Why doesn't power on a rolling course relate to analyticcyling formula?

[Blackie]

This question is quite important to me, as where I live, there is not a single long flat stretch of road. So my LT work is done on rolling/"lumpy" courses.

This evening I did 10miles, and my polar s725 watch (I know, there's about +/-5% accuracy in them), showed an average of 303W. The time taken was 26.57min. Its an out and back course, so theoretically the average gradient is 0%. (In fact its extremely up and down, with total vertical climb of 180m, so I aim to push hard on the hills, and recover a little on the descents, as there seems little point in revving at >120rpm in 53x12). My weight is 76kg, with a bike wt. of 10kg.

I assume a CDA of 0.5 (I don't know if this is correct: I am 1.86m (or 6ft 1in) and did this on a road bike, mostly in the drops, with no aero gear or tribars). There was very little wind, at sea level and the temp was 71F.

The analyticcycling.com calculator tells me my speed should have been 11.96m/s, roughly 27mph, compared to the actual 22.4mph achieved. Time wise, that is 5min difference. What's happened to those 5min? To get to the actual time, without playing with other variables, I need to assume a CDA of 9.9 - does that look right? (Though tall at 6' 1", my BMI is 22.0).

Cheers for any thoughts

 

[RapDaddyo]

This question is quite important to me, as where I live, there is not a single long flat stretch of road. So my LT work is done on rolling/"lumpy" courses.

This evening I did 10miles, and my polar s725 watch (I know, there's about +/-5% accuracy in them), showed an average of 303W. The time taken was 26.57min. Its an out and back course, so theoretically the average gradient is 0%. (In fact its extremely up and down, with total vertical climb of 180m, so I aim to push hard on the hills, and recover a little on the descents, as there seems little point in revving at >120rpm in 53x12). My weight is 76kg, with a bike wt. of 10kg.

I assume a CDA of 0.5 (I don't know if this is correct: I am 1.86m (or 6ft 1in) and did this on a road bike, mostly in the drops, with no aero gear or tribars). There was very little wind, at sea level and the temp was 71F.

The analyticcycling.com calculator tells me my speed should have been 11.96m/s, roughly 27mph, compared to the actual 22.4mph achieved. Time wise, that is 5min difference. What's happened to those 5min? To get to the actual time, without playing with other variables, I need to assume a CDA of 9.9 - does that look right? (Though tall at 6' 1", my BMI is 22.0).

Cheers for any thoughts

You're probably quite fortunate to live in an area with rolling terrain as most riding is done on rolling or hilly terrain unless you live in a really flat part of the world. The problem you are having is that you are looking at the course as though it was flat when in fact it is not flat. If you run the numbers at AC.com as separate segments (e.g., x miles at y% grade), you will probably end up with something that looks like your results. Look at it this way. Take an extreme example of an out/back course: Alpe d'Huez. If you rode Alpe d'Huez as an out/back at 300w (assuming you had a huge gear and didn't fly off the road at 90mph), do you think you would have the same time as a board-flat course of the same aggregate length? You lose more on the upgrades than you gain on the downgrades. BTW, your pacing strategy (push it on the upgrades, float on the downgrades) is sound.

 

[Blackie]

You're probably quite fortunate to live in an area with rolling terrain as most riding is done on rolling or hilly terrain unless you live in a really flat part of the world. The problem you are having is that you are looking at the course as though it was flat when in fact it is not flat. If you run the numbers at AC.com as separate segments (e.g., x miles at y% grade), you will probably end up with something that looks like your results. Look at it this way. Take an extreme example of an out/back course: Alpe d'Huez. If you rode Alpe d'Huez as an out/back at 300w (assuming you had a huge gear and didn't fly off the road at 90mph), do you think you would have the same time as a board-flat course of the same aggregate length? You lose more on the upgrades than you gain on the downgrades. BTW, your pacing strategy (push it on the upgrades, float on the downgrades) is sound.

RapDaddyo, you're spot on the money, and the numbers work. When you put the case of Alpe D'Huez, I realised I had been an idiot. This is great news - I'm not as **** at time trials as I thought.

One further question - in training LT, does it matter whether you do LT rides on flat or rolling courses ie on this one I regularly see maybe +/-100W on steeper parts depending if going up/down, even though on flatter parts I try to pace it close to threshold? (Not that there's much I can do about the terrain, and doing the sessions on a turbo is boring). Just curious as to whether such wide ranges is optimum. Cheers

 

[RapDaddyo]

RapDaddyo, you're spot on the money, and the numbers work. When you put the case of Alpe D'Huez, I realised I had been an idiot. This is great news - I'm not as **** at time trials as I thought.

One further question - in training LT, does it matter whether you do LT rides on flat or rolling courses ie on this one I regularly see maybe +/-100W on steeper parts depending if going up/down, even though on flatter parts I try to pace it close to threshold? (Not that there's much I can do about the terrain, and doing the sessions on a turbo is boring). Just curious as to whether such wide ranges is optimum. Cheers

Personally, I think you're better off training and testing on rolling terrain, because that's the real world (unless you're a trackie). It takes a little more concentration because power is constantly being modulated, but that too is real world. And I think it's natural to go harder on the upgrades because when the bike slows down it doesn't "feel right." When I am trying to maintain a constant power, I always have to watch myself on climbs (where I would normally increase power) and descents (where I would normally back off a little). The good news is that what you're doing is the optimal TT pacing strategy, so that's a good thing. Cheers.

 

[Eldrack]

If you are doing a course where you start at point A and end up back at point A via a circular route hills will slow you down. This is because air resistance is proportional to Velocity cubed. You go slower uphill and therefore encounter less air resitance than going on the flat, you are also doing work to gain gravitational potential energy. On the way down hill your gravitational potential energy goes into kinetic energy (makes you go faster) but the air resitance shoots right up as you go faster, the increase in air resitance being greater than the decrease when you were going up the hill leading to an overall greater power output required on the hill course than the flat one.


Same is pretty much true for wind. It speeds you up going in one direction and slows you down going in the other. The increase in air resitance with the wind behind is greater than the decrease in air resitance with the wind in your face hence you go slower. I hate wind. It is intensely irritating especially as it is very strong where I live.

Based on this I reckon if you increase your power output slightly going up the hills and into the wind and decrease it slightly going down and with the wind behind but keeping the same overall average power you should do better but i'll wait until I have my physics degree before I put a firm answer behind that.

 

[bikeguy]

Blackie, riding with a Cda of 0.5 would require 544 watts at 27 mph. Check
your calculations again. That's not including rolling resistance!

At 22.4 mph, cda of 0.5 would require approx 311 w, not including rolling resistance. Since your power output was 303 watts, I estimate your cda
is 0.43, although it's probably a bit lower because you were on rolling terrain.

Oh yeah, the speed you rode at is quite good.



-Bikeguy

 

[Blackie]

Thanks for those replies.

Bikeguy, thanks for trying to figure out the numbers. I played with them, and based on analyticcycling assumptions including rolling resistance, my CDa was 0.7, which fits in their range of 0.4-0.7.

Its interesting that reducing this number makes a huge difference to speed. Any ideas to get it down when road racing? I've sometimes seen vets wear skinsuits in crits - does it really make a big difference? Finally, is there any data on the benefits from aero gear in TTs re time saved/drag. Cheers