Garmin Connect vs Strava

I Bikepacked the Pennine Bridleway over 3 days last week (well most of it, I started in new mills so I could ride from the doorstep) and out of interest recorded the rides on my edge 530, with heart rate from my instinct 2.

The difference in how garmin and strava interpreted the data was crazy- essentially garmin saying ‘crack on you’re fine’, and strava saying ‘holy crap you’ve overstretched.’

The days were 70-90km with 2000m climbing on a fully loaded bike, often lasting over 8 hours which is quite a lot for me, especially 3 days in a row.

Here’s screenshots of both after day 3. Don’t judge my irregular training in the build up I kept getting sick 🤢

While I wasn’t totally done at any point on the ride I definitely exerted way more than in any of the weeks building up to the trip, so I feel like strava is more on point with the assessment.

Anyone else use both and see similar patterns? In other news the pbw was excellent, particularly the last day in the dales which was the most fun I’ve had on a bikepacking bike.

I gave up looking at those data sets ages ago, never found them at all helpful or representative

Pick one and stick with it – I’d go with Garmin myself, but I’m locked into that for plenty other bits and pieces.

I find the instinct to be useful for running and it sometimes helps to stop me overtraining when I’m tired. The 530 is great for navigation (did the 3 day trip without needing a charge and had 40% batt left), but the garmin fitness tracking metrics are a bit weird when you combine them from 2 devices.

but the garmin fitness tracking metrics are a bit weird when you combine them from 2 devices.

You should set one as your “primary wearable” and allow it to do the calculations. You also need to have “Physio Trueup” configured to amalgamate the data from multiple devices.

OTOH, if you’re really interested in this subject then I’d also suggest taking a look at intervals.icu. It’s free but donations are requested.

Garmin seems to chuck weird statements at me when recording on my edge 530 (which has just died horribly at random with a flickering screen). It tells me I’m unproductive most of the time – regardless of what I do. I think when I’ve been Zwifting this confuses it even more.

The message that goes with unproductive is often ‘your load is optimal but you’re de-training’.

I think Strava is quite basic on that screen – just telling you are against your recent trend.

I’ve not seen the Garmin comparison before, but Strava’s fitness assessment is somewhat limited I think.

I used to ride a lot 10+ years ago and was in fairly decent shape, I’ve spent a couple of months gradually getting back into things and am hideously slow compared with back then.  Obviously.  On Friday I went up a climb and my GPS told me that my PB was 17 minutes, back in 2012, and it took me 23 minutes.  But Strava says my fitness is 47 somethings now, yet I can see that my fitness never went above 46 somethings in 2012 and was actually 38 somethings when I went up that climb 6 minutes faster. LOL

As above imo, Strava is best used as a guide rather than relying on the data. GARMIN seems more accurate if you feed in enough information.

1135 Relative Effort is, surely, a humblebrag! 😄😄

“…But Strava says my fitness is 47 somethings now…”

Strava says my fitness has recently increased from 8 to 12. At least I’m in double figures now. How high does it go? Please tell me that 47 puts you up there with the likes of Pogacar 🙁

The Strava “fitness” number goes much higher than 47. Mine is currently 67 but it’s been higher. I’m not going to be racing any time soon, but I’ll be happy if it stays above my age for a few years (currently 65).

I’m not sure about Garmin (and I might be wrong about this) but I think the Strava weekly effort compares you to the previous couple of weeks, if that, and doesn’t reflect on anything else. So, if you taper before an event and then do some big days it thinks you’ve gone crazy when really you just rested and then did some efforts.

LOL, I’ve no idea what it goes up to, my only benchmark is the last 12 years of rides I’ve fed it.  I am definitely slow these days.  I checked though, mine went from 0 to 12 with 4 slow rides totalling about 60 miles!  I even looked at all the parameters on my account, if my weight was set to something silly maybe that would skew it, but everything is correct.  My training zones are set off my max heart rate and it’s not showing that I’m 100% in the top zone or anything like that, in fact it’s picking up the opposite.  Think it’s just random rubbish somehow.

More seriously, if you look at the graph, if you rode once a fortnight or something it might struggle to get your fitness  score up because each ride adds less than the two weeks off takes away, but obviously doing 25 rides a year would leave you fitter than if you did none.  So there is an underlying principle that you get fitter by riding your miles shorter but more often.

I’m not convinced the algorithm is good enough to be particularly useful, though I find it vaguely helpful if your riding is a mixed bag; you can’t easily compare rides to see how your training is going when some rides are dead flat, or dead hilly or off road etc as you need to combine time, distance and elevation to some extent.

The weekly effort chart I do quite like, it successfully predicted my burn-out last year and I’ve done a better job of getting fit this year by using it to not get carried away. It hates it if I ride lots of climbs and that’s what kills my legs, it’s hilly round here so it’s hard not to get carried away, especially when the weather is nice!  It’s currently squealing about last week though, because I rode Monday, Thursday, Sunday.  If I’d moved one of those rides back/forward a day it wouldn’t be bothered, so maybe a rolling analysis would be more useful to smooth that out.

If I was using it more extensively, there is a data API you can access, it seems reasonably straightforward to bounce the data into an SQL server and build out some more relevant metrics.

My Strava fitness hit 200 when my power meter recorded me averaging 400/800W for 4 hours (can’t remember the power number, it was booked).

50 if I’m doing 5 hours/100km a week.

80 when I was doing the above and 2x spin classes plus 1x circuit training.