How Strava works

Chuchote

How does Strava show your identity if you sign up and show others a run? (In other words, does it use a nickname or identify you.) And is it easy to exclude an area around your home, and how wide an area do people exclude?

Veloce150

Chuchote wrote:

How does Strava show your identity if you sign up and show others a run? (In other words, does it use a nickname or identify you.) And is it easy to exclude an area around your home, and how wide an area do people exclude?

You can use a nickname and you can choose the visiibility of your runs. You can choose who can 'follow' you.

Part of the fun of being 'public' is seeing who else was on your route and how they did. That said the comparisons are based on time and not watts/kilo, so their rankings should not be taken seriously.

You can set an area (say 1.5m) to avoid pinpointing your house.

Probably best not to publish detailed descriptions of your expensive bike. Keep it generic.

Beasty

You can keep all or selected rides completely private, meaning no-one else can see them. That way you can see how you compare with others by looking at leaderboards, and also keep track on your own performances over specific segments/rides

You can also create your own segments which again can be public (in which case anyone who does not keep their rides private can appear on the leaderboard), or private, in which case only your own rides appear

nialljf

if i may jump in here, how accurate are strava's estimated power algorithms?

Beasty

nialljf wrote: »

if i may jump in here, how accurate are strava's estimated power algorithms?

Pure guesstimates. Can't take into account factors such as wind, position on the bike, or any drafting benefits. The only time I think you could consider them something approaching a decent guess is when you are doing a circuit solo

rtmie

Beasty wrote: »

nialljf wrote: »

if i may jump in here, how accurate are strava's estimated power algorithms?

Pure guesstimates. Can't take into account factors such as wind, position on the bike, or any drafting benefits. The only time I think you could consider them something approaching a decent guess is when you are doing a circuit solo

Interesting, I always took them to be pure guesstimate in all cases. Given that the majority of my spins are solo, is there a suggestion that for solo spins they are of some use/

Beasty

rtmie wrote: »

Interesting, I always took them to be pure guesstimate in all cases. Given that the majority of my spins are solo, is there a suggestion that for solo spins they are of some use/

On the flat with no wind maybe. Otherwise you may as well plug some details into the Bikecalculator to get a better guesstimate

magicbastarder

one thing i noticed soon after starting to use strava is that their estimates for calories burned are much lower - typically 50% lower - than other online calculators.
e.g. if you search for calorie calculator in google, and pick any one at random, and tell it you weigh 75KG and have done one hour of cycling at 25km/h, its estimate for calories burned would be typically twice what strava would tell you for a similar recorded spin.

Veloce150

nialljf wrote: »

if i may jump in here, how accurate are strava's estimated power algorithms?

About a third less than what my power meter would say.

On a similar note, I've ridden with people who are just using the app and for the exact same ride they'd have a stat for climbing twice mine.

Strava is one part game and one part training aid.

The best bit is being able to compare your power output on climbs or rides on different dates, ignoring time. This is only useful if the power data is real, nor estimated.

magicbastarder

Veloce150 wrote: »

On a similar note, I've ridden with people who are just using the app and for the exact same ride they'd have a stat for climbing twice mine.

is this possibly due to the deliberate inaccuracy built into civilian GPS though?
strava can only use the GPS data being fed from the device - any such inaccuracy is from the device, not strava.

Veloce150

magicbastarder wrote: »

is this possibly due to the deliberate inaccuracy built into civilian GPS though?
strava can only use the GPS data being fed from the device - any such inaccuracy is from the device, not strava.

by a factor of 200%?

magicbastarder

here's an altitude accuracy test where the raw data output by a single device varied by over 100m on a flat track. most devices recorded a greater than 25m elevation change - which if you allowed for a 1m change in actual altitude, would be 25 times that covered.
http://www.singletracks.com/blog/gps/gps-elevation-accuracy-test-smartphone-apps-vs-dedicated-gps/

granted, the inaccuracies will tend to cancel out over a longer ride with greater amounts of climbing. an interesting comment though:
"When it comes to calculating elevation gain and loss over a ride, online services are important tools for improving accuracy. Two of the units in our test–the Garmin Edge 500 and Magellan Cyclo505–produced fairly-accurate raw elevation data, but the data still wasn’t nearly as accurate as the “calculated” elevation gain and loss generated by Strava."

as mentioned, strava has to work with the data provided to it. on criticism i would have of strava is that when a segment is created, it uses the GPS data recorded by the person creating the segment, which can be inacurate; but there's no further finessing of that data as more cyclists travel the segment, which might average it out to something closer to reality.
e.g. there's one segment i know which appears as a 142m climb for me on strava, but in reality i think is about 10m.

it's the road out of the naul towards oldtown; the segment is 0.4km and is listed as 29.7%. it's called R122 climb

Macy0161

magicbastarder wrote:

one thing i noticed soon after starting to use strava is that their estimates for calories burned are much lower - typically 50% lower - than other online calculators.

