Does my CdA suck?

Triathlon Forum
1

I live in a new area in town and did 3x15min in a flat area (never done that before). It’s not pancake flat but pretty close. There was a light wind but I was riding a loop. I had to slow down a little bit on some turns/stop signs due to cars but I kept my speed up pretty well.

I was on my p2c with training wheels and training tires with a regular helmet.

Here are my average watts, normalized power and average speed for each interval.

Interval 1
AP 272
NP 275
AS 22.6

Interval 2
AP 279
NP 283
AS 23.0

Interval 3
AP 293
NP 295
AS 23.8

I’m 6’6 and 182 lbs. I thought these speeds looked slow based on the amount of power I was putting out. I have raised my front end recently because of neck problems and these results make me want to lower it back down but my neck hurts just thinking about it.

Are these speeds pretty slow for the amount of power I was putting out or are they right on par? I was thinking of doing the same workout with all aero gear next week to see if there is a big difference.

Thanks everyone.

2

depending on what your training tires are, you can lose a lot right there. Otherwise…position, position, position

3

I have a gator skin on the back and a specialized mondo pro on the front. Front wheel is a neuvation aero something and the back wheel is a dt swiss 1.1 32 spoke with a PT

My friend did the same workout on a similar course and he put out less power but his speed was pretty close to mine.

4

You’ve got to remember your a TALL person. To me, those numbers look pretty reasonable, considering your height. I’d retest with multiple positions if you think that there is time on the table still.

5

you’re 6’6", 182lbs? I have a flagpole outside…you and it should be friends.

But seriously, thats a lot of height that can get exposed.

6

You have a powermeter and time…

Why not test it instead of looking at power/speed ratios?

G

7

I have a power meter not sure about the time, i guess it’s worth skipping a workout to test. I was going to do a moderate ride today but I guess I can just test instead.

What way should I test? I know there is a excel file on the wattage forum that has you doing 1 min repeats on a flat course recording speed and time. Is that a pretty good way? Any better way?

Luckily I have major spacers to drop so I can just do a few runs and drop a spacer, then do a few more and drop some more.

8

Har Har :). But seriously everyone loves to sit behind me during road races :).

9

Har Har :). But seriously everyone loves to sit behind me during road races :).

I don’t know about that, at 6-6 and 182, you are not that wide. I don’t benefit from a draft that is above my head. I’d find it more of a nuisance, as i couldn’t see over the top by raising my head. You wouldn’t know about that though. :stuck_out_tongue:

10

Any better way?

Hmmm. Maybe. Depends on the venue you have available to you.

11

I have a mile stretch with pretty good roads that is pretty flat although there is a slight rise. There aren’t many cars on it either but a car may pass me during a test. The weather calls for no wind today which is really weird for Reno.

I was planning on doing the “regression” model that’s on the forum but I see there is a Chung Method too (I’m guessing that’s you). It looks like I just drop my ride file data into the first five columns of the spreadsheet and it produces numbers for me. Is this correct? Can I do this with any ride or does it have to be under certain conditions/style of ride.

Thanks!

12

I have a power meter not sure about the time, i guess it’s worth skipping a workout to test.

testing is training and training is testing - a tempo day is a great day to test.

What way should I test? I know there is a excel file on the wattage forum that has you doing 1 min repeats on a flat course recording speed and time. Is that a pretty good way? Any better way?

Luckily I have major spacers to drop so I can just do a few runs and drop a spacer, then do a few more and drop some more.

As Robert mentioned, the venue kinda defines the methodology (IMHO). Robert’s VE method works great with a great venue. A closed loop with known elevation or a “half-pipe” work well.

Regression also does great as well, but I find it less "workout"y than the VE method.

g

13

I have a mile stretch with pretty good roads that is pretty flat although there is a slight rise. There aren’t many cars on it either but a car may pass me during a test.
If it does, you’ll need to toss that run out.

14

okay cool. I’m pretty sure I read that for the regression method you ride the exact same distance for every test. I’m reading Alex’s spreadsheet and it doesn’t mention it but seems implied from the data.

Does anyone else have any words of wisdom for this test? I’m thinking the only aero gear I will wear is a helmet as my changing position will adjust how the helmet aligns to my back. What speeds would work best? 15, 19, 23, 26? Three passes at each speed?

I think I’m going to be looking mostly down during these tests too since I’ll have to pace my MPH by looking at my PT. Hopefully this doesn’t mess things up too much.

