Go Back  Bike Forums > Bike Forums > Road Cycling
Reload this Page >

Strava....WTF moment.

Notices
Road Cycling “It is by riding a bicycle that you learn the contours of a country best, since you have to sweat up the hills and coast down them. Thus you remember them as they actually are, while in a motor car only a high hill impresses you, and you have no such accurate remembrance of country you have driven through as you gain by riding a bicycle.” -- Ernest Hemingway

Strava....WTF moment.

Old 09-11-18, 05:04 PM
  #1  
Senior Member
Thread Starter
 
bruce19's Avatar
 
Join Date: Jul 2006
Location: Lebanon (Liberty Hill), CT
Posts: 8,695

Bikes: CAAD 12, MASI Gran Criterium S, Colnago World Cup CX & Guru steel

Liked 1,393 Times in 811 Posts
Strava....WTF moment.

I've got a 6.35 mi. TT w/approx. 250 ft. of elevation that I try to do once a month. Last year, at age 71, I did it at 17.4 mph. This year, now 72, I was hoping to hit 18.0 mph by Sept. For reasons that I don't understand, I actually hit 18.4 mph in July So, today I went out and rode it hard. It was a muggy but windless day. According to my Garmin Edge 200 I did it in 20.03 min. for a 19.0 mph average. I uploaded to Garmin Connect and, again, it gave me 20.03 min. for 19.0 mph. But, when it went to Strava, 6 seconds were added making it 20.09 min. which yielded 18.9 mph. Now I don't really care about the .1 mph but why does Strava do that? Just curious.
bruce19 is offline  
Old 09-11-18, 05:29 PM
  #2  
Senior Member
 
MagicHour's Avatar
 
Join Date: Aug 2013
Location: NYC
Posts: 877
Liked 12 Times in 10 Posts
To hazard a guess: Strava and Garmin have slightly different algorithms for calculating avg speed over a distance? and/or could be dependent on other factors like speed sensor vs gps speed.
MagicHour is offline  
Old 09-11-18, 05:30 PM
  #3  
Non omnino gravis
 
DrIsotope's Avatar
 
Join Date: Feb 2015
Location: SoCal, USA!
Posts: 8,553

Bikes: Nekobasu, Pandicorn, Lakitu

Liked 1,731 Times in 958 Posts
Welcome to... Strava Math®!!

Only half joking. I've had Strava add a few seconds to rides where I literally never unclipped or stopped moving. On Sunday, GPS: 58m56s; Strava: 59m06s. Never stopped once. Today, GPS: 4h45m39s; Strava: 4h45m46s-- but I had over an hour of non-moving time. Strava just makes it up as it goes along.

I think part of it is due to GPS latency-- Strava "smooths" segments in an effort to remove jittery GPS data. Some days there's more than others.
__________________
DrIsotope is offline  
Old 09-11-18, 05:30 PM
  #4  
Senior Member
 
Join Date: Sep 2011
Posts: 4,764
Liked 232 Times in 173 Posts
is the recorded file exactly the 6.35 mi distance only? Probably the difference in elapsed vs moving time. Make it a segment on strava so it will automatically compare your attempts
redlude97 is offline  
Old 09-11-18, 06:07 PM
  #5  
Senior Member
 
bobwysiwyg's Avatar
 
Join Date: Jan 2016
Location: 961' 42.28° N, 83.78° W (A2)
Posts: 2,344

Bikes: Mongoose Selous, Trek DS

Liked 319 Times in 189 Posts
My Strava WTF moment is a bit different. My wife and I headed up to the U.P. of Michigan for a two week vacation, a trip of about 530 miles.
While up there I noticed my cell battery going down faster than usual even for this area.
A couple of times I did a "close all," did a restart and finally found Strava had somewhere running since we left Ann Arbor. Tried ending and uploading to get out after and delete the event. Over 800 miles, and total climbs I would love to have done.. no, survived! 😀
bobwysiwyg is offline  
Old 09-11-18, 06:14 PM
  #6  
Senior Member
Thread Starter
 
bruce19's Avatar
 
Join Date: Jul 2006
Location: Lebanon (Liberty Hill), CT
Posts: 8,695

Bikes: CAAD 12, MASI Gran Criterium S, Colnago World Cup CX & Guru steel

Liked 1,393 Times in 811 Posts
There's something I'm not understanding. If my on bike unit says I've ridden for 20:03 and it uploads to Garmin Connect at 20:03 how can Strava change that? Where did that 6 seconds come from? I really don't get it.
bruce19 is offline  
Old 09-11-18, 06:17 PM
  #7  
Senior Member
 
CliffordK's Avatar
 
Join Date: Nov 2014
Location: Eugene, Oregon, USA
Posts: 27,549
Liked 4,590 Times in 3,409 Posts
Many Garmins have an issue that they tend to track once every 5 seconds, but Strava likes once a second data points.

