Wow, quite an endeavor! I'm troubled or confused by the ride data to vs from the hill.
The ride to the hill is -135m? The ride back home from the hill therefore should be +135m? Or you have no idea which of any of these figures is correct? The troubling part is evidently depending on what you look at the range is from 92m to 196m?

Also curious, did you ever also try Garmin's elevation correction available by enabling in Connect?

Good job. I have a question: does the Garmin data include the actual elevations or only the differences in these 0.4m units?

Garmin has 'actual' elevation numbers (sometimes more accurate than others of course).

Jeffrey Friedl's Blog » The Voodoo of Elevation Gain and Strava (and How I Get Around It)

In that case instead of adding up all of the delta's it would be more accurate to:

1. Determine the altitudes of all local minima and maxima of elevations.
1a. I would smooth the data first, but that's not necessarily required
1b. Just graph the elevations and choose the tops of humps and bottom of valleys

2. Sum local maxima minus the sum of local minima
2a. the same number of each
2b. start from a local minimum if we're interested in amount climbed.

local maximum is the point where the ride shifts from going up to going down, and vice versa for the minimum.

Adding the deltas is OK if that's all the information you have, but it will always build up errors from rounding and measurement variations, especially when it's already smoothed (such as the .4m increments).

That's strange. Before the change Strava elevations were always low for my rides. Now they will be lower?

I've never had Strava give me an inflated elevation number, ever. This just means I'm going to get boned on elevation gain even more than I have in the past. For while the Garmins I've owned have been consistent, they also read consistently low-- I've done a climb into the local mountains half a dozen times, and know the listed elevations for both the start and the finish: 1,320ft at my house, and 6,140ft for the end point. That's 4,820 feet of absolute veritcal, if the road just constantly climbed from point A to point B. It of course does not, and there are several dips and drops along the way. Strava has never given me more than 4,600 feet on any trip. If I were to use altitude correction, it would likely chop off another couple hundred. But hey, Strava already rounds everything down except calories, so we shouldn't be at all surprised.

Garmin reports Total Ascent and Total Descent at the end of a ride. For a round trip these two numbers should be equal.

As we are talking about cumulative elevation gains the Total Ascent should never be a negative number. The ride to the hill has 135m of Total Ascent even though I start at 85m elevation and end at 24m. Because there are a lot of hills between those two points the elevation gains going up the hills adds up to give the Total Ascent. If there was a straight road between the start and end points then the Total Ascent would have been zero.

The ride back home has a total of 196m elevation gains. In theory, the Total Descent for the ride to the hill should equal the Total Ascent (196m) for the ride back home. Likewise, the Total Ascent (135m) for the ride to the hill should equal the Total Descent for the ride back home.

I don't use Garmin Connect (I'm not that bored) so can't tell you about their elevation correction methods.

I suspect that what Garmin does is this:

Receives the current altitude.
Compares that to the last recorded altitude.
If difference is less than 0.4m the last recorded altitude is written as the current altitude
If the difference is more than 0.4m then the current altitude is written.

This eliminates small variations which you might get riding on a flat road. For example, consider the following 10 seconds of elevation data:
23.4, 23.5, 23.6, 23.5, 23.6, 23.4, 23.3, 23.4, 23.5, 23.6

Garmin will record this as:
23.4, 23.4, 23.4, 23.4, 23.4, 23.4, 23.4, 23.4, 23.4, 23.4

Only when the elevation gets above 23.8 or below 23.0 will the recorded data change.

By recording data like this Garmin is reducing minor variations and effectively smoothing the data.

Any rounding errors will actually work themselves out. If you start a climb at 25m and end at 135m Garmin should always give you 110m elevation gain as long as each elevation data point is greater than or equal to the previous point.

If we look at what strava is doing consider this set of elevation data:
23.4, 23.5, 23.6, 23.7, 23.8, 23.9, 24.0, 24.1, 24.2, 24.3

We can see there is a total of 1.0m elevation gain but because of the way Strava is adding things up, because the difference between each point is less than 0.15m Strava discards the data so Strava would report this set as 0m elevation gain.

