# Map my run - gain

Not calculating gain correctly?

15 messages
19/09/2013 at 11:50

I'm trying to plan routes that are as hilly as possible for training runs over the next couple of months but the gain showing on mapmyrun once I've plotted the route doesn't seem to be right. Either that, or I am fundamentally misunderstanding what "gain" is. I take gain to be, basically, all the elevation gains added together, with elevation loss not counted. So if I run up a hill that takes me from 150ft to 350ft and then down again, I have just done a route with a gain of 200 feet. If I do it twice, it's 400 feet. If I do another hill, up and down, that takes me from 200ft to 300ft, I've done 500 feet overall. Is that correct?

If it is, why might this be happening? I just plotted a 6.5 mile route with 3 climbs on it that mapmyrun shows as having a gain of 574ft. However, when I check the elevation gains for each hill, by hovering the mouse over the lowest, then highest points, noting the differences and then adding them all together, I come up with 732 feet. It also doesn't even notice a 4th climb at all. One hill actually has two peaks, with quite a drop between the two of them, but it just shows it as one constant climb to the main peak. I can understand this is just rubbish mapping but I don't understand why, when it's own map shows 732 feet of gain (as I understand it), it would come up with a figure of 574 feet?

If it is mapmyrun being crap, are there better sites for that sort of thing? Strava, Endomondo?

cougie    pirate
19/09/2013 at 12:11
They're pretty accurate for distance - but not for gain. I'd not sweat over it.
19/09/2013 at 12:24

The thing is cougie, I can understand if the mapping just wasn't picking up the correct elevations for each point on the map, (as it doesn't on at least one section), but I don't quite get why it doesn't add together the elevations properly. That's what made me think I might be misunderstanding gain.

I'm training for an event with a lot of gain and I want to make sure I'm getting enough time going up and downhill and I like to know how the terrain I'm training on compares to what I can expect from the event. I don't expect it to be accurate to the foot, but this seems to be 20-25% out.

19/09/2013 at 12:27

Why does it matter so much?

19/09/2013 at 12:45

I'm not sure Ric. So much as what?

It doesn't matter a huge amount to me. I'm not losing sleep over it or anything. I'm just curious as to whether I'm understanding gain properly and if there are any sites that can track it a bit better than mapmyrun seems to.

All fairly minor stuff really so don't worry about me, I'm coping with it OK.

19/09/2013 at 12:54

One way to find out. Go out and do it.

By that I mean go out and get a 'hard copy' off a Garmin, and then cross check it with the 'theoretical' mapmyrun.

A few runs should yield enough emperical data to be able to factor in any errors or innacuracies.

19/09/2013 at 13:21

I will do Ric. The reason I noticed the gain looked so wrong was because a run that I actually did, containing just one of the hills, was showing as having nearly the same gain. I thought that if I added in the other couple of hills too, I'd have a nicely hilly but fairly short route, for a midweek hilly run (I get bored going up and down the same hill repeatedly) so was a bit confused by the figures I was getting.

19/09/2013 at 13:29
I have read about this on several forums while searching for an answer yo this question. It is allegedly yo do with the way they round things up or down. It seems they round up when drawing up the map and round down when you run the route using their app. If you then upload the route, it apparently rounds it down again.
I am a figures person too as it helps compare routes for difficulty, but to be honest it is often like comparing apples and oranges.
I just use it as a guide though because not all gains are equal :-/
19/09/2013 at 20:47

I don't put  too much reliance on Elevation from mapmyrun, GPS, or barometric altimeters. One you have run the hill once you can assign an effort level to it.

If you are really keen on the stats then a running watch with altimeter (barometer) with gps correction would probably be the best set up.

I walked up a high mountain - barometric altimeter used to swing about with changes in the weather. You will get similar with gps inaccuracy.

20/09/2013 at 00:26
20/09/2013 at 09:08

Assuming they use the standard set of elevation data produced by NASA (I think), elevation data is only available to a resolution of 90 metres.  This can often explain missing hills / dips as the sample points do not actually align with the route or points on the hill.

It’s also possible that when calculating the gain, they take a number of samples at points along the route that is slightly different to what they use to draw the elevation graph with.  With would lead to the discrepancy between their total and your calculation.

Alternatively, mapmyrun may just be crap.

20/09/2013 at 09:29

I started looking into this when I was training for the Beachy Head marathon. I thought it would be nice to be able to compare the difficulty of different runs according to elevation gain and actual numbers, but quickly realised how inaccurate such measurements can be (in my case, whether it was worth believing anything the Garmin might be telling me.)  In the end I stopped sweating the details and just made sure I got a few hilly runs in, mainly a combination of hilly fartleks, and off-road long runs with a few hills thrown in towards the end on tired legs.

Once you're used to running over particular hills you can start to progress the difficulty of your runs, e.g. a medium-long hilly fartlek run for me might include 3 reps of 4 different hills, running hard up each hill, steady on the flat in between and easy coming down.  Progress the run by doing 4 reps on each, etc...

Edited: 20/09/2013 at 09:29
20/09/2013 at 09:57

You could try this website: http://www.bikehike.co.uk/mapview.php instead. I use it to check the elevation data and far as I know its fully accurate.

Edited: 20/09/2013 at 09:58
20/09/2013 at 10:25

Thanks for all the helpful answers. I will be running on these hills regularly so the data I get won't really effect what I do and where I go too much, but it's helfpul to know how my runs are comparing to what I can expect from my race in December, which has 2,300m of ascent.

B_Kins: Thanks for the link, that's a really good, simple site. I plotted the same run on it and came up with a much better elevation profile and more realistic looking figures.

cougie    pirate
20/09/2013 at 10:51
Your race ascent probably won't be 100% accurate either

We'd love you to add a comment! Please login or take half a minute to register as a free member
15 messages