This becomes an issue on segments where the closest points are either before or after the start line.

In many cases, Garmin users get an advantage, but I presume they are just as likely to get penalized.

Take this segment, for example. Short and sweet!!!

https://www.strava.com/segments/12973075

The KOM ride: 47s, 34.2MPH
On the analysis page, it shows Maximum speed for the segment, 34.0 MPH, and average of 32.6 MPH

Note, Strava put the maximum speed as LESS than the average segment speed.

The issue in this case is that Strava put the end points for the ride inside the actual end points for the segment. So, the Garmin course is about 5 seconds shorter than other rider's courses. Then uses the segment end points to calculate the segment speed.

It makes it dang hard to catch the KOM.

I presume the OP has a similar, but opposite effect, where his Garmin end points are outside of the actual segment, so he had an extra 6 seconds of riding calculated in time, but not distance.

That difference will likely disappear when viewing the Strava Analysis page.

I don't know if those 5 seconds between data points is fundamental to Garmin, or is a configuration option. It might be worth looking at, depending on whether battery life is an issue.

BTW:
Congrats on the 19 MPH segment.
CliffordK is offline  
Old 09-11-18, 06:32 PM
  #8  
Senior Member
 
Join Date: Sep 2011
Posts: 4,764
Liked 232 Times in 173 Posts
Originally Posted by bruce19
There's something I'm not understanding. If my on bike unit says I've ridden for 20:03 and it uploads to Garmin Connect at 20:03 how can Strava change that? Where did that 6 seconds come from? I really don't get it.
Garmin connect tells you total time, moving time, and elapsed time, which one are you referring to? If moving time then the calculations can be different
https://support.strava.com/hc/en-us/...e-Calculations
redlude97 is offline  
Old 09-11-18, 06:37 PM
  #9  
Senior Member
 
WhyFi's Avatar
 
Join Date: Jan 2010
Location: TC, MN
Posts: 39,521

Bikes: R3 Disc, Haanjo

Liked 9,462 Times in 4,673 Posts
Strava Live segment results on your Garmin/Wahoo aren't Strava official. Strava even discards any info from your wheel-based speed/distance sensors, instead relying solely on their own parsing of your GPS data.
WhyFi is offline  
Old 09-11-18, 06:45 PM
  #10  
Senior Member
Thread Starter
 
bruce19's Avatar
 
Join Date: Jul 2006
Location: Lebanon (Liberty Hill), CT
Posts: 8,695

Bikes: CAAD 12, MASI Gran Criterium S, Colnago World Cup CX & Guru steel

Liked 1,393 Times in 811 Posts
Originally Posted by WhyFi
Strava Live segment results on your Garmin/Wahoo aren't Strava official. Strava even discards any info from your wheel-based speed/distance sensors, instead relying solely on their own parsing of your GPS data.
Thank you. Garmin Connect shows me three times......"time", "moving time" and "elapsed time." Why "time" is different from the other two, I don't understand.
bruce19 is offline  
Old 09-12-18, 07:19 AM
  #11  
Senior Member
 
rgconner's Avatar
 
Join Date: Mar 2017
Location: Sacramento, CA
Posts: 1,156

Bikes: Curtis Inglis Road, 80's Sekai touring fixie

Liked 11 Times in 9 Posts
"People assume that time is a strict progression of cause to affect, but actually, from a non-linier, non subjective point of view it is more like a big ball of wibbily wobbly timey wimey...stuff " - Dr. Strava
rgconner is offline  
Old 09-12-18, 07:27 AM
  #12  
Non omnino gravis
 
