BestBikeSplit of by 50 watt

Jun 10, 19 7:28

Post #1 of 20 (3052 views)

For any gives speed, I seem to need way more power than any online tool estimates, including bestbikesplit (bbs) which I currently use.

E.g. for my yesterdays 72k/800m/3 hours tour I had an avg. of 190 and a NP of 220 watts.
If I plot that course into bbs, it estimates that I should be able to do it with an average of 140w. Other tools are similarly optimistic.

I've spent a lot of time searching the issue, so far without success:

- I got a professional bikefit tuned to a rather aero position.
- I wear tight tri clothing
- I use race tires at 8 bar
- I use a mid/high end bike with ultegra components.
- I got my power meter compared to a second device, values are quite similar.
- I freshly lubricated the chain

I'm out of ideas. How can it be that my power requirements for a given speed are 50 watt above calculation?

My average CdA is calculated to be around .45 on a hilly course, which makes sense for a 1.90m guy at 88kg. Yet, to achieve realistic predictions I have to set the calculator to a CdA of around .75, which is unheard of on any race bike.

For my first IM next month, the difference is between calculated 6 hours and real-life 9 hours missed cutoff.

I am happy to provide any files or information that might help.

Thanks!

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 7:43

Post #2 of 20 (3003 views)

Is you power meter accurate?

Re: BestBikeSplit of by 50 watt [sgy] [ In reply to ]

Jun 10, 19 7:57

Post #3 of 20 (2970 views)

sgy wrote:

Is you power meter accurate?

I got my power meter compared to a second device, values are quite similar.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 7:58

Post #4 of 20 (2963 views)

What type of power meter are you using? What do you have your CdA set as in BBS?

Re: BestBikeSplit of by 50 watt [stevej] [ In reply to ]

Jun 10, 19 8:10

Post #5 of 20 (2919 views)

stevej wrote:

What type of power meter are you using? What do you have your CdA set as in BBS?

Garmin Vector 3 (Pedal based).

I experimented with different CdA settings:
- BBS race analytics yields an avg. of .40 to .44
- Bike measurment data entered into bbs yields around .43
- Online calculators yield, based on weight etc. .40

All quite close. In detail, race position is slightly below .4 and climbing position slightly above .5

CdA required to replicate my power usage for a give speed is .75.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 8:37

Post #6 of 20 (2854 views)

Zsolt85 wrote:

stevej wrote:

What type of power meter are you using? What do you have your CdA set as in BBS?

Garmin Vector 3 (Pedal based)....

The crucial detail is whether this is single or double pedal measurement. And when you compared to a 2nd device, what was that?

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 8:43

Post #7 of 20 (2832 views)

Have you only compared a single ride on BBS or have you done several?

There are many variables to account for including those you've already mentioned (weight, position, clothing, bike, etc...), so assuming you've entered all of those accurately, the big ones I'd look at next are weather and course accuracy. Have you accounted for the actu a l weather conditions and are you sure the course data is accurate and complete. It's not a l ways obvious if it's not. Best thing to do is upload the xroures from a bunch of rides and model them all against actual time and power observed to validate whether this is a cosistent mismatch or a one off.

Re: BestBikeSplit of by 50 watt [Ai_1] [ In reply to ]

Jun 10, 19 8:46

Post #8 of 20 (2821 views)

Ai_1 wrote:

Zsolt85 wrote:

stevej wrote:

What type of power meter are you using? What do you have your CdA set as in BBS?

Garmin Vector 3 (Pedal based)....

The crucial detail is whether this is single or double pedal measurement. And when you compared to a 2nd device, what was that?

It's single pedal. I was thinking about getting the right pedal, but those "upgrade" pedals seem to be out of stock everywhere. I compared it to a smart trainer (wahoo kickr or so), which showed 2-5 watts less.

Re: BestBikeSplit of by 50 watt [Ai_1] [ In reply to ]

Jun 10, 19 8:56

Post #9 of 20 (2802 views)

Ai_1 wrote:

Have you only compared a single ride on BBS or have you done several?

