Sunday, April 05, 2009

New Skool

OK, a follow up to this earlier post ("Old Skool") about a local time trial (TT) I raced last month on my normal road bike.

On that ride I completed the not quite 25km undulating Calga TT course in 42:55.

I also referred to the importance of the 3 P's of riding fast TTs (well as fast as one can go):

1. Power to the pedals
2. Piercing the wind
3. Pacing the course

With power meter data, the environmental conditions known and some special mathematical wizardry, I am able to quantify each of those three Ps from my ride last month:

1. Power:
My power average was 264 watts (normalised power 268 watts).

2. Piercing the wind:
I estimated a Coefficient of Drag x Frontal Area (CdA) of 0.334 m^2. The lower the CdA, the faster you go for the same power output.

3. Pacing
My Pacing Optimisation Score was 0.990, which ranks between "excellent" and "best in class" and means that in order to attain "best in class" level of pacing, there was another ~ 13 seconds of time savings to be found on course. Those savings can be found by dosing your effort carefully on the course depending on the terrain.

Of these 3 P's, the biggest gains (in a month) were going to come from improving #2: Piercing the Wind.

My pacing is already pretty good (but always room for improvement).

As for power, well that was a bit of an unknown for a couple of reasons, one of which I'll get to in a moment. The other reason was I recently had a bad head cold and needed a full week off training. That's never a good thing when you are training well to improve your power. If you have a good amount of training behind you, it often doesn't hurt your power much, provided you allow yourself to recover properly and don't start riding hard too soon and end up prolonging the illness.

So what about piercing the wind?

After my Old Skool post, a generous offer was made by a former coaching client of mine to loan me a TT specific bike (for a while until I can sort out my own rig). That was an offer too good to refuse, so last week the bike arrived and yesterday was my first and only chance to work on getting the set up right. It even has a Powertap power meter so that was a big bonus :D.

So it was off to Centennial Park for some time riding and making adjustments to the saddle position, the bars, arm rests and so on until I felt I could ride the bike OK. Main challenge was being able to pedal without the prosthetic hitting my arm on the upstroke. It's really annoying. I got it to a stage where it was hitting slightly but not enough to ruin a ride. I will have a solution for that, which I'll write about in another post (some news coming about my new legs).

Sometimes when you go from a road bike position to a TT bike position you can lose some power as you are not used to the different joint angles and so on. Typically you are looking to maximise your aerodynamic gains without much sacrifice in ability to produce power (in the end it's maximising speed that matters). That can take quite some time to optimise as you need time to adapt to the new bike position. I didn't have that luxury as the TT was today.

Here's the loaner bike:

Bike has an aero bar set up and 38mm deep carbon rims, so not a complete aero set up (which would have a rear disk wheel and a deep section front wheel). Also, I am not as yet using an aero helmet - I used the same standard road helmet as last time as well as a skin suit.

So what happened this time?

Conditions today were very similar to last time: calm to very little wind with the same air density at 1.179 kg/m^3 (different temperature, barometric air pressure and humidity between each day but all the variations cancelled each other out to end up with air that was the same density). In other words, the two TTs can be readily compared.

My race time was 41:14, which is 1 minute 41 seconds faster than last month.

So how did the 3 P's compare to last time? Here are the numbers (with previous month's TT numbers in brackets). They allow us to assess how much each component of the Three Ps contributed to my extra speed.

1. Power:
Average: 263 watts (264 watts) - basically the same power
Normalised: 269 watts (268 watts)

2. Piercing the wind:
CdA: 0.286 m^2 (0.334 m^2) - a 14% improvement

3. Pacing the course:
Pacing Score 0.991 (0.990)
Time lost compared to Best in Class pacing: 8 seconds (13 seconds) - so a 5 second improvement through better pacing

So it's pretty clear that the vast bulk of speed improvement was due to my improved aerodynamics, all achieved simply because I was using a bike that enabled me to ride in a much more aerodynamic position. Now if you ever wondered why some riders obsess over aerodynamics - well there's your answer!