I think I’m going to try three setups. Current setup, one with no spacers, then one with half the spacers put it. If there is enough daylight I’ll go for the in between setups.

I’ll report my findings to this thread.

15

I have a mile stretch with pretty good roads that is pretty flat although there is a slight rise. There aren’t many cars on it either but a car may pass me during a test. The weather calls for no wind today which is really weird for Reno.

I was planning on doing the “regression” model that’s on the forum but I see there is a Chung Method too (I’m guessing that’s you). It looks like I just drop my ride file data into the first five columns of the spreadsheet and it produces numbers for me. Is this correct? Can I do this with any ride or does it have to be under certain conditions/style of ride.

The venue determines the method. If you have flat (or constant grade) then you can do the regression type runs at controlled speed (i.e., field test classic). If you don’t, you’ll need to use a method that accounts for changes in grade. VE allows that. In both cases, wind and cars can mess you up.

In field test classic, you need to get a pretty wide range of speeds for the control section. Same thing with VE: you’ll want to get a wide range of speeds, but you can either get that wide range by applying power or by letting gravity help you. That’s why a half-pipe course works well: you can get pretty high speeds on the descending parts while slowing down at the ends.

16

In field test classic, you need to get a pretty wide range of speeds for the control section. Same thing with VE: you’ll want to get a wide range of speeds, but you can either get that wide range by applying power or by letting gravity help you.

To “pry apart” CdA and Crr in a manner analogous to the regression method, don’t you need speed to vary significantly at any given point on the loop?

Or to (rhetorically) put it another way: what would the VE profile for a loop look like if you rode it at a constant speed?

17

okay cool. I’m pretty sure I read that for the regression method you ride the exact same distance for every test. I’m reading Alex’s spreadsheet and it doesn’t mention it but seems implied from the data.

Does anyone else have any words of wisdom for this test? I’m thinking the only aero gear I will wear is a helmet as my changing position will adjust how the helmet aligns to my back. What speeds would work best? 15, 19, 23, 26? Three passes at each speed?

I think I’m going to be looking mostly down during these tests too since I’ll have to pace my MPH by looking at my PT. Hopefully this doesn’t mess things up too much.

I think I’m going to try three setups. Current setup, one with no spacers, then one with half the spacers put it. If there is enough daylight I’ll go for the in between setups.

I’ll report my findings to this thread.

That sounds pretty ambitious. Like any experimental protocol, you tend to get better with practice so if this is your first attempt I’d cut out the equipment changes and just try to get familiar enough with the procedure so you can do it well and consistently.

There’s no requirement in field test classic to ride the exact same distance each run. The real key is that each run be as consistent as possible within the run. Basically, you don’t want accelerations or changes in grade to be too variable – it’s possible with the regression method to correct for changes in speed and non-flat venues but I’m not sure any of the extant spreadsheets will do this for you automatically. Anyway, that’s why you usually want a flat venue and to hold your speed constant.

Having a short “test” section that you do in both directions helps because when you do your high speed runs you’re not going to want to maintain that speed for a long time.

Holding your speed constant is good because you don’t have to take the accelerations into account, but rather than trying to chase speed on your computer display to ± 0.1 km/h it’s often easier to show cadence and ride to that (and don’t change gear during the test section).

Not needing to ride the exact same distance also makes it easier to “mark” the section. Hit the interval marker way before you enter the test section, and make sure you get up to the proper (constant) speed, hold it through the test section, then hit the interval marker way past the end. When you get to the analysis, just go backward from the second marker to find the parts that were done at constant speed.

For this first time, I’d pick one pretty slow speed (like, 15 km/h) and one pretty high speed (like, 35+ km/h) and just repeat each one a few times. At this stage, don’t worry about “filling in” all the speeds in between – you’re trying to learn.

18

To “pry apart” CdA and Crr in a manner analogous to the regression method, don’t you need speed to vary significantly at any given point on the loop?

Nope.

19

To “pry apart” CdA and Crr in a manner analogous to the regression method, don’t you need speed to vary significantly at any given point on the loop?

Nope.

Not even if the actual elevation is unknown, and/or the quality of the pavement varies at different points?

EDIT: I should add that I’m thinking in terms of the VE profile, not just trying to match up the total change in elevation.

20

and don’t change gear during the test section).

Or between tests, if using an SRM (or Quarq).