Garmin would record this as something like this:
23.4, 23.4, 23.4, 23.4, 23.8, 23.8, 23.8, 23.8, 24.2, 24.2

This would give a total gain of 0.8m. Sure it's off by 0.2m but due to this method the error would only ever be +/- 0.4m regardless of the size of the hill (in theory).

That is awesome. It has been my impression that the Strava elevation numbers are pretty over-inflated. Almost everyone claims more elevation gain than I do on group rides. Interesting to note that even my conservative numbers are probably quite inflated.

I have been riding along a cliff on Catalina Island & had my Garmin track show me as being in the ocean for parts of the ride. That was pretty amusing. I should go back & look at elevation gain data from that one, because there's probably 100 vertical feet difference between where I actually was and where my Garmin thought I was. Although I'm pretty sure my elevation gain data is all barometric data from the Garmin unit itself.

Interesting. But again: I don't really care that much.

I know a guy who thinks that accuracy in a power meter isn't important as long as it's consistent -- but he's pretty obsessed with his Strava accumulated elevation gain.

This isn't adding up the rounded increments like you were doing, but it is looking at absolute elevations. Which is what I said to do, with local minina and maxima.

If instead Garmin is just doing the step and ignoring anything below a .4m change, until it reads a higher change, then it's potentially missing a lot Anything less than .4 meters, up and then down, isn't recorded. Potentially every other sample could be missing a rise and a drop of up to .4 meters.

I'm also from South Cal. I've been using Garmin 800.

1. Palos Verdes Areas: mostly no strava inflation. 10-20% inflation on the ~800 ft Via Coronel climb. ~20% percent inflation on the "Big One". ~30% inflation on Ganado.

2. Hwy 39 from foothill to Crystal Lake: Strava algorithm would inflate 4500 ft to ~6500 ft.

3. GMR ~20% inflation

4. However, GMR-GRR-Baldy Ski Lift: very little inflation

5. Newport Coast, Pelican Hills, Ridge Park: 30 - 40% inflation

6. Santiago Canyon Rd from West to East, ~20-30% inflation, from East to West, not as much..

..........

I need to swap Garmins with you, I guess.

Last week, I did an 80 miler on Wednesday, 60 miles of it with another rider. My ending elevation, 3,250ft. His, 3,350ft. So apparently, I managed no elevation change in those 20 solo miles. Actually, negative 100ft.

Looking over the numbers, I seems I get a consistent average of elevation under-reported ~5%, and distance by 1%. Sounds like nothing, until you extrapolate it out over the past couple years-- Strava has eaten about 175 miles and 38,000ft of climbing. Jerks.

Yesterday I did my first big ride with elevation gains since 12/17, including Newport Coast Hills in Orange County, CA. To my surprise, I lost ~10% of my Gamin 800 elevation after letting Strava calculating it. I've done same route tens of times and I always gained >30% more elevation using Strava.


The monthly strava climbing challenge will become a lot more difficult this year.

Has anyone considered a class-action lawsuit against Strava? This miscalculation of altitude change must be causing severe depression worldwide among roadie types. If nothing else, money from the lawsuit could help offset the cost of all that Xanax that makes the reality of lower elevation gain bearable. This is SERIOUS!!11!!

In addition, I burned 600 - 800 less calories due to the loss of elevation in my big ride. I had to skip dessert last night.

That's my experience also. Though I never used my 810 at sea level, here around my home (~900m) the readings are usually +/- 5-10m apart. I only see more significant fluctuations when I leave home and in the middle of the ride or on the way back it starts raining hard, but that usually happens typically in the summer, when you get those intense and quick summer downpours with quite a barometric fluctuation.

Other then that, the last time I checked (about 3 years ago) Strava was about 10% more generous in terms of average speed, distance and altitude when compared to an Garmin eTrex, my then brand new 810 and a Echowell cyclocomputer. And at least here, of the people I follow through Strava, it's really easy to see who uses Strava on their cells because of the complaints of lost data and/or weird rectified routes on their ride maps.