Searched about but not found anything that answers this particular question….
I’ve found a great local section of trail and have become a little obsessed with a certain Strava segment. At the beginning of the week this particular segment was showing the correct name on my ride after downloading but since then the last 3 days have thrown up various (old?) names for the same route and Strava actually thinks I’ve been riding an adjacent segment on one ride!

My question is is there a less than perfect gps signal strength in this area that is throwing up all these variables at the end of the ride. Would a dedicated garmin like the basic edge 25 have a better capability with regards to keeping a stronger gps signal than my iPhone 6s? The area is heavily wooded which probably doesn’t help…..

Hard to say. iPhone does well in the accuracy tests I’ve been able to find, but that’s not necessarily the same as the strongest signal.

With an edge 800 I have problems starting a ride while in the trees but if I connect first it seems to maintain accuracy with no problems.

No comparison. Dedicated GPS device every time.

Phones are better than they were, but they’re packed full of clever software intended to maximise runtime, and they do this by aggressively managing power use. Your phone doesn’t give a rats’ ar$e about your Strava accuracy, it’ll dial down GPS sampling and mix in triangulation from nearby Cell-towers if it makes the battery last longer.

Generally a device like a Garmin has quite a slow startup time especially if it’s moved a long way from the last place it started, it’s a long time since it last started (because has to find satellites again from scratch) and overhead cover can obstruct signal, but once it’s locked you usually need to go in a tunnel or similar to shake it off.

An iPhone can be as good as a dedicated GPS

Biggest issue is people don’t compare like with like. Shove your bar mounted gps in your shorts pocket or bury it in your rucksack and it’ll give you flaky results too.

Give your phone a good sky view and run a decent app and it’ll give you great results. For example the Strava app used to only record a point every 4s. I think that’s now changed.

It’s not a single GPS signal but multiple signals from a constellation of satellites. Those satellites aren’t at a fixed point in the sky. They are all at fairly low altitudes and they all have different orbits. This means that at any one location you might get a good set of signals as the satellites are well above the horizon and well spread out. Ten or fifteen minutes later and the satellites might appear to be clustered together or close to the horizon and your device will struggle to work out an accurate location. Sometimes you might only have four satellites in view, sometimes there might be double that.

If the location of your segment is in a wooded area or a very narrow steep sided valley then that’s going to affect the quality of the signals being received as well.

I generally find my TomTom running watch is more accurate than my Garmin Edge 500.

But as to worrying about segments, come on now 🙄

I tried a Lezyne Super GPS, got home and checked Strava and loads of the segments weren’t showing. When I zoomed in on the ride the GPS had plotted the correct track shape but it wasn’t on the path so Strava thought I had been riding beside the path hence no segments.

Next ride I used the Lezyne again but also turned on Strava on my iPhone – when I compared at the end the iPhone was much more accurate.

Needless to say I have been using my iPhone ever since but I am considering going for a Garmin.

Too much margin for error on pretty much all consumer GPS devices (isn’t it something like 20m?).

Have in the past run both phone and Garmin on the same ride just to see what the variation is. Never saw any consistency. Sometimes both would pick up the same segment, sometimes the phone would ‘see’ one the Garmin didn’t, sometimes the Garmin would ‘see’ one the phone didn’t. Sometimes the phone gave faster times, sometimes the Garmin did.

I only really use Strava to compare my rides with my rides, so as long as I consistently use the same device it’s all good.

Cheers all, that all makes sense and very interesting… I think I may look out for a lower end garmin and compare but I’m not losing any sleep over it. I’ll happily admit that I very rarely go out for a ‘potter about’ and unless I’m literally on my last legs crawling back to the van with my bike then I feel like it wasn’t a great ride. Strava clicks with my competitive nature whilst on the trails….At 47 and with 3 KOM’s and a handful of 2/3 placings so far on well used segments I’ll sing it’s praises all day long whilst I’m still feeling this good with my efforts at my age!

I’m surprised there was only one Strava hater, kind of expected more clever dicks 🙂

Ghostracer Pro app allows you to select the GPS tracking frequency, up to every 1 second.

Interesting.

If you’re obsessive about GPS signal quality, some of Garmin’s ETrex units allow you to connect to Russia’s GLONAS system simultaneously with the American GPS service as well as the European EGNOS (European Geostationary Navigation Overlay Service) augmentation system (satellites and ground stations I believe).

I’ve not idea though how this effects accuracy in real terms though (I’ve got an ETrex 3.0 and you’ll still find signal blackspots).

So do the latest iPhones…

I worked in the Middle East on a seismic surveying crew. We were using Trimble GPS units that cost (this is 25 years ago) in the region of $20,000 each. These were augmented by a base station at a very precisely known location providing corrections, at the time the Selective Availability signal was still switched on. Without the corrections you’d see your position move by a couple of hundred metres at every cycle of the signals. Turning the differential system on and the accuracy was in the order of 3cm in all three axes within 15 seconds. (Trimble came out with an upgrade to improve the accuracy to 1cm in 3 seconds for the cost of $10,000 per unit!)

However even with this very expensive setup you’d still get times when you couldn’t get a reliable fix and you’d just have to wait for a better configuration of satellites before continuing. The signal “blackspots” move around the surface of the globe, if your device allows it then you can view the visible satellite constellation to verify this. The general usage Garmin devices like the Oregon have this screen but my 510 doesn’t.

In ideal conditions Garmin claim 3 metre accuracy.

I have an RTK base station set up in my local woods just to remove all doubt.

Thanks again!

To clarify, I’m interested rather than obsessed with how Strava works (and sometimes doesn’t) and have learnt more stuff here from like minded riders than a google search that threw up stuff from all sorts of Internet ‘experts’ around the globe….

Garmins 3 metre accuracy is good to know, although many trails in my local woods (Stanmer park in Brighton by the way) run parallel with others at less than that distance so I guess on a bad day things are going to get unclear somedays….

If you look at the list of segments at the bottom of an activity on the desktop site, there is a link to ‘don’t see the segment you are looking for’

It gives you a list of partial matches and the reasons they didn’t give a match.

Yes I found that and it said I deviated from the route at 31% which is why it didn’t read the route I was on. Rubbish of course as you can’t stray from this particular segment!

I don’t think an iPhone is at all bad when it comes to accuracy of where you are at any particular time, so for route planning / following it is perfectly fine (as long as it doesn’t run out of battery). However if you are using it for measurement of speed, stats like that I found that it doesn’t seem to react quickly enough to measure brief changes in speed – for example using it when I was skiing it could work out my average speed over a say 2-3 minute run, but did not react quickly enough to the fairly sudden changes in speed, having what appeared to be about a 2-3 second delay. A dedicated device like a sports tracker seems to do this much better, presumably something to do with refresh rate or summit technical like that…

22mph is 10m/s (sorry about the mixed units) if as above you have an accuracy of 20m on a gps and add in a sample rate of 2s it’s quite easy to see how errors arise especially on short segments. Or how your mate who was slower on the trail can beat you on Strave.