Time Trial by power

Since 2001, over 90,000 cyclist's have joined Cycling Forums to discuss topics from general cycling to equipment, training, racing and travel or vacation destinations (especially in europe during the tour de france). We also feature an great deals in our online store, 100's of articles, classifieds and product reviews.

Hello

I'm looking for some info about time trial by power.
My first theory is that I should keep the power output at a fixed point relative to FTP depending on the length/expected duration of the route.
On second thought, it might not be so simple as the route could be hilly and the wind may change.

Hills - the slower I ride, the less energy (than average) I'm going to waste on breaking the wind. So my conclusion is that I should spend more energy going uphill than when riding on the flats.

Wind - the same goes for wind. I should spend more energy when having headwind and less in the tailwind.
This also means that I should spend less energy when going downhill.

Does this make sense to you - and if it does, how much more or less energy should I at the different parts of the route?

You pretty much have it. If you search this site there are some discussions in the archives. Just make sure you don't go so hard that you pop during the "hard" portions.

... My first theory is that I should keep the power output at a fixed point relative to FTP depending on the length/expected duration of the route.
... wind. So my conclusion is that I should spend more energy going uphill than when riding on the flats.
... I should spend more energy when having headwind and less in the tailwind.
This also means that I should spend less energy when going downhill....Both your theories are good depending on the course and the conditions. But basically you should be trying to hold your best average power for the duration. On a rolling course or one with a lot of wind changes you should do as you suggest, work a bit harder into the winds and uphill but overall you're still targeting your best estimated sustainable power for the duration. That means the hard sections can't be much harder than that average and you shouldn't back off too much in the "easier" sections. You shouldn't for instance, coast on gentle descents and try to make it up by killing the hills you're still bounded by AP(and NP) for the duration so don't stray too far from that average.

Perhaps the most important way to use a PM during a TT is to keep one eye on the meter for the first five minutes or so to make sure you're not going out too hard in the excitement. If you can stay at or slightly below your best AP for the duration during those opening minutes you're in a much better position to just listen to your body and pace accordingly for the rest of the effort. If you go out too hard(really common problem with race day excitement) it'll be really tough to pace later on when it really counts.

Good luck,
-Dave

Hills - the slower I ride, the less energy (than average) I'm going to waste on breaking the wind. So my conclusion is that I should spend more energy going uphill than when riding on the flats.

Wind - the same goes for wind. I should spend more energy when having headwind and less in the tailwind.
This also means that I should spend less energy when going downhill.
Your plan is pretty sound. Do as Beerco recommends and search the archives (try "variable pacing strategy").

One area that I think you need to look at again is:
I should spend more energy when having headwind and less in the tailwind.
When you think about it, I believe that you'll want to do the opposite. The logic is similar for why you would want to lift your effort (wrt FTP) on uphill sections vs downhill sections. You get more bang for your buck when you can limit the % of effort used to overcome wind resistance - if I got that right.

Dave

When you think about it, I believe that you'll want to do the opposite. The logic is similar for why you would want to lift your effort (wrt FTP) on uphill sections vs downhill sections. You get more bang for your buck when you can limit the % of effort used to overcome wind resistance - if I got that right.
Not exactly. It doesn't matter whether the resistance is wind, gravity, or rolling friction, but you'll get the best bang for your energy use whenever there's a lot of 'it' (ie, resistance).

One way to think about it is that the baseline for comparison would be an isopower output. The isopower rider is going to lose a lot more speed when the resistance is high than they gain when the resistance is lower. Now, think about how a variable power approach could be constructed to minimize that effect and you'll see that it's best to go hard during the slow sections of the course and cruise during the fast sections.

Your plan is pretty sound. Do as Beerco recommends and search the archives (try "variable pacing strategy").

One area that I think you need to look at again is:

When you think about it, I believe that you'll want to do the opposite. The logic is similar for why you would want to lift your effort (wrt FTP) on uphill sections vs downhill sections. You get more bang for your buck when you can limit the % of effort used to overcome wind resistance - if I got that right.

Dave
mantra: when you start to slow down, push harder :D

Thanks for your replies, and for the tip on searching the archive. There are some great posts.

The biggest problem, I still haven't found a solution to is how harder or easier should I go, on the different parts of the course. It seems almost impossible to answer that, since there are a lot of variables. The gradient, the wind velocity, length of the various course parts, my physics e.g. I know the avg. power, but if I take I go say 20w easier for 1 minute, how much harder can I climb the hill without blowing on the way home.

Calculating it seems impossible, so I guess the best thing is to have some good rule of thumps.

Thanks for your replies, and for the tip on searching the archive. There are some great posts.

The biggest problem, I still haven't found a solution to is how harder or easier should I go, on the different parts of the course. It seems almost impossible to answer that, since there are a lot of variables. The gradient, the wind velocity, length of the various course parts, my physics e.g. I know the avg. power, but if I take I go say 20w easier for 1 minute, how much harder can I climb the hill without blowing on the way home.

