a borrowed one my 8.01 mapmyrun loop came to 8.19. Is this a standard deviation that other people see or is this a fluke?

I pretty much always found my Garmin to be 1% short. All the time. So, in your case, that 8.01 mile course might actually be 8.09 miles. Either way, it's quite close. You have to realize that triangulating position, estimating pace and calculating total distance based on satellite to ground data is a pretty remarkable feat. In other words, be happy you're only off by a trivial amount. From what I understand MapMyRun just uses more advanced statistical analysis (read that to mean geometrically-based data sets with complex computational algorithms). Regardless, they're still just estimates. The only way to get a real measurement with a wheel. Even bikes are misleading.

Load your run data points into Google Earth and you can see if it made routing errors. My older GPS's made left to right errors that could be relevant on curvy courses, but my 310XT nails the road. Make sure you have all the firmware updates and if you have a 310xt, set it for "every second" data recording.

I've been keeping an excel file of my runs with the GPS distance and the USATF route mapper distance. I have a new Garmin 210. The GPS and USATF have been as close as 0% difference to as much as about 1.5% difference, with most runs coming in around 0.5% difference. That's essentially the same distance in my book. However, it still doesn't answer the question of how these two distance measuring tools compare to the actual on earth distance.

I have also used an older Garmin 201 and that had a much wider range of differences, up to 5%.

What ever the answers maybe, I feel like either of these are more accurate than the distances displayed on treadmills and stationary bikes at the gym.

Get a Garmin FR60...the foot pod-non GPS model.Ninety-eight percent accurate right out of the box, but you can calibrate it on a track and get the accuracy close to 99%! I love it, and use it more than my 305.

Great review here:

http://www.dcrainmaker.com/2010/02/garmin-fr60-review-in-depth.html

the used to be cancer guy wrote:

The only way to get a real measurement with a wheel. Even bikes are misleading.

It's the other way around.

The calibrated bicycle with Jones Counter is the most accurate method of measuring a running course.

A measuring wheel is NOT accurate, as it is highly susceptible to bouncing and constant side to side movement, which doesn't happen on a bicycle, where the wheel has a constant body weight and is kept in a straight line due to the assist of the rigid back wheel.

David Katz and maybe some others use wheels, but I think only for cross country courses. You still need to use the same principles either way.

I agree with Brian above. I have used both and see very little difference. I have also used both to measure race routes that I have run and they all come up really close.

When you finally convert to racing on the trails (which you will) you'll understand it makes no difference anyway!!

Just go by effort and enjoy the experience.

Honestly when I was in good shape my watch and my legs were as accurate as anything else. I could tell exactly how far miles were based on how I feel. The fact is for most runs it really doesn't matter and we are too stressed about exactly how far we run instead of getting in an enjoyable run. 6.01 5.99 not difference in real world.

foot loose wrote:

Get a Garmin FR60...the foot pod-non GPS model.Ninety-eight percent accurate right out of the box, but you can calibrate it on a track and get the accuracy close to 99%! I love it, and use it more than my 305.

Great review here:

http://www.dcrainmaker.com/2010/02/garmin-fr60-review-in-depth.html

Thanks for posting that really nice write up. I'm almost through it, and then going to read the ratings on Amazon.

I have a question about "the next time you synchronize your watch to your computer, the data is automatically downloaded to the computer and then transmitted to Garmin Connect."

How is the synchronizing done and how is the info sent to garmin? Do you need a windows computer for that, or can I use the watch without transmitting the info?

roads trails and treadmills wrote:

I agree with Brian above. I have used both and see very little difference. I have also used both to measure race routes that I have run and they all come up really close.

When you finally convert to racing on the trails (which you will) you'll understand it makes no difference anyway!!

Just go by effort and enjoy the experience.

That and accuracy degrades on trails with twists and turns. My run to run accuracy on a 5.2ish mile trail at Lory State Park is only .04-.05. If you understand how it records this makes sense. Small left to right errors placing you on the globe make little difference if your course has broad curves or is straight, but twists and switchbacks are a bit harder to deal with and, of course, it can't measure accurately at all on a track because left to right errors make a huge difference on turns (plot a track run on Google Earth and you'll see why). And if you go through a switchback, it draws a line between the two points it plotted and doesn't know what you did in between. Although I haven't taken the 310XT on a track and "every second data recording" coupled with a firmware update that claims to improving smoothing may eliminate some of that weakness.