Ive just been out and pretty much powered round a 20 mile on road loop on my mountain bike bit of a tail wind on first half and head wind on way back, was trying to keep average speed above 20mph but fell slightly short because of headwind and came in at 19.6 just.

average power was 339w with segments on strava done with est power of over 500w but when I check against some people I am following they have done that segment faster but with half the est power – the only thing I can think of is that they must have either a. been holding on to the back of a truck (marty mcfly style) or b. there was a hell of a tailwind behind them.

tbf Ive just looked and they did it on a road bike and I was on a full sus 26″ wheel and was only 5 seconds slower – is there a formula for working out the time/speed difference between 700c/29″ road tyre/mtb tyre and 26″ mtb tyre?

is gamrin/strava inaccurate or is there another reason for this? am I alone in thinking that strava is not a wholey accurate measure of ones speed and power?

Strava power estimates are notoriously inaccurate. An average of 20 mph is good going on a full suss MTB but 340 Watts for an hour would put you at the top end of amateur racers. If you’re just a regular reasonably fit rider I’d say a good hour effort would be more like 200-250W.

thanks, is power another good method for measuring training performance? should I look at getting a power meter to measure more accurately?

I have been back on the bike just 1 year now after having smoked for the last 20 odd years, going to start xc racing next year after a few trial races earlier this year.

Commuting a bit slower than you were going, at a low Z2 pace, I reckon my MTB (27 km/h, fast-ish knobblies at 30 psi) is about 3 km/h slower than a road bike (30 km/h 23 mm slicks, 90 psi). This is gauged over a few rides. The effort feels the same and my HR is pretty much the same. Just for information, the powertap on my road bike tells me that the average power at that speed is roughly 175 W.

Wind speed and direction can make a huge difference to how much power you need for a given speed. Strava doesn’t know anything about this, so its estimated power could be way off.

Also depends on the drag from your tyres, how aerodynamic you are, your weight, and the bike weight including luggage. Plus any GPS inaccuracies etc. So it can’t really be calculated accurately (unless you have a power meter).

so the fact that people go faster than me on a segment (2 min)(only by a few seconds) but use at least a 100w less power does not mean they are necessarily faster than me? right?

It could be that there was a tail wind, they were on a 29er or road bike etc.

so the only way to truly measure whether I am faster than these people is to race them? I suppose I need to do the same route a few times a month and record any weather conditions, dietry consumption prior to the ride, type of tyre i used etc and compare my own results to see if I am getting faster?

I think it may also assume that if you select bike type as mountain bike that you are riding off-road and the pretty much un-estimatable rolling resistance that goes with that.
As an example a road ride last weekend it says I averaged 153 watts, if I change the bike to a mountain bike it says 278 watts.

Assuming that there isn’t much wind and you’re riding on roads on a road bike and you’re getting accurate height data, and you’re not just measuring over a short segment then strava will produce a reasonably accurate figure. There’s a lot of caveats in that though 🙂

well Strava can only calculate on basis of what data you put in, if the bike or rider is heavier / lighter than stated, that’ll affect things too. Don’t stress it!

yeah dont think ill stress it as power meters in whatever variety cost an arm and a leg!!

ignore it, ill jump in here, i have a MTB power meter

this is from last Tuesday, first pic, is a lap of the Sherwood Pines Red with a Stages Powermeter on my Tallboy C

second pic, is 20 minutes later, on a Surly Krampus, no powermeter

same Strava segment “Sherwood Pines Red Route 2014”

so the fact that people go faster than me on a segment (2 min)(only by a few seconds) but use at least a 100w less power does not mean they are necessarily faster than me? right?

I think the Strava power algorithms for road are getting better these days, and given fairly ideal conditions, if you’ve entered in all your data correctly (weight mostly), and are reasonably aero on a bike, it’ll give a figure that’s not a million miles out from what my power meter says (I’ve got a Power2Max, and a PowerTap, also have a PowerCal but that’s a bit different.) However if any of those don’t hold true, especially off road, then don’t even bother.

Edit: dirtyrider… 100 watts 🙂

strava power average doesn’t take into account headwind/tailwind ect. It uses a power formula which is different for road and mtb rides. If you use your mtb on a road ride it assumes your riding off road. You can have fun editing your rides and swapping bikes around. A 190w road ride becomes 345w when I swap my road bike for my mtb.