There are many variables to account for including those you've already mentioned (weight, position, clothing, bike, etc...), so assuming you've entered all of those accurately, the big ones I'd look at next are weather and course accuracy. Have you accounted for the actu a l weather conditions and are you sure the course data is accurate and complete. It's not a l ways obvious if it's not. Best thing to do is upload the xroures from a bunch of rides and model them all against actual time and power observed to validate whether this is a cosistent mismatch or a one off.

I measured a total of about 6 or 7 rides, totaling about 20 hours of ridetime. (Most of my training was on Zwift, so that's all I got right now.)

The error is even simpler to observe:
A simple calculator like this one tells me that applying 200 watts should give me 30 kph. I rather get 24.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 9:18

Post #10 of 20 (2759 views)

Zsolt85 wrote:

Ai_1 wrote:

Have you only compared a single ride on BBS or have you done several?

There are many variables to account for including those you've already mentioned (weight, position, clothing, bike, etc...), so assuming you've entered all of those accurately, the big ones I'd look at next are weather and course accuracy. Have you accounted for the actu a l weather conditions and are you sure the course data is accurate and complete. It's not a l ways obvious if it's not. Best thing to do is upload the xroures from a bunch of rides and model them all against actual time and power observed to validate whether this is a cosistent mismatch or a one off.

I measured a total of about 6 or 7 rides, totaling about 20 hours of ridetime. (Most of my training was on Zwift, so that's all I got right now.)

The error is even simpler to observe:
A simple calculator like this one tells me that applying 200 watts should give me 30 kph. I rather get 24.

The single sided meter is a prime suspect for inaccuracy. It helps that you compared it against a total power reference but as it was a smart trainer it's hard to know if that comparison would stay true on an outdoor ride. Outdoors your pedal balance may differ, especially if it involves different durations where fatigue may be a factor on pedal balance as could be climbing and varying efforts and cadences.
Were your outdoor rides generally quite flat, mountainous or a mix? Was it windy on most of those rides?

Your description of yourself and your gear compares fairly closely with me.
On my tri bike I can average 35km/h at touch over 200W on a flat course or 40km/h at about 280W. I think BBS thinks I should be slightly faster but it's fairly close (I realise I'm not terribly aerodynamic at present). On a hilly course I can easily be down to 27km/h at 200W, I don't have a experience of a mountainous 280W ride to reference, plus they will have much more varied efforts.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 9:20

Post #11 of 20 (2754 views)

This might be your answer. When you compared, what type of powermeter did you compare it to? Sounds like a power imbalance making the assumption of "Total Power = Left Leg Power x 2" invalid.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 9:25

Post #12 of 20 (2745 views)

8 bar, 115 psi? That's pretty high on real (ie rubbish) roads. Try playing around with the crr on bbs.

Re: BestBikeSplit of by 50 watt [SteveM] [ In reply to ]

Jun 10, 19 10:01

Post #13 of 20 (2711 views)

SteveM wrote:

8 bar, 115 psi? That's pretty high on real (ie rubbish) roads. Try playing around with the crr on bbs.

Good point. Will probably go back to 7 bars.
I recently went up from 7 bar to 8, hoping to fix the issue. Actually BBS accounts about 1 - 2 % of total resistance to crr (with rest being airo drag and gravity), so even the worst conditions would not explain more than 3-5%.

Re: BestBikeSplit of by 50 watt [FishOutofWater] [ In reply to ]

Jun 10, 19 10:03

Post #14 of 20 (2706 views)

FishOutofWater wrote:

This might be your answer. When you compared, what type of powermeter did you compare it to? Sounds like a power imbalance making the assumption of "Total Power = Left Leg Power x 2" invalid.

Very good point. I just assumed that my right leg is stronger, so if anything, garmin should underestimate my power output, not overestimate it. Though I see that difference even on fresh legs and easy rides.

Re: BestBikeSplit of by 50 watt [Ai_1] [ In reply to ]

Jun 10, 19 10:07

Post #15 of 20 (2695 views)

Ai_1 wrote:

Zsolt85 wrote:

Ai_1 wrote:

Have you only compared a single ride on BBS or have you done several?

