Home Training for a Triathlon

How much can I trust Strava?

So I got my first bike in 15 years a week ago, did 20K on Wednesday and just tried my hand at 60K (my arms hurt far more than my legs). This is all in aid of an Olympic length tri but that's neither here nor there.

So Strava tells me that in the 60K I burnt nearly 1,000 calories, which seems like an excuse for a fry up but also seems implausibly high. Worse, in both rides it told me I managed to hit 60Km/h and, seeing as I don't remember riding off a cliff, I don't think that much happened.

Not sure how much you can see without being me but here's a link https://www.strava.com/activities/369400542/embed/e5de895126c8a64c0f1a640043326ddb578b9737

So how much faith should I put in those numbers?


  • risris Posts: 1,002

    strava use little algorithms to interpret your data, based on filling in the gaps from the gps unit (which usually records a data point every 1-10s, depending on how you have it set up). it can mean that it is inconsistent with other datasets - my garmin will give me different ave speeds as strava is smoothing out more. 

    i tend not to get hung up on it, though, it's only a place to record it, and garmin connect/endmondo use the raw / dodgy data point stuff, with all the auto pauses and the like. 

    the calories burned thing will, again, be using a ratio or algorithm that is probably crudely based on the weight you put in. I find that most of the sites (endo/strava) over-guess the calorie burn - when i use a hrm on a workout the numbers are a lot lower, which is probably because the interpretation is different!


  • shadebugshadebug Posts: 23

    So really I need to be investing in a heart rate monitor and a cadence/speed sensor?

  • risris Posts: 1,002

    even with a speed/cadence sensor the data isn't perfect, i would say it is a nice to have, but only if you are going to keep with the sport and the training etc. what's useful is being able to view trends rather than the individual data sets.

    hr information is very useful, particularly for running if you are trying to work out and manage a low hr restrained training cycle. 

    i feel like the calorie stuff is a bit of a red herring, at least for me. i suspect that the devices would need to know a lot more about my physiology to know accurately what my calorie burns are. i don't even look at the numbers in strava/endomondo - weight loss isn't a priority for me so i figure it's irrelevant!

     edit - you might this link at strava's community interesting:

    thread at strava's community

  • Andrew4Andrew4 Posts: 190

    As Ris said, Strava use a lot of cheats and generalisations to make their algorithms work. The calorie data will be really poor, just an estimate based on your weight, speed, and any vertical climb it thinks is in there (note Strava's vertical climb estimates are awful as they "flatten" the satellite image which means switchbacks can look like climbs and descents when its all climb or all descent, I can't find the link but there are several articles summarising this). HR data is the best way to measure energy input and how hard you are working. It still isn't perfect but it is measuring a physiological input rather than a physical output.

    That said, how long did your 60k ride take you? 1000 doesn't seem an unreasonable number, my 920xt slates me as circa 500kcal per 40 minutes (approx. 25k) so give or take 1000 might not be unreasonable for a reasonably gentle 60k

  • I would have thought 1000 cal for for 2:40 was fairly close to the mark. As others have said though, Garmin figures vary greatly from Strava, and I don't worry too much about them. 500 to 600 cal for an hour is a reasonable average. But it very much depends on your weight, the terrain and how hard you ride.
Sign In or Register to comment.