tedder987

Is there anyone with a power meter that has calibrated it against the strava average power estimates from the app? Just curious how (in)accurate they are.

rubiksoval

Consider that Strava doesn't know if you have a headwind or tailwind, if you're in an aero tuck, or sitting bolt upright, if you're wearing a skinsuit or a parachute, if you're in a group or solo, if you're riding gatorskins or corsa speeds, if the road is silky smooth or gravel, etc., etc., etc.

In short, pretty much useless.

tedder987

Yeah I figured that the precision would be lousy but is it also inaccurate? Strava does seem to know what way the wind is blowing though since it lists that on my rides along with the outside temperature.

rubiksoval

Yes. For all of the same reasons unless you ride with the exact same clothing/equipment on the exact same course in the exact same position every time. And maybe up a long, steady climb. That'd probably be the best for accuracy.

Seattle Forrest

How accurate it is defends on a lot of factors and some of them change. How useful it is depends on what you want the numbers for.

In general the guesstimated average power over the course of an entire ride isn't that bad. It's not respected enough for bragging rights at the pub unless your friends use a Strava power meter too. Most people who buy a direct force meter do it to measure their effort during intervals and other workouts. Strava is absolutely not reliable for this purpose. If you want it for calories, it's probably better than what you'd get from HR.

alcjphil

The power estimates it gives when I do mountain bike rides are a joke. It takes much more effort to do 15 kph on singletrack than it does to do 25 kph on the road on my road bike
yet my road rides give much higher estimates than my mountain bike rides

caloso

It's completely meaningless.

WhyFi

And you have your bikes set up in Strava as mtb and road, respectively? I don't have a PM on my gravel bike, but once I change the ride info to the gravel bike from my road bike, which is the default, the average typically isn't too far off of what I'd guess based on HR and RPE.

guachi

My Strava estimates all greatly underestimated my rides compared to the same ride with a power meter.

rower2cyclist

Strava power estimates are pretty useless, a joke...I mean like Seinfeld would've shot an entire episode on them if the show was still around.

tedder987

So I have a MTB as well as the road bike and both are in strava as such. part of why I asked the question is that on moderate MTB rides I would get average power estimates about 20% higher than for much more intense road rides. Also for reasons I cannot explain, they are usually 10-20% lower on the road on cold days (40-50 F) than on war ones, even when the effort feels higher.

If they are consistent and repeatable (and have some decent correlation with a real power meter) then I might use it to gauge how my effort was relative to how it felt, but right now I am leaning towards it being not useful even for that purpose.

dmanthree

It usually estimates my power really low, so it must be way off.

Kidding aside, the power readings bear no resemblance to reality. They might be spot on every once in a while, but that's the old "stopped clock is right twice a day" effect.

WhyFi

If you were to ride the same course regularly, a stopwatch would likely be the better tool, but if you mix up ride courses like most people, the average power estimate might provide some kind of measuring stick to show performance trends. I wouldn't take it at face value for any given ride but, as a tool, it's probably worth slightly more than what you paid for it. Slightly.

But you sound like you want a power meter, so just get a power meter.

Photobadger

It would be better if it wasn’t there. I have ridden the same bike with and without a power meter and Strava isn’t even in the ball park. Way low for me (not sure the formula understands what a block a 200cm rider really is &#128527. As mentioned many times above, lots of reasons this is the case—Strava isn’t bad, the problem just requires data it doesn’t have.

caloso

I remembered that a couple of years ago I did an experiment where I was riding into a steady headwind. I ran Strava on my phone so it just estimated power. Then I uploaded the same ride from my Garmin which had power data from my Quarq PM and compared the estimate to the actual power data. It was off by a ton.

https://www.bikeforums.net/19545157-post30.html

tedder987

Caloso, thanks, that's exactly the sort of actual data I was hoping to see. Photobadger, I hope I have the same problem with underestimates, I am ~190cm tall and I can't really be as weak as Strava says (I hope)

WhyFi

Keep in mind that this is purely in to a headwind and without any corresponding tailwind overestimate - pretty much worst possible scenario.

That said, it seems like you would like a power meter, so go buy a power meter (I'm an enabler ).

tedder987

I think a power meter would be fun, but based on my use of the heart rate monitor (and even my cadence sensor) it just isn't worth the money. Plus I need to get new wheels first anyways as my old ones are getting noodly.

I wonder if there could be a business case for renting out power meters to non-racers like me who want to take a look but not invest. Swapping a crankset in would be pretty easy in many cases, pedal-based meters would be even easier.

WhyFi

You could probably buy a used one and then sell it off without taking too much of a loss, if any. I don't know if The Pro's Closet has any, but it might be worth a check. Hell, even if you can't find a used one, you can probably find a compatible crank arm for ~$300 new and then sell it for $250 or so.

fooferdoggie 

Even when it gets imported power data it is off. the day I managed 203 watts average I exported that from my bike and strava had the ride from my garmin without a power meter. strava showed the usual 120 or so average power but the imported ride with the power meter data showed about the same. the graph showed the power data but the average was only slightly higher then without a power meter.

burnthesheep

With correct overall weight in Strava, while going uphill without any blustery wind.........it's not too bad. Otherwise, can be useless.

It doesn't know local wind. So outbound it may show you doing 300w at 23mph in a nice tailwind then inbound only 100w at 15mph in a headwind.

*grain of salt*

Psimet2001 

No value.

WhyFi

Pfffft - that's just a wrongheaded oversimplification.

Seattle Forrest

I don't know if they're still in business, there was a place on the internet that rented then for $75 a week. Some local shops do, too.

spilot101

Mine are underestimated by 80~100w