There are many variables to account for including those you've already mentioned (weight, position, clothing, bike, etc...), so assuming you've entered all of those accurately, the big ones I'd look at next are weather and course accuracy. Have you accounted for the actu a l weather conditions and are you sure the course data is accurate and complete. It's not a l ways obvious if it's not. Best thing to do is upload the xroures from a bunch of rides and model them all against actual time and power observed to validate whether this is a cosistent mismatch or a one off.

I measured a total of about 6 or 7 rides, totaling about 20 hours of ridetime. (Most of my training was on Zwift, so that's all I got right now.)

The error is even simpler to observe:
A simple calculator like this one tells me that applying 200 watts should give me 30 kph. I rather get 24.

...
Were your outdoor rides generally quite flat, mountainous or a mix? Was it windy on most of those rides?
...

Thanks for your numbers! It helps me verifying that the estimations of bbs are indeed realistic and the problem is on my end.
My rides are usually slightly mountainous, with the wind being low and all over the place. I always do round trips, with elevation and partially wind cancelling out.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 10:37

Post #16 of 20 (2642 views)

Elevation and wind do not cancel out because it's a loop. Do the maths to convince yourself!

As a really simple example:
Let's say you're doing a 10km out and back route that's simply 5km up a 6% climb and 5km back down again. Let's say you typically average 30km/h on my roadbike at 200W. But up this gradient you only manage 15km/h at 200W. So the climb takes 20mins.
On the descent you might do 60km/hr and get down in 5mins but that's still 25mins compared to 20mins on the flat.
In fact you'd already used your 20mins by the time you finished climbing. I think these are pretty realistic numbers but regardless the principle should be clear?

The problem is it's the time spent at a given speed that matters, not the distance. So a slow 5km does more damage than a fast one can compensate. There are some exceptions but they are not the norm.

Re: BestBikeSplit of by 50 watt [Ai_1] [ In reply to ]

Jun 10, 19 10:48

Post #17 of 20 (2614 views)

Sorry, I misphrased that.
What I've ment is that unknown variances to wind direction, wind speed and slightly off elevation profile will have less effect on a round trip (or on the cumulative results over multiple roundtrips) then on one-way trips:

On round trips my elevation profile tend to end up +/- 10 meters to where it started, so there is no major systematic error, i.e. I don't severly underestimate elevation gains.
Winds in my area tend to be quite constant over the day, so a round trip will reduce the effect on unknown exact wind speeds greatly, compared to a one-way where you could have all head- or tailwind.

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 11:48

Post #18 of 20 (2532 views)

I own 4 kickrs and they are all different in watts. gen one is the real ego booster reads 25 to 30 high haha.. best thing i ever did was throw out my garmin pedals and got a quarq.

Re: BestBikeSplit of by 50 watt [Cookiebuilder] [ In reply to ]

Jun 10, 19 12:40

Post #19 of 20 (2481 views)

Cookiebuilder wrote:

I own 4 kickrs and they are all different in watts. gen one is the real ego booster reads 25 to 30 high haha.. best thing i ever did was throw out my garmin pedals and got a quarq.

You own 3 kickrs more than anyone should need. :)

Re: BestBikeSplit of by 50 watt [Zsolt85] [ In reply to ]

Jun 10, 19 22:57

Post #20 of 20 (2326 views)

Zsolt85 wrote:

Cookiebuilder wrote:

I own 4 kickrs and they are all different in watts. gen one is the real ego booster reads 25 to 30 high haha.. best thing i ever did was throw out my garmin pedals and got a quarq.

You own 3 kickrs more than anyone should need. :)

That is a valid point ;-) But it is true, not all Kickrs are very reliable. The 2018 version seems very accurate though.
One problem with Kickr trainer is that the pedal velocity is not very constant during a pedal cycle resulting in wrong values for different powermeters that base power calculation on constant pedal velocity.

The best thing to do - regardless of the accuracy of your powermeter is upload some fro; training rides in the BBS analyzer tool and let it calculate corresponding cda and crr for you. This will give you something to work with even if your power meter is off for some reason as long as it is always off by the same amount.

Good luck!
Sam

Our Partners

transition-DOT-area-AT-slowtwitch-DOT-com

Advertise with us

©1999-2024 Slowtwitch, Inc., and Slowtwitch.com
Reproduction of material from any Slowtwitch.com page
without written permission is strictly prohibited.