Strava still gives me a bit more than my Garmin though for cycling. Both have access to the hrm data, not sure strava uses it for calories though. Strava way higher for running than Garmin though.

tunney

magicbastarder wrote: »

is this possibly due to the deliberate inaccuracy built into civilian GPS though?
strava can only use the GPS data being fed from the device - any such inaccuracy is from the device, not strava.

The civilian v military gps difference ended since years ago. They are both are accurate as each other.

magicbastarder

i was reading up on that - one thing retained is that if your speed exceeds something like 1000mph, civilian GPS stops functioning. i've only ever hit that speed coming down off howth with a stiff tailwind, though.

apparently the inaccuracy with altitude and GPS is a fundamental artefact of the way GPS works.

Dermot Illogical

Veloce150 wrote: »

On a similar note, I've ridden with people who are just using the app and for the exact same ride they'd have a stat for climbing twice mine.

I did a spin around Inishowen earlier this year, including an overly optimistic attempt at Mamore Gap, and the elevation on Strava (using a Garmin 500) matched perfectly the elevation recorded by the altimeter in my Fitbit.
Strava mobile app might be different? It's possibly using a triangulation/gps mix.

Unknown Soldier

Chuchote wrote: »

How does Strava show your identity if you sign up and show others a run? (In other words, does it use a nickname or identify you.) And is it easy to exclude an area around your home, and how wide an area do people exclude?

To set up Strava it is better to do it on a PC, rather than use the app. Much more settings available, and a lot easier too.

Veloce150 wrote: »

On a similar note, I've ridden with people who are just using the app and for the exact same ride they'd have a stat for climbing twice mine.
.

I find that Strava robs people of about 15-20% of climbing.

A lot of the Garmin Edges come with barometric altimeters, which most of the time are fairly accurate. But yes there can be inaccuracies between people and devices. I tend to use set elevation points for my gaff/work and a few places. Otherwise you have to give it a bit of time to acclimatise.

A handy way to see Elevation data for "studying" is if you do a Sportive/spin with a lot of people, and using Strava on a phone look at the people you did the spin with.

The summary screen is usually their name, distance and metres climbed.

Having just checked the Reservoirs Dogs spin I did a few months ago, most that I rode with did in and around 1,950 to 1,975 over the 123km'ish route. But there are a few showing the same distance but 1,750'ish metre climbed.

I'd say they were using their phones/strava.

All that aside, when I sideload my ride data from Strava to RWGPS I get totally different climbing figures

moonshadow

magicbastarder wrote: »

i was reading up on that - one thing retained is that if your speed exceeds something like 1000mph, civilian GPS stops functioning. i've only ever hit that speed coming down off howth with a stiff tailwind, though.

apparently the inaccuracy with altitude and GPS is a fundamental artefact of the way GPS works.

I presume you got KOM on that segment

bren_mc

When you upload to Strava initially, the elevation reading has a little link beside it with a question mark, indicating that it may not be accurate. When you click the link, you'll see a "Correct Elevation" button. Clicking this will cause the figure to be recalculated (more accurately I presume).

https://support.strava.com/hc/en-us/articles/216919447-Elevation-for-Your-Activity

Wishbone Ash

bren_mc wrote: »

When you upload to Strava initially, the elevation reading has a little link beside it with a question mark, indicating that it may not be accurate. When you click the link, you'll see a "Correct Elevation" button. Clicking this will cause the figure to be recalculated (more accurately I presume).

https://support.strava.com/hc/en-us/articles/216919447-Elevation-for-Your-Activity

I find that correcting the elevation often reduces it by a huge amount - sometimes up to 75%. It particularly noticeable on a group ride where my elevation sometimes ends up at a fraction of my club colleagues.

On the other hand, if corrected on rides at high altitudes, such as riding in the Alps etc. it sometimes adds significantly to the original figure - a ride with 3,000m becomes 3500m or higher after 'correcting' it.

