Device accuracy and reliance

[The Ouzo]

We all know that our devices are not 100% accurate and all sorts of things affect their accuracy whilst we ride.

But I thought this was interesting from a group ride this week. A couple on a Tandem put this screen shot on their strava.

Same segment, the faster time was the person at the back.

 

 

 

[DieselnDust]

how long was the segment?

where as the segment?

How and where on the bikes were the bike computers mounted?

Did both gps use the same sampling model or were both set to smart?

[Mamil]

That's the cart before the horse?

It's a short and fast segment which probably amplifies the effect the GPS reading error 

I just KNOW that this is the reason that I have so few (almost no) KOMS

[Skubarra]

4 minutes ago, Mamil said:

That's the cart before the horse?

It's a short and fast segment which probably amplifies the effect the GPS reading error 

I just KNOW that this is the reason that I have so few (almost no) KOMS

Opposite for me... I have a KOM on a short Tygerberg trails downhill segment that I know for a fact is a GPS reading error but is just "slow" enough to look realistic (unless you are aware of my lack of descending skills)....  

[The Ouzo]

13 minutes ago, DieselnDust said:

how long was the segment?

where as the segment?

How and where on the bikes were the bike computers mounted?

Did both gps use the same sampling model or were both set to smart?

Its a short segment on a flat open piece of road.

I cant answer the rest of the questions, i dont even know if both devices were from the same manufacturer.

 

I just thought it was interesting as we know there are accuracy errors and this clearly shows them.

How and why those errors occurred is a discussion I  dont know enough to get involved in.

Edited September 26 by The Ouzo

[DieselnDust]

There is always going to be errors between devices over short segments because of how the bike computer samples or how often it calculates its position. Clarity of the signal it is seeing is important and a riders body is a large enough obstruction to cause an error. The further you travel , the more accurate the overall distance and elevation will be. I had a similar discussion about this on my cycling club chat recently, also a couple on a tandem.

Edited September 26 by DieselnDust
added between devices

[splat]

The person on the back of the tandem was sheltered by the guy on the front. Less wind resistance. hence faster.
Aero is everything!

[gerriemtb]

Maybe they gave the person in the back the faster time because they are a paid subscriber to Strava and the person in front isn't?

[Jase619]

The wife and myself both have Garmin FR245s, and we have occasionally gone out of our way when we run together to ensure that everything is the same, start time, when we set the watches to start acquiring GPS, and we have never had it line up, hers is always slightly behind mine. Which I am okay with haha, but gives her lots of anger issues.

[DieselnDust]

sometimes when the stoker tells the pilot "give it gas" they take it literally and the gps records the extra windspeed as distance gained....

[bleedToWin]

By chance the one gps recorded one or two fewer samples on entry (starting the clock late) and again on exit (stopped the clock early). It's all to do with how Strava decides you have started or finished a segment and your gps sampling rate. No sampling errors here.

[dave303e]

3 hours ago, The Ouzo said:

Its a short segment on a flat open piece of road.

I cant answer the rest of the questions, i dont even know if both devices were from the same manufacturer.

 

I just thought it was interesting as we know there are accuracy errors and this clearly shows them.

How and why those errors occurred is a discussion I  dont know enough to get involved in.

1- flat open roads should not be segments, basically the wind will have the decider on most KO(flats)Ms.

2- Honestly I am in the GIS industry and those calculations and the speed of them are still incredible to me. Let alone when you factor in more and more signals. But there are so many variables. One could be on GPS, the other could be on Glonass only or both or Galileo. You could have device settings, device antennae, device direction, interference from metal objects or signal reflectance. The list of things that can interfere goes on and on....

Grand scheme of things it doesn't count because it is a tandem anyway but there is always going to be error and 99% of the time it is an acceptable margin of error.

[peetwindhoek]

49 minutes ago, dave303e said:

1- flat open roads should not be segments, basically the wind will have the decider on most KO(flats)Ms.

There’s even an app that shows you the best routes to take advantage of the wind!
https://mywindsock.com/

[ChrisF]

2 minutes ago, peetwindhoek said:

There’s even an app that shows you the best routes to take advantage of the wind!
https://mywindsock.com/

 

does it warn you of the return trip ...

 

 

Flew out to Klipheuwel on Monday, return along Adderley was less fun ... 

[nonky]

@OP the following youtube channel may be of interest to you (he mostly tests watches):

 https://www.youtube.com/@TheQuantifiedScientist