DrIsotope's Avatar
 
Join Date: Feb 2015
Location: SoCal, USA!
Posts: 8,553

Bikes: Nekobasu, Pandicorn, Lakitu

Liked 1,731 Times in 958 Posts
__________________
DrIsotope is offline  
Old 09-12-18, 08:35 AM
  #13  
Senior Member
 
wphamilton's Avatar
 
Join Date: Apr 2011
Location: Alpharetta, GA
Posts: 15,279

Bikes: Nashbar Road

Liked 341 Times in 228 Posts
I think it's a boundary condition due to the sampling rate. I've seen something similar in a short segment, where my times were rounded up to 5 second intervals, or 3 second intervals depending on the device, yet I could see other times down to the second.
wphamilton is offline  
Old 09-12-18, 08:45 AM
  #14  
- Soli Deo Gloria -
 
Join Date: Aug 2015
Location: Northwest Georgia
Posts: 14,779

Bikes: 2018 Rodriguez Custom Fixed Gear, 2017 Niner RLT 9 RDO, 2015 Bianchi Pista, 2002 Fuji Robaix

Liked 736 Times in 469 Posts
Strava is an order of magnitude tool.

A +/- variation is to be expected.

I don't think they guarantee accuracy.


-Tim-
TimothyH is offline  
Old 09-12-18, 08:48 AM
  #15  
Non omnino gravis
 
DrIsotope's Avatar
 
Join Date: Feb 2015
Location: SoCal, USA!
Posts: 8,553

Bikes: Nekobasu, Pandicorn, Lakitu

Liked 1,731 Times in 958 Posts
If you have VeloViewer, you can look at all of your efforts over any given segment, and watch the recorded length of that segment vary by at least 10%. It's the luck of the GPS ping some days. If a "measured" 1 mile segment is 0.975 miles on one effort and 1.025 on another, you can have a better PR with a lower average speed.

Or as I like to say, "When is a mile not a mile? Strava!"
__________________
DrIsotope is offline  
Old 09-12-18, 08:51 AM
  #16  
6-4 Titanium
 
Join Date: Jul 2018
Posts: 330
Likes: 0
Liked 36 Times in 31 Posts
This would explain the 50 year olds averaging 27mph. 71 year old averaging 19mph now. In B4 72 year olds averaging 27mph sitting in.
MyTi is offline  
Old 09-12-18, 09:01 AM
  #17  
Senior Member
 
wphamilton's Avatar
 
Join Date: Apr 2011
Location: Alpharetta, GA
Posts: 15,279

Bikes: Nashbar Road

Liked 341 Times in 228 Posts
Originally Posted by DrIsotope
If you have VeloViewer, you can look at all of your efforts over any given segment, and watch the recorded length of that segment vary by at least 10%. It's the luck of the GPS ping some days. If a "measured" 1 mile segment is 0.975 miles on one effort and 1.025 on another, you can have a better PR with a lower average speed.

Or as I like to say, "When is a mile not a mile? Strava!"
The segment keeps the same length regardless, so Strava just divides the invariant segment length by the time, so the variation you're seeing there won't impact the segment. Except at the very end, if the final position reading is skewed off from the endpoint in one direction or another, Strava may think that you've finished earlier or later than you actually did. Which would make it off by one time sample interval.

The recorded ride length will vary which makes for some amusing results in Strava average speeds, but that's a different question
wphamilton is offline  
Old 09-12-18, 09:13 AM
  #18  
Non omnino gravis
 
DrIsotope's Avatar
 
Join Date: Feb 2015
Location: SoCal, USA!
Posts: 8,553

Bikes: Nekobasu, Pandicorn, Lakitu

Liked 1,731 Times in 958 Posts
Originally Posted by wphamilton
The segment keeps the same length regardless, so Strava just divides the invariant segment length by the time, so the variation you're seeing there won't impact the segment. Except at the very end, if the final position reading is skewed off from the endpoint in one direction or another, Strava may think that you've finished earlier or later than you actually did. Which would make it off by one time sample interval.
This is an ever-so-slightly uphill 2-mile segment. The PR has a lower average speed than 2-4. Because it GPS logged shorter.

__________________
DrIsotope is offline  
Old 09-12-18, 09:22 AM
  #19  
Senior Member
 
wphamilton's Avatar
 
Join Date: Apr 2011
Location: Alpharetta, GA
Posts: 15,279

Bikes: Nashbar Road

Liked 341 Times in 228 Posts
Originally Posted by DrIsotope
This is an ever-so-slightly uphill 2-mile segment. The PR has a lower average speed than 2-4. Because it GPS logged shorter.