Just to put the aerodynamic changes into perspective,
that's over 4 seconds per kilometre faster for the same power.

The nice thing about this is that there are more aerodynamic improvements to be made, and one would hope that my fitness will improve and that I'll have more power available once I adapt to the TT position. As for pacing, well I seem to have that pretty well sorted.

One final comment on the day. Last time I experienced some problems with my leg fitting becoming loose and painful in the latter stages of the TT. I didn't experience the same problem today. I packed extra foam into my leg this time and conditions were a little cooler which more than likely meant less sweat build up inside the leg liner. It still works loose gradually over time but it was much better today and no significant pain.

My next TT will be at the end of April, when I tackle the challenging Mooball TT course in northern NSW. That is part of the 2009 National Paracycling Road Race Championships. Should be a hoot (although I wish it were a flatter course).


Unknown said...

Great post Alex - best of luck for the Mooball TT.

Just curious - do your models provide any rough ideas in time reductions as average power goes up on this course, assuming everything else is equal?

For example, say your average power in a few months increased by x watts. How would that translate to time at Calga?

Now I know this is asking for real rough and rubbery numbers, but do you have a feeling how much time would be saved if you had a rear disk wheel and/or aero helmet?

Thanks for the great post!

Alex Simmons said...

Oh I can readily put in another power number and get an indication of the time reduction. For instance, if I was capable of say, 5% more power, my time would be about 58 seconds less on that course.

As for the value of a disk wheel or an aero helmet, well until I test the difference in CdA, that's a bit hard to be quite so specific.

I will have an aero helmet on loan for my next TT at Mooball. I doubt I'll have a disk though.

But for instance let's say that combined they provide for a 5% improvement in total CdA. That would equate to a reduction of 32 seconds (at same power). But I don't know the actual drag reduction that such equipment might make in my case.

If I could produce 5% more power and achieve a 5% reduction in CdA, then my Calga time would be reduced by 90 seconds, just breaking into the high 39-min mark.

Unknown said...

Thanks Alex.

Was there any weight difference between the TT bike and the road bike you used last time? Does that play a part in the scheme of things?

Also - I'm really interested in the pacing aspect. Do you have any references on this? Intuitively, I've always felt going a little harder up the hills at Calga then slightly easing off on the descents seems to work well, but I am only guessing.

Then of course, there is the bastard of all time right at the end, Blood Hill. :)

I got my best time at Calga when I really hit Blood Hill pretty hard, but boy was I in agony after that!

Dan Bill said...

Hi Alex, this is my first comment on your blog but i have kept on eye on it for a while and i think some of your write ups are absolutly fantastic. I do have one question though, how do you insert screenshots of your training peaks files as this is something i am trying to do but cant seem to be able to.


Alex Simmons said...

Mass of equipment has a much smaller impact in TT performance, unless of course the TT has a lot of climbing.

In this instance, the two bikes weighed the same anyway (I hopped on the scales while holding each and there was no difference).

What matters most is the power to CdA ratio.

But for the sake of argument - what would the performance advantage be on the 25km Calga course if the TT bike was 1kg lighter (all else the same)?

A: 6 seconds.

Alex Simmons said...


I just use a screen capture tool* to save the pics to my hard drive then use Blogger's picture upload tool.

*I use Faststone Capture but it also works if you use the PrintScreen button and paste in MS Paint, and crop the picture to suit.

Unknown said...

Wow - just 6 seconds! Converting my older Aluminium road bike with proper TT handlebars/shifters sounds a lot more attractive now.

Any comments on how you derived the ideal pacing strategy? I am assuming there are some interesting mathematics behind it?

Alex Simmons said...

See this thread on the Google Groups Wattage forum:

That references a link to a paper I wrote on the subject.

And yes there is a lot of maths underpinning the models!

robert said...


Found the POI pacing model document, very applicable to a TT route I've been working on. Wondering if the spreadsheet model is available to help break this route into segments. Great blog.