Campaign for real averages

[Dogtrousers]

Only time I will be worrying about averages this year is keeping above 19km/hr to stay ahead of the broom wagon in the Wales Velothon lol.

I have the same worry.

I did the pancake flat Great Escape audax in 11 hours flat (Real average 18.6 km/h) which suggests that I have a big problem at the uber lumpy Velo Wales. RidewithGPS is keen to make me feel good, by telling me that I averaged 21.8 km/h, which I know is just nonsense.

Fortunately I saw this coming and requested (and got) an early start time. I'm still not all that confident tho.

 

[Shut Up Legs]

But does it give average speed? I don't think it does, only moving average speed. Same as RWGPS.
Why would they omit this key metric, leaving us to work it out for ourselves, when they do show the never-never-land "moving average"?

My guess is that it's because none of these sites will admit Audax even exists.

 

[NorvernRob]

Have noticed recently there are people i ride have higher average speeds on Strava. I know i am faster than a lot of them, so i tweaked the wheel circumference in my wife's garmin, we did a ride on Sunday, Her garmin showed 6 miles more than mine and a higher average speed even though i got to the pub 20 mins sooner. So unless you have measured the wheel circumference and put that in your garmin, your average speed could be wrong, i don't really worry about other people's averages, but it's good to keep an eye on my own. At the end of the day you are only cheating yourself if you put the wrong circumference in. It's alright i am just thinking out loud.

I might be missing something here, but Garmins measure speed via GPS, so it doesn't matter if you're riding a 29er or a Penny Farthing the speed will be consistently accurate.

 

[Tin Pot]

Use MapMyRide.

I only heard about "moving time" nonsense when I met a Stravite.

Plus, you can wind up Homeopaths by opining the wonders of MMR.

 

[derrick]

I might be missing something here, but Garmins measure speed via GPS, so it doesn't matter if you're riding a 29er or a Penny Farthing the speed will be consistently accurate.

Not if you have a speed sensor on the wheel.

 

[Tim Hall]

I might be missing something here, but Garmins measure speed via GPS, so it doesn't matter if you're riding a 29er or a Penny Farthing the speed will be consistently accurate.

As I understand it, greater accuracy is achieved by using a wheel sensor as well, as it makes a better allowance for the errors due to a wiggly route between GPS meaurements.
(x post with derrick)

 

[deptfordmarmoset]

No. As I've said several times, I don't want to faff around working it out. Why should I? The technology is there to present it, but it doesn't.

I want ridewithGPS to present this useful metric, just like it does the fluffy, and of questionable meaningfulness, "moving average".

If I gave you a spreadsheet, you could simply enter your time and distance and the machine will do it for you. But what will you do when you ride for an hour, have lunch for an hour and ride back in an hour? Do you include the lunch in your average and knock 50% off your average? Or is it 2 rides? Tommy Godwin during his marathon year on a bike maintained a cumulative average speed of just over 8.5mph. However, he also slept, shat and showered at the same speed. Which is fair enough for the extreme nature of the challenge but not really a useful indication of his on the bike average speed. If you're keeping traffic delays out of the average, what exactly do you include?

 

[huwsparky]

Personally I think a moving time is more useful than an elapsed time. Generally I don't really stop but the average if I did stop for a chat or coffee etc would put it wildly out against my other rides which would render that information useless if I was actually bothered about number crunching.

BTW I don't really care how fast i go but you may as well not even have an average time if you take into consideration total elapsed time, it wouldn't mean anything for analysis purposes.

 

[Dogtrousers]

If I gave you a spreadsheet, you could simply enter your time and distance and the machine will do it for you. But what will you do when you ride for an hour, have lunch for an hour and ride back in an hour? Do you include the lunch in your average and knock 50% off your average? Or is it 2 rides? Tommy Godwin during his marathon year on a bike maintained a cumulative average speed of just over 8.5mph. However, he also slept, shat and showered at the same speed. Which is fair enough for the extreme nature of the challenge but not really a useful indication of his on the bike average speed. If you're keeping traffic delays out of the average, what exactly do you include?

As I said, I don't want to faff around with a spreadsheet or a calculator. I don't want to have to work it out for myself. I don't see why the ride logging sites such as ridewithgps or strava only present moving average, when they have the figures there to present the real average. What's more, they have the cheek to call it "average speed" when it's nothing of the sort. It's a made-up figure just to flatter me.

