Strava and Power

okgo

That Jim Ley, gets everywhere!

I am going to do a comparison tomorrow morning using a real powertap on my garmin, and using the phone app at the same time with its estimates. The one thing it obviously cannot get right is the wind.

doyler78

okgo wrote:

That Jim Ley, gets everywhere!

I am going to do a comparison tomorrow morning using a real powertap on my garmin, and using the phone app at the same time with its estimates. The one thing it obviously cannot get right is the wind.

Well make sure you weigh yourself, your bike and anything that will be attached to you or your bike so that you get the total weight involved correct (yeah the scale could be wrong but you can only work with what you have)

Alex_Simmons/RST

t5nel wrote:

Alex_Simmons/RST wrote:

t5nel wrote:

What do people make of the calculators at analyticcycling.com? For hill climb they tend to give me results about 15% lower than strava. Rest of the time wind has got to be the biggest factor that make strava inconsistent. I am pretty sure it is about as good as it can be given the dataset they have.

I reckon you must be about 4'7" tall. It's about as good as I can get with the dataset I have.
What's the point of estimating things so badly?

I do not make money by selling power meters so I will let others make up their own mind about objectivity

I might sell power meters, but my information is based on evidence, which is always the best means for people to make up their own minds. At least I declare my interest. I don't know yours.

My point isn't about power meters, but rather that people might (and do) rely on these wildly inaccurate estimators to aid their weight control plans.

As for some reasonable guidance on rate of energy expenditure not requiring power meter, well I provide that sort of info too you know:
viewtopic.php?p=17563529#p17563529

which is of course based on knowledge of actual data from power meters.

RoadMeridaBen

Alex_Simmons/RST wrote:

I can only imagine Strava power numbers exist purely for some light comic relief.

Are you saying I didnt average 520 watts for my last 25 mil TT at the weight of 66kg? :shock:


:P

Alex_Simmons/RST

RoadMeridaBen wrote:

Alex_Simmons/RST wrote:

I can only imagine Strava power numbers exist purely for some light comic relief.

Are you saying I didnt average 520 watts for my last 25 mil TT at the weight of 66kg? :shock:


:P

tweetymike

Yesterday I had a full VO2Max and blood lactate threshold test done at Winchester Uni, as part of a critical power research program I'm a lab rat for. It said my threshold for blood lactate is around 240W. This is on a £35k SRM bike with a darned fancy blood analyser thingy and a sample being taken every 3 minutes as the power increased.

Going back over my 'baseline' 1hr / 30km Strava segment that I do maybe twice a month, Strava has consistently pegged me at between 230 & 240W over the hour, over the last 6 months rides. It's a ride I do as fast as possible, and I've got a good feel now for how hard I can go initially on the flat without blowing up 40 minutes in on the hills.

So for my money, for a *free* app that only needs a phone, Strava is pretty good at power estimation. This is potentially invaluable info, and if I had a HRM that could show accurate average power (or a correlating metric) I'd definitely use it on long rides to ensure I was staying away from The Wall.

Trev The Rev

You might as well use tarot cards as use Strava power figures.

Just because one chap has had a real test done which comes out similar to Strava power numbers over one particular route does not mean the numbers are anywhere near worth even looking at.

I am often critical of power meters and the way they are used by some people but honestly you get what you pay for. Strava power numbers are free entertainment. Power numbers must be real, accurate and repeatable, if they are not, they are worthless.

Herbsman

Anyone know if Strava calorie estimates are more accurate when you upload activities with power data?

jimmurray

As it's derived from the uploaded power figures it should be 'spot on'.

Herbsman

jimmurray wrote:

As it's derived from the uploaded power figures it should be 'spot on'.

It'll only be within about 10% due to variations in metabolic efficiency

ShutUpLegs

Herbsman wrote:

jimmurray wrote:

As it's derived from the uploaded power figures it should be 'spot on'.

It'll only be within about 10% due to variations in metabolic efficiency

I'll wait for Trev to confirm its relevance

Trev The Rev

ShutUpLegs wrote:

Herbsman wrote:

jimmurray wrote:

As it's derived from the uploaded power figures it should be 'spot on'.

It'll only be within about 10% due to variations in metabolic efficiency

I'll wait for Trev to confirm its relevance

I don't count calories, I eat by feel.

It is the most amazing system. It automatically tells you when you need to eat and how much to eat, it takes into account all your training, no matter what sort of training, it has this amazing way of making you want to eat a wide variety of foods and makes sure you eat all the right things in the right quantities to maintain your correct weight.

Herbsman

Yes, but I'm not bothered about my weight, or calorie intake. I am just curious as I find it interesting to see how much energy I use during a ride. I 'eat by feel' too... i.e. eat when I am hungry and try not to eat crap like biscuits and crisps.

Trev The Rev

Herbsman wrote:

Yes, but I'm not bothered about my weight, or calorie intake. I am just curious as I find it interesting to see how much energy I use during a ride. I 'eat by feel' too... i.e. eat when I am hungry and try not to eat crap like biscuits and crisps.

Once you start looking at calories you start to interfere with the natural mechanism. Thinking about avoiding biscuits is the thin end of the wedge.

Herbsman

Trev The Rev wrote:

Herbsman wrote:

Yes, but I'm not bothered about my weight, or calorie intake. I am just curious as I find it interesting to see how much energy I use during a ride. I 'eat by feel' too... i.e. eat when I am hungry and try not to eat crap like biscuits and crisps.

