Why would the horiz distance be useless? Consumer grade GPS devices can't even measure the distance around a 400m track accurately so I'm not sure why you think they'd incorporate elevation into the distance calculations.

Yup, that's why I included the "some argue about accuracy" in my post. But IME, the elevation gains have been close on rides like our GMR to the shack (8 mile climb 2200 ft) to what those using Garmins record. But again, I'm not really into exact numbers as long as they are similar. Plus I'll get a Garmin sooner or later.

When it comes down to it, I us it as a reference for my rides. Only to categorize them as "this will be ok" and "this is going to hurt".

BTW, are you using a Garmin?

For what it is worth, I have been using www.ridewithgps.com and find that the grade data and total elevation gain/loss values are much more accurate than www.mapmyride.com.

Does that work without a GPS as well? If so thanks, I'll take a look at it, MMR seems very limited in its view.

I'm new to map sites so I'm still picking up info on them.

No Garmin but I did recently get one of them fancy smart phones. Strava's rendering of the route recorded on my Droid said 4.4% average on that quarter mile with a peak grade of 6.7%.

I've yet to ride that section with iMapMyRide running on the phone. I'd imagine the gradients would be slightly more accurate than the gradients from their online maps.

I will agree with this as well.

Yes. RidewithGPS.com is very good for mapping routes. I like the site a lot and do think the grade info for the mapped routes is much more useful for the short and steep hills around here than MapMyRide. But RidewithGPS.com gives crazy elevation readings from my Garmin Forerunner 305. Sometimes about double what it shows for the same route mapped on the same site. Also, their latest software update has made the site not play well with Firefox, so I have to open it in IE to map routes.

No, I think it really represents the tangent of the average climb angle, rather than the solution of a right triangle using Pythagoras.

Ok, wikipedia says it's the percent rise divided by run. If the base of a right triangle is horizontal then it is the run. The rise is the vertical leg of the right triangle. Divide rise by run and multiply by 100. The hypotenuse does not enter into the calculation.

Web sites and GPS readings can be inaccurate, sometimes by quite a bit.

The % grade road signs show the steepest grade on the road, not the average. They're to let drivers of heavy rigs know what they'll have to get up/down.

Thanks, I'll give it a try.

I could easily be getting in over my head here , but the reason I used the hypotenuse was to take known values (the distance traveled and the elevation gain) to solve the unknown value (the run).

Rise over run works and is so simple, it is what we use to determine departure climb gradients in air transport. Whether the GPS thinks the world is flat doesn't matter a damn. Can I lose an engine and still get me and the passengers and crew over that hill is all that counts! If it is good enough for that, it is good enough for riding a bike.

I duplicate my routes on RWGPS and MMR. Rwgps doesn't load on my Iphone and MMR does. It seems as if Iphones don't like Adobe Flash.

Fine and dandy. I have no argument with anyone using whatever tool they wish to enrich his/her ride. I never made any reference to GPS and a flat world. Please read the post I replied to (#2) which said the calculation was rise over run using distance traveled in place of run. My response was (not so) simply that distance traveled was not the same as run. It was a response about definition.

I agree that using rise over distance traveled is a close enough approximation for practical use. This whole discussion is much ado about nothing.

I wasn't pickin' on ya BD. Just adding to your post. Someone mentioned "flat earth" and GPS earlier and while true, it is so insignificant it isn't important. Much like this thread!

Ok, okay, ok... let's all agree to go out and buy a meter! How accurate they may be, your legs will tell you how accurate they are. When it starts to hurt more, I bet the grade is getting steeper!

Or you are getting a flat?

I don't understand why you would take the square root of the distance and then square it.

Fixed it to reflect my own problems.
All this talk of slopes, triangles and such has me yearning for some pi(e).

It's useless unless your trying to compute the %grade on a hill, think about it this way, you really don't care if point A and point B are 400m apart, if your 40m higher up you want the running distance, 401.996m because that is how far the wheels have turned. A GPS actually makes a lousy odometer, for a simple reason, it needs to see 3 satellites to do a ground based measurement, it needs 4 to do elevation, if it has fewer it will do an approximation. It's rather simple, if you have elevation to do the calculation, based on a horizontal measurement, thanks to an ancient greek Mathematician, named Pythagoras. GPS isn't all that accurate anyway, I've heard more stories of people joyfully following the GPS until they are so lost, they need a local to figure it out. For thousands of years, all you needed was a compass and a half decent map.

Rise divided by run is correct; % grade is important in several contexts to my profession and industry. Reality is that you can use hypotenuse if you want. Using a 10% grade as an example, that gives a 5.71 degree angle, the cosine of which is .995. That means that your answer will change out in the 2nd or 3rd decimal point. Big fat deal for biking calculations.

You engineering types just don't get the big picture, you are so engrossed in the finer details.

The real issue is "Is there a piece of blueberry pie at the top of the hill?" If there is, the hill becomes a lot less steeper. So, please don't forget to add the "blueberry pie" coefficient into your calculations.

I'm an analog kind of guy who feels trapped in a digital world.

Big chain ring = easy.
Middle chain ring = medium.
Granny ring = OMG!

Bearing in mind that none of this really matters...

This is stretching the capabilities of my very non-engineer type mind. My last math class was in highschool in 1973.
It's hard to write this without knowing how to type formulas.

It's all in the grouping. Subtract the elevation change squared from the distance traveled squared and then take the square root of that.

elevation change / the square root of (the distance traveled squared minus the elevation change squared)