My friend claims she ran 28 miles during the Chicago Marathon. Could she possibly be that bad at tangents?

Chicago is notoriously bad for GPS. The tall buildings and tunnel running really mess with your watch. That said, if she ran over 4 hours she was probably in a sea of fatties making wide turns. I'd guess she probably did run an extra mile or so.

If she lined up at a back corral, she could be back a half mile from the start.

Likely just didn't turn on or off her GPS at the start or finish line.

I've lived in Chicago for 15 years and never heard so many people complain about the race distance (according to their Garmin).

They used the same course that they have for years. But for some reason, this year many runners have mentioned the distance being off.

Link to her Strava or MapMyRun etc GPS data would be helpful to explain it.

But I'd imagine her GPS trace goes all over the place in the midst of difficult signal issues. You'd have to be going some to make every mile nearly 8% longer, even with waggling through crowds and hi-5-ing supporters on route.

There was a serious problem with my Garmin GPS readings (Fenix 3) this race. This is my 6th Chicago Marathon, and I've used Garmin (Forerunner 405, 210, 610, and now Fenix 3) as well as Polar for the earlier ones. I also live 6 blocks south of Grant Park and run in the city regularly.

Usually, the only time the GPS drops out is at the start of the race as you cross under Randolph Street, adding around .25 mile to the final measurement. I was showing an average race pace between 2-3 minutes per mile FASTER than my actual pace, and by the time I reached mile 7, my Garmin was already showing over 9 miles covered.

The final time that showed on my Garmin was exactly matched with my official race time, but the Garmin showed a final distance of 28.47 miles.

Mine had me at 26.55 at the end of the race. It was usually saying I was going 5-10 seconds faster per mile than I actually was. Either that or I was just running extra, because I know I was weaving a bit

year is 2015 and people still don't know how GPS works. come on.

care to elaborate?

care to elaborate?

justcurious wrote:

care to elaborate?

Tangents do not add any noticeable distance. If your GPS reading is long then it's GPS error.

http://imgur.com/sXiN2oW

They should make the GPS watches know that you're running the Chicago Marathon, and base the distance off of known checkpoints. If your location is at the 15 mile mark of the race it should assume you have been running for 15 miles. This would be more accurate than the current method.

HardLoper wrote:

justcurious wrote:

care to elaborate?

Tangents do not add any noticeable distance. If your GPS reading is long then it's GPS error.

http://imgur.com/sXiN2oW

They should make the GPS watches know that you're running the Chicago Marathon, and base the distance off of known checkpoints. If your location is at the 15 mile mark of the race it should assume you have been running for 15 miles. This would be more accurate than the current method.

While I would agree that GPS error is most likely the largest factor in why someone records a race distance longer than the measured race distance, I think you are understating how much not running the tangents adds. Where I live, our city blocks are about 350' east-west and 380' north-south. I have no idea how that measures up against other cities so maybe your mileage varies. If you run around a block on the inside edge the whole way, you run 1460'. Our streets average about 50' wide. Run around that same block in the middle of the street the whole way, you just ran 1560' Run around that same block on the outside corner, you just ran 1660'. That's almost a 14% difference.

I know nobody runs the outside of every corner, and I know courses are usually more meandering than an around the block measurement. But I know there are tons of places that runners add distance. For example, let's say you take a left onto a straight road for 1 mile, then at the end of that road you take a right. Racing the shortest course would be to run the tangent between those corners - a very slow move from one turn to the next. Let's say the street is 50' wide, and there is a water stop 100' from the end, on the left side of that street. If you get close to the water stop, running a good tangent, and then decide you want a cup, to get over to the water and back on course adds close to 100'. Or if you recognize that water stop is going to be on your left when you turn onto the street you stay on the left but then you're not running the tangent, so you've added around 50' (only the distance to get back on course, not the "go to the water then come back" total).

The marathon I work for has ten 90° turns the first 3.5 miles. With average street widths of 50' through that section, even if you just run through the middle of those turns instead of the tangents you add around 250'. I could easily see somebody adding .1-.2 before you even start taking into account weaving around other runners or going wide to get to a water table.

By the way, I love your idea of having a GPS watch with race courses already in them, and having the brains of the watch check your known location against the coordinates of the course map. That would be a big step forward!

your arithmetic is OK, but your application of it is quite nonsensical

Mine (a Garmin 225) read 26.65. I started in the first corral - very little dodging - and stuck pretty close to the tangent line. As the other poster stated, that first mile was like .25 of it. I guess the rest was margin of error and some minimal weaving. For what it's worth...

hrmmm wrote:

I've lived in Chicago for 15 years and never heard so many people complain about the race distance (according to their Garmin).

They used the same course that they have for years. But for some reason, this year many runners have mentioned the distance being off.

They obviously should all complain to the RD. just kidding.

I've actually had a horrible time getting my Garmin to pick up satellites in the Chicago area lately. Maybe that has something to do with it. Any GPS tech folks here that could explain this?

HardLoper wrote:

justcurious wrote:

care to elaborate?

Tangents do not add any noticeable distance. If your GPS reading is long then it's GPS error.

http://imgur.com/sXiN2oW

They should make the GPS watches know that you're running the Chicago Marathon, and base the distance off of known checkpoints. If your location is at the 15 mile mark of the race it should assume you have been running for 15 miles. This would be more accurate than the current method.

While I'd agree tangents are unlikely to have added two miles, not running them does add noticeable and quantifiable differences. For example while measuring a local half marathon I biked the course once "normally" (no tangents) just to be sure I had all the turns spotted properly, then went back and did two rides doing full SPR. The non-SPR ride was nearly 3 tenths longer. I could easily see someone not paying attention to tangents adding 1/2 mile to a marathon, which even for the fast guys is a couple minutes.

Satellite coverage varies from day to day and hour to hour. Even if you're standing in an open field you might get a couple of one or two hour periods a day where there are only a few satellites in orbit that are visible from your location. Throw in lots of trees and buildings, and a GPS receiver might not get enough signals to calculate a location or may come up with an inaccurate one. Signals also tend to bounce off of shiny buildings ("multipathing"), which will further throw off a cheap handheld GPS.

So it's possible that the satellite coverage and geometry was just really bad in Chicago for a couple hours of this year's race, and that's why everyone's Garmins were suddenly way off.

A tad long wrote:

Mine (a Garmin 225) read 26.65. I started in the first corral - very little dodging - and stuck pretty close to the tangent line. As the other poster stated, that first mile was like .25 of it. I guess the rest was margin of error and some minimal weaving. For what it's worth...

The first few miles of the Chicago Marathon goes under tunnels and along huge skyscrapers.... you're not gonna start getting an accurate reading until the 5k point where you get outside of downtown.

pm222 wrote:

The final time that showed on my Garmin was exactly matched with my official race time, but the Garmin showed a final distance of 28.47 miles.

I'm glad it dose something right for $200. It can tell time.

Then again people are now paying $500 for a apple watch.

every heard of research?everyone can do it,though few will.

justcurious wrote:

care to elaborate?