Once you start looking at calories you start to interfere with the natural mechanism. Thinking about avoiding biscuits is the thin end of the wedge.

Oh my life. :roll:

I have never heard such rubbish. Show me some evidence that looking at calories burned during cycling, purely out of curiosity towards the energy requirements of the activity, will interfere with the 'natural mechanism'.

I never said avoiding biscuits was the basis of my diet. Do you want me to list every thing I eat so you can judge whether it's good enough?

Trev The Rev

Herbsman wrote:

Yes, but I'm not bothered about my weight, or calorie intake. I am just curious as I find it interesting to see how much energy I use during a ride. I 'eat by feel' too... i.e. eat when I am hungry and try not to eat crap like biscuits and crisps.

You "try not to eat biscuits and crisps", so you do admit you do eat crap. Now you are hooked on counting calories you are already in need of help.

Herbsman

Trev The Rev wrote:

Herbsman wrote:

Yes, but I'm not bothered about my weight, or calorie intake. I am just curious as I find it interesting to see how much energy I use during a ride. I 'eat by feel' too... i.e. eat when I am hungry and try not to eat crap like biscuits and crisps.

You "try not to eat biscuits and crisps", so you do admit you do eat crap. Now you are hooked on counting calories you are already in need of help.

You need to take a good hard look at yourself.

Gruge

I have one big fat 15kg hardtail and a slimmer 10kg drop bar. I weigh an entirely porky 95kg. Over my 25km commute, the meter reckons I average 270W on the hardtail but 141W on the drop bar. That doesn't even work even if you're just taking into account the bike. If you count my weight then 5kg difference between bikes should barely affect things.

BrandonA

It amuses me that Strava have a premium feature which reviews your power. Unless you have a power meter it is completely worthless, if you've got a power meter it is probably worthless as you'd use a specialist site to review your power.

Alex_Simmons/RST

madasahattersley wrote:

BrandonA wrote:

It amuses me that Strava have a premium feature which reviews your power. Unless you have a power meter it is completely worthless, if you've got a power meter it is probably worthless as you'd use a specialist site to review your power.

Not necessarily, the power analysis tools on Strava are for the large part every bit as good as WKO+ etc, but much cheaper and much more sociable. Cyclists aren't a very sociable bunch though, better keep those power numbers hidden!

Until your subscription goes beyond 5-6 months, then you are paying more.

The feature currently does not permit historical changes of threshold power, which is a pretty basic requirement for many of the tools they have copied.

Do they at least acknowledge they are copying other people's work?

maryka

madasahattersley wrote:

BrandonA wrote:

It amuses me that Strava have a premium feature which reviews your power. Unless you have a power meter it is completely worthless, if you've got a power meter it is probably worthless as you'd use a specialist site to review your power.

Not necessarily, the power analysis tools on Strava are for the large part every bit as good as WKO+ etc, but much cheaper and much more sociable. Cyclists aren't a very sociable bunch though, better keep those power numbers hidden!

The only Strava power feature I find more useful than standalone power software (WKO, GC) is the peak power chart for each ride showing me exactly on the map where any given duration peak power happened. But you still can't create laps post-ride and you certainly don't get anything beyond average power for any selected section of your ride -- no intensity, no X-power/NP, no variability numbers.

But there are so many things wrong with the Strava power analysis that when my yearly premium subscription expired back in Feb I didn't bother renewing it. A few off-the-chart rogue data points and your low duration power "records" are forever sky-high and can't be fixed. I've got 5 years worth of data in Strava but they only carry the last 3 years for power comparison purposes (useless to me as 2010 was my last best season before having my daughter, trying to compare 2013 rides with 2010 ones is what I want now and can't have). Also as Alex says, no way to change FTP over time.

The worse is that they have spent loads of time developing useless features like the "intensity bar" and don't seem to pay much attention to user-requested features like lap data for turbo rides (yes, if you hit the lap button on a turbo ride then upload, you will not have any laps to look at...) And don't get me started on how they've ruined the graphs and data for runs!

If I pay for premium again it will be for stuff like the new Routes feature, certainly not for power features.

And yes, if you don't have a powermeter, the estimated power numbers are pretty useless.

GGBiker

Personally I have found it to be surprisingly accurate vs my powertap, usually within 10-20w on climbs which is excellent for an estimate. The calculations are not that complicated as long as the data input is accurate. Wind is probably the biggest factor but mostly I'm riding in winds from 0-15mph so not a very significant factor. Also wind is less important on climbs.

I have checked a few TT times against the calcs on an app called Bikecalc, it is surprisingly accurate compared to my powertap and good for working out the power required to hit a certain time.

Alex_Simmons/RST

GGBiker wrote:

Also wind is less important on climbs.

It's not unimportant though, indeed wind has a large impact on climbing speed:
http://alex-cycle.blogspot.com.au/2013/07/windbags.html

GGBiker

Most people would be happy with +/- 10% for an estimate, especially a free estimate.

If I look at power figures on Strava and compare where actual power values (from a meter) lie in the results hierarchy vs the estimates they mostly look sensible, especially if you know the riders and their approx weight.

Of course it is not particularly useful for training purposes. Strava is useful for indicating increasing fitness on runs you do frequently as you are unlikely to see constant improvement due to improving weather conditions alone!