garmin or stop watch when racing

[Jwolf]

Used my Garmin this past weekend during the Manitoba 1/2 Marathon and paid the price, as I did not compenstate for the extra distance, by not running the tangents and, in the end I ended up running an extra 300m, which ended up being an extra 1 minute on the Official Chip Time .... ooops ...

I have never done a race that's not like this-- Garmin is always longer than race distance by about 1-2%. Isn't that pretty well expected?

What's the actual accuracy rating of a Garmin? I don't think it is 100% so you should expect some variation due to its inaccuracy.

It doesn't actually MEASURE distance-- it finds position and calculates distance (and from that, speed). It's supposed to be accurate to the nearest metre in pinpointing position (losing satellite reception temporarily makes this worse), but these errors can cancel themselves out and there is also distance smoothing in the calculation.

But IN PRACTICE (that is, Garmin calculated distance vs. a Jones-wheeled measured distance) it is good to about 1-2% accuracy. Not due to calculation error but because you probably don't run exactly the line the course is measured on. Error can be even higher if the Garmin doesn't get a good signal the whole time.

[MichaelMc]

What's the actual accuracy rating of a Garmin? I don't think it is 100% so you should expect some variation due to its inaccuracy.

It doesn't actually MEASURE distance-- it finds position and calculates distance (and from that, speed). It's supposed to be accurate to the nearest metre in pinpointing position (losing satellite reception temporarily makes this worse), but these errors can cancel themselves out and there is also distance smoothing in the calculation.

But IN PRACTICE (that is, Garmin calculated distance vs. a Jones-wheeled measured distance) it is good to about 1-2% accuracy. Not due to calculation error but because you probably don't run exactly the line the course is measured on. Error can be even higher if the Garmin doesn't get a good signal the whole time.

Have to disagree. "Garmin® GPS receivers are accurate to within 15 meters on average. " this is off of their webpage: http://www8.garmin.com/aboutGPS/ this also gives an extensive list of the many sources of error.

1 reading per second which can be off by 15 meters each time. Many of those errors cancel, but some (the ones left and right of your course) add up instead, giving your route a very fine zig-zag. Moral of the story: nothing is perfect.

[turd ferguson]

Used my Garmin this past weekend during the Manitoba 1/2 Marathon and paid the price, as I did not compenstate for the extra distance, by not running the tangents and, in the end I ended up running an extra 300m, which ended up being an extra 1 minute on the Official Chip Time .... ooops ...

In the future, will use the mile and km markers on the course, and adjust accordingly on my Garmin ...

Have you actually looked at your race and checked it in google earth or sporttracks and looked for the extra 300m. It sounds like you're assuming you actually ran 300m too far.

Whenever I've had a race off by that much, I can usually find the spot in the track that went goofy - usually near buildings or underpasses - and straightening that part of the track gets the error in line.

[Double Bellybuster]

Garmin is always longer than race distance by about 1-2%. Isn't that pretty well expected?

This seems long. I make a point of running the inside path at all times and am usually 0.0% to 0.3% over race distance.

2% would be huge - that would put a marathon over 43K.