That appears to be a 3rd party analysis, not the Strava leader board. I am almost 100% certain that Strava segment leader boards are derived from the single recorded distance, which is the GPS distance from the ride from which the segment was created.
wphamilton is offline  
Old 09-12-18, 09:30 AM
  #20  
Senior Member
Thread Starter
 
bruce19's Avatar
 
Join Date: Jul 2006
Location: Lebanon (Liberty Hill), CT
Posts: 8,695

Bikes: CAAD 12, MASI Gran Criterium S, Colnago World Cup CX & Guru steel

Liked 1,393 Times in 811 Posts
Originally Posted by rgconner
"People assume that time is a strict progression of cause to affect, but actually, from a non-linier, non subjective point of view it is more like a big ball of wibbily wobbly timey wimey...stuff " - Dr. Strava
Quantum physics makes me crazy. At least in this dimension.
bruce19 is offline  
Old 09-12-18, 09:38 AM
  #21  
Senior Member
Thread Starter
 
bruce19's Avatar
 
Join Date: Jul 2006
Location: Lebanon (Liberty Hill), CT
Posts: 8,695

Bikes: CAAD 12, MASI Gran Criterium S, Colnago World Cup CX & Guru steel

Liked 1,393 Times in 811 Posts
This has been helpful. Garmin Connect says.....Time: 20:04....Moving time: 19:52...Elapsed Time: 20:09. Questions remain...to wit: If I never stopped moving, what is "elapsed time" and "time?" FWIW Garmin Connect also says my average speed was 19.0 mph and my average moving speed was 19.2 mph. Strava took the Elapsed Time and gave it 18.9 mph.
bruce19 is offline  
Old 09-12-18, 09:44 AM
  #22  
Senior Member
 
WhyFi's Avatar
 
Join Date: Jan 2010
Location: TC, MN
Posts: 39,521

Bikes: R3 Disc, Haanjo

Liked 9,462 Times in 4,673 Posts
Originally Posted by bruce19
This has been helpful. Garmin Connect says.....Time: 20:04....Moving time: 19:52...Elapsed Time: 20:09. Questions remain...to wit: If I never stopped moving, what is "elapsed time" and "time?" FWIW Garmin Connect also says my average speed was 19.0 mph and my average moving speed was 19.2 mph. Strava took the Elapsed Time and gave it 18.9 mph.
For segments, Strava always looks at elapsed time (as they determine it from looking at the GPS data). How Garmin defines Time, Moving Time and Elapsed Time is irrelevant, but I would be curious about the difference between Time and Elapsed Time, too (well, if I had a Garmin).
WhyFi is offline  
Old 09-12-18, 10:06 AM
  #23  
Senior Member
 
99Klein's Avatar
 
Join Date: Nov 2012
Location: Lincoln Nebraska
Posts: 1,088

Bikes: 99 Klein Quantum, 2012 Cannondale CAAD10 5, Specialized Tarmac Comp, Foundry Thresher, Fuji Sportif

Likes: 0
Liked 1 Time in 1 Post
Welcome to Strava. Their number always differ from Garmin, Endomondo, MMR... Now if you really want to kick a$$ at segments, leave your cycling computer at home and use an iphone. Side, by side, they are always faster segments.
99Klein is offline  
Old 09-12-18, 11:21 AM
  #24  
Newbie racer
 
Join Date: Feb 2018
Posts: 3,406

Bikes: Propel, red is faster

Liked 1,571 Times in 974 Posts
Hit the lap button on your GPS. Won't fix any silly Strava stuff, but will give you an exact milepost for doing your TT. If you know it starts at that mailbox and ends at the railroad, just do your lap accordingly.
burnthesheep is offline  
Old 09-12-18, 12:11 PM
  #25  
Senior Member
 
Join Date: Sep 2011
Posts: 4,764
Liked 232 Times in 173 Posts
Originally Posted by bruce19
This has been helpful. Garmin Connect says.....Time: 20:04....Moving time: 19:52...Elapsed Time: 20:09. Questions remain...to wit: If I never stopped moving, what is "elapsed time" and "time?" FWIW Garmin Connect also says my average speed was 19.0 mph and my average moving speed was 19.2 mph. Strava took the Elapsed Time and gave it 18.9 mph.
if you have autopause on, then the difference is in the time between when you slow down but autopause hasn't started. If you turn off autopause I bet your time and elapsed time will now match up.
redlude97 is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Your Privacy Choices -

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.