Wishbone Ash

Wishbone Ash wrote: »

.....On the other hand, if corrected on rides at high altitudes, such as riding in the Alps etc. it sometimes adds significantly to the original figure - a ride with 3,000m becomes 3500m or higher after 'correcting' it.

Hence the reason why Strava has a club called 'Honest Climbers'.

https://www.strava.com/clubs/honest_climbers

MB Lacey

Chuchote wrote: »

How does Strava show your identity if you sign up and show others a run? (In other words, does it use a nickname or identify you.) And is it easy to exclude an area around your home, and how wide an area do people exclude?

You decide what your username on Strava is - it can be made up or your actual name. This will be your Strava name - which you can edit at anytime.

Yes very easy to set up numerous privacy zones around home, work, wherever - it's either a 500m or 1km circular radius I can't remember.

Wishbone Ash

MB Lacey wrote: »

......Yes very easy to set up numerous privacy zones around home, work, wherever - it's either a 500m or 1km circular radius I can't remember.

It used to be 500m or 1000m but now it's 200,400,600,800 or 1000m.

You won't register on any segment which falls partly within a privacy zone. A privacy zone doesn't have to be a specific zone around your actual home/work address. You can create the zone using another address once your home/work is still inside. I live on the edge of my privacy zone and created it thus as otherwise I wouldn't be included in a few popular segments near me.

2RockMountain

bren_mc wrote: »

When you upload to Strava initially, the elevation reading has a little link beside it with a question mark, indicating that it may not be accurate. When you click the link, you'll see a "Correct Elevation" button. Clicking this will cause the figure to be recalculated (more accurately I presume).

https://support.strava.com/hc/en-us/articles/216919447-Elevation-for-Your-Activity

Kinda strange, no? Why wouldn't it just correct all uploads, without requiring you to correct it?

On a more general question, how does Strava treat time stopped at red lights? Does this count in time taken towards segments?

Dermot Illogical

Yes

Wishbone Ash

2RockMountain wrote: »

...how does Strava treat time stopped at red lights? Does this count in time taken towards segments?

Yes, it would have to as to do otherwise wouldn't make sense. On a long segment, a person could rest during it which wouldn't be fair to someone who didn't rest.

magicbastarder

2RockMountain wrote: »

On a more general question, how does Strava treat time stopped at red lights? Does this count in time taken towards segments?

i think for your overall average speed for a cycle, time stopped at lights is not counted. but as mentioned above, it is included for segments.

you'll see an 'elapsed time' and a 'moving time' for an upload.

crosstownk

2RockMountain wrote: »

On a more general question, how does Strava treat time stopped at red lights? Does this count in time taken towards segments?

As already stated, the time is the complete time from the beginning of the segment to the end.

What puzzles me is why segments are created with multiple traffic lights. It makes no sense to me.

magicbastarder

also, i don't understand people who create really short segments in urban areas.
most segments i pay heed to are generally my own private ones - often loops i'd do, so you get a good sense of how fast you're going; because it's a loop, the effects of wind or altitude changes would generally cancel out.

Wishbone Ash

crosstownk wrote: »

...What puzzles me is why segments are created with multiple traffic lights. It makes no sense to me.

I suspect they are created by someone who manages to get through several sequences of green lights on a deserted route at 4am.

crosstownk

Wishbone Ash wrote: »

I suspect they are created by someone who manages to get through several sequences of green lights on a deserted route at 4am.

Possibly. Although I did create a segment that covers most of my commute across the city just to be able to see the shortest/longest times it takes me to get to and from work. It's a private segment and not one I'd compete against - it's just to satisfy my own curiosity.

Incidentally, with regards to privacy zones, if you pass through your privacy zone rather than stop, you will be awarded any segment times that are relevant. However, if you stop/save the ride then the segment won't count.

Luxman

I saw a segment for the Gerry brannigan spin. It was 58k or something mad. The whole spin. I think the WW200 is a segment too

Chuchote

How anonymized is the "Include my anonymized public activity in Strava Metro and Heatmap" setting? Would your rides and your routine be available from it to someone who wanted it?

RobertFoster

Both are completely anonymous. The heatmap shows popular routes based on the number of rides recorded. Metro is similar, but also provides more in depth data (number of cyclists, type of ride, wait time at intersections). It's still group based though. You can't pick out individual details.