Dirtyrider: Is the surly listed as a mtb in your strava profile ? If not the default is road and strava will assume your ride is just a slow road ride.

^^ ah no, its not, now updated, not sure that function was there when i added it

however,

here is the same segment from late June 2014, same bike, Tallboy C, listed as a MTB in my profile, battery was flat in my stages

OP.. You can do a 20mph average on a mtb on the road for 20 miles….
Thats awesome.. post up our strava results

OP.. You can do a 20mph average on a mtb on the road for 20 miles….
Thats awesome.. post up our strava results

not sure how? heres a link to my ride if it works:

link to strava ride

Oh.. thought it was going to include hills 😆
Funnily enough was reading this the other day as im a very light rider and was also interested how my watt effort was so low on certain climbs in comparison.

How Does Strava Calculate Power Data?

Oh.. thought it was going to include hills

lol there were a few small hills, well small slopes. there was a headwind though lol

will have a read of that article though

If you want something that gives you a better idea of power but is much cheaper than a proper power meter then you might want to consider a PowerCal. It uses heart rate and works on rates of change to calculate a power figure. I was very sceptical but it got recommended on here and the dcrainmaker review was positive. I needed a new HR strap anyway so thought I’d take a punt. On the road the 30 second power figure isn’t too far off what a power meter says, and the NP for the ride is even better, much much better than the Strava figures, at least IME. Handy to get the TSS and NP if I’m out on a bike without a proper PM. Anyway if you need a new HR strap anyway then it’s well worth considering as it’s not much more cash (70 quid from Evans.)

It’s a good effort anyway hills or not Darren 😀 so it gets some kudos from me.

cheers taxi

Mr Blobby: Thanks for the Powercal write up. I’m in that exact position: need a new HRM / strap anyway and thought the Powercal might be worth a go for a few extra quid. I’m not expecting PM accuracy, more like some rough figures for TSS etc. on top of the HRM data.

jimification, that’s just what I use it for when riding without a proper PM. I get the impression that how close it gets can vary quite a bit from person to person, but it’s pretty good for me, maybe I’m just a good fit for their algorithms. They don’t claim PM accuracy, but when riding along on the flat at a steady pace the 30s average does get close enough for me to generate meaningful NP and TSS values. I think it’s closer than the TSShr that TrainingPeaks generates, and way better than the power figures from Strava.

I am still puzzled by how it does it though as, for example, you could have someone who’s powerful doing a steady 350W at 150bpm, and someone who’s not going 200W at 180bpm. I don’t know how it works out which you are but it seems to do a pretty good job of it.

I have a powertap & a powercal, I did a 1hr 30 training ride using both (2 separate garmin 705’s) with a 20 min sweetspot in the middle, the powercal was 5 w different on on the SS and within 1w on the whole ride.

I superimposed the 2 in training peaks, whilst roughly similar the Powercal displayed much greater peaks & troughs throughout the ride as well as some comedy super high power readings for no apparent reason.

This was over one ride, a definitive test would be over several rides, having said that I did a stage race in SA earlier in the year and used a powercal, the numbers were fairly consistent and about what I would expect (based on my training sessions over 2 years on a powertap)

The powercal is good for a basic overview on longer sessions, I wouldn’t bother so much for shorter efforts though, i’ve tried it for 5min efforts & the readings are all over the place, it gets more accurate for average power over longer rides.

Not tried the Strava power thing, looks a bit too vague to me

I have a powertap & a powercal, I did a 1hr 30 training ride using both (2 separate garmin 705’s) with a 20 min sweetspot in the middle, the powercal was 5 w different on on the SS and within 1w on the whole ride.

That is pretty impressive, you must be an even better fit than me! Times when I’ve done similar rides with both the PC gets to within about 10 to 20 watts.

I superimposed the 2 in training peaks, whilst roughly similar the Powercal displayed much greater peaks & troughs throughout the ride as well as some comedy super high power readings for no apparent reason.

Yes, is a bit funny. I’ve had the 5 sec and the 30 sec side by side on the garmin and the 5 sec is all over the place but the 30 sec seems to come out about right (at least for longer steady efforts.) Much like the dcrainmaker review shows.

Dang, that’s interesting, will investigate further.