Take the example I gave above. It's massively unlikely that I could maintain 22km/h in a 200k time trial, yet ridewithgps has managed to scrape together enough brief stops to make it look like that. The reality is that, even if I scrimped on the stops, and got lucky with the traffic lights or rode on closed roads, the lack of rest would get to me, and by reducing my rest times the overall average would come down. So the silly moving average figure that RWGPS gives me is misleading when I come to consider whether I can get round Velo Wales inside the cutoff. I know that the cut-off speed of 19km/h is going to be unlikely/very hard, and I know this by calculating real averages and ignoring moving ones.

If I took an hours lunch in the middle of a two hour ride, I'd just accept that it was a slow ride, because it was. I was out for 3 hours.

The only thing you can predict from a moving average is how long it would take you to do a given distance in an imaginary world where there are no traffic delays and you are immune to getting tired and needing a rest. It sounds like a nice world but sadly I'm unlikely to get to visit it.

 

[deptfordmarmoset]

As I said, I don't want to faff around with a spreadsheet or a calculator. I don't want to have to work it out for myself. I don't see why the ride logging sites such as ridewithgps or strava only present moving average, when they have the figures there to present the real average. What's more, they have the cheek to call it "average speed" when it's nothing of the sort. It's a made-up figure just to flatter me.

Take the example I gave above. It's massively unlikely that I could maintain 22km/h in a 200k time trial, yet ridewithgps has managed to scrape together enough brief stops to make it look like that. The reality is that, even if I scrimped on the stops, and got lucky with the traffic lights or rode on closed roads, the lack of rest would get to me, and by reducing my rest times the overall average would come down. So the silly moving average figure that RWGPS gives me is misleading when I come to consider whether I can get round Velo Wales inside the cutoff. I know that the cut-off speed of 19km/h is going to be unlikely/very hard, and I know this by calculating real averages and ignoring moving ones.

If I took an hours lunch in the middle of a two hour ride, I'd just accept that it was a slow ride, because it was. I was out for 3 hours.

I'm still not getting your point of view. What is the ''real'' average? Should a commuter include the hours before returning at the end of the day if getting off the bike for an hour's break is included? One thing I do know: for my in mostly flat town rides, even on a light bike, I struggle to keep a rolling average of 12 mph. On a run a couple of years ago on a heavier bike, I got to Whitstable via the North Downs, with a rolling average higher than 14mph. Out of town is faster even for an old slowcoach like me. For Velo Wales, you'll need to give yourself an idea how much time you have to get to the cut off points - and then keep rolling as continually as you can without killing yourself.

 

[mjr]

Personally, if at home, I start my tracker at the point where I would pick up the car keys if I was going by car (so before picking up the rack pack or bar bag and hooking/unhooking the basket/panniers if needed) and stop it when about to enter the door of my destination (so after parking). I find that overall time the most useful for planning other rides of similar distance and terrain (including urban terrain - lights and so on). A moving average would be misleadingly quick.

 

[PhilDawson8270]

Personally, if at home, I start my tracker at the point where I would pick up the car keys if I was going by car (so before picking up the rack pack or bar bag and hooking/unhooking the basket/panniers if needed) and stop it when about to enter the door of my destination (so after parking). I find that overall time the most useful for planning other rides of similar distance and terrain (including urban terrain - lights and so on). A moving average would be misleadingly quick.

It's only strava that gives this "moving time"

But if you're planning for trips, you can look at the elapsed time on Strava, or you can look at it on your gamin itself, or on garmin connect.

The only thing that Strava doesn't give itself is actual average speed, but you can get that from the GPS device itself, or via the garmin site.

 

[Drago]

Ignore all the hours I spend sweating, grunting and pushing, and I spend only 12 seconds a day on the loo.

 

[Archie_tect]

I use Cyclemeter so I can put rides on mycycinglog, half the time I forget or it doesn't record them properly, or the battery dies, but other than that I'm not bothered.

 

[Globalti]

Anybody who uses "real" averages must be completely off their chump. What happens if you stop for half an hour for lunch or to fix a breakdown or whatever? The meaning of average speed is lost. You can only compare your speeds on different trips by comparing the average speed at which you moved. My cycling buddy is faster than me up hills so sometimes has to stop and wait for me so when we cycle for an hour on a Wednesday evening he averages 17 mph and I average 16.8. This is a true and accurate measure of our relative speeds.