Calculating it seems impossible

Actually, it's pretty easy - the hard part is executing such an optimal (theoretically, anyway) pacing strategy "on the fly". I therefore view such endeavors as simply a way to get in the ballpark, with further refinement coming from experience and/or pre-riding the course in question.

lActually, it's pretty easy - the hard part is executing such an optimal (theoretically, anyway) pacing strategy "on the fly". I therefore view such endeavors as simply a way to get in the ballpark, with further refinement coming from experience and/or pre-riding the course in question.Any rule of thumb where I could start, say go 5% slower on flat a couple of miles before the hill, then 10% harder up the hill - or something like that

It's not easy getting experience in the real world. I have to be equally rested, the wind conditions have to be the same for me to compare two tests. I really don't see how I can get the correct experience.

Anecdotally, I hear more and more (educated and respected riders) advocating a constant power approach. I think Chris Boardman did this. In my own experience I also seem to have achieved better results/watt when aiming to hold a constant power.

Could it be that as soon as you consciously attempt to ride to a variable schedule, you introduce too much variance. i.e. with the best will in the world you tend to vary a fair bit anyway in anything other than a flat straight windless course. Aiming to vary more on top of that then puts you too far either side of threshold, and you end up putting less actual J's into the system (even if you are applying them a bit more efficiently).

Anecdotally, I hear more and more (educated and respected riders) advocating a constant power approach. I think Chris Boardman did this. In my own experience I also seem to have achieved better results/watt when aiming to hold a constant power.

Could it be that as soon as you consciously attempt to ride to a variable schedule, you introduce too much variance. i.e. with the best will in the world you tend to vary a fair bit anyway in anything other than a flat straight windless course. Aiming to vary more on top of that then puts you too far either side of threshold, and you end up putting less actual J's into the system (even if you are applying them a bit more efficiently).Since we naturally push harder up hill and into a headwind, I suspect it's more to do with how much one should hold back their effort on climbs/headwinds than looking to encourage them to push even harder.

i dont have anything like the real numbers or explaination of the physiological principles that demonstrate this right now, but i beleive i read something stating that it takes your body 4 times as long to recover from a unit of power generated above threshold than it does for power generated within.

seemingly this would support the idea that the greatest energy release over a distance is actually achieved by riding at a constant power equal to or very close to your threshold power.

putting it more simply this is equal to the way i have always ridden time trials...bumping HR upto and off threshold but maybe going over by 2-5 beats on a drag or hill.

probably the biggest contribution of power meters in time trialling is in ensuring you dont go way too hard in the first few minutes... i rode my first TT with an SRM a coulple of weeks ago and even with this firmly in mind i was amazed how much excess power i wanted to put out from the start line.

Post deleted
(http://www.ncbi.nlm.nih.gov/pubmed/11083127?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA)

i dont have anything like the real numbers or explaination of the physiological principles that demonstrate this right now, but i beleive i read something stating that it takes your body 4 times as long to recover from a unit of power generated above threshold than it does for power generated within.

seemingly this would support the idea that the greatest energy release over a distance is actually achieved by riding at a constant power equal to or very close to your threshold power.

putting it more simply this is equal to the way i have always ridden time trials...bumping HR upto and off threshold but maybe going over by 2-5 beats on a drag or hill.

probably the biggest contribution of power meters in time trialling is in ensuring you dont go way too hard in the first few minutes... i rode my first TT with an SRM a coulple of weeks ago and even with this firmly in mind i was amazed how much excess power i wanted to put out from the start line.These abstracts might prove an interesting read:

http://www.ncbi.nlm.nih.gov/pubmed/17497402?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA

http://www.ncbi.nlm.nih.gov/pubmed/11083127?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA

http://www.ncbi.nlm.nih.gov/pubmed/17133287?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA

seemingly this would support the idea that the greatest energy release over a distance is actually achieved by riding at a constant power equal to or very close to your threshold power.That may be the case but time trials are decided by the stop watch, not maximal energy expenditure.

ok now i'm a convert... variable power it is!

ok now i'm a convert... variable power it is!

There was a time trial in colorado last weekend. 6 Mile out/ 6 mile back with a screaming tailwind out/ headwind back. It was a crosswind, so I'm not sure of the portion parallel to the course. Out was nearly -1% gradient, back +1%. The time splits were typically in the ratio of about 1.5:1 between back and out.


Intuitively, it would make sense to push it with the tailwind and conserve slightly with the headwind. I ran some simulations with the kreuzotter.de calculator and they all indicated that it was slightly advantageous to push it with the headwind. This is with power -5% out and +3% back (to make average power constant). A greater power variation will have greater results.
However, depending on the wind, the difference isn't huge. About a 2 s advantage for 5mph wind and about 10s for a 15 mph wind. I don't trust the calculations that much as they predict a speed that's much too high on the outward leg (can't spin that fast).

A power variation of +/- 10% is probably too big, unless you've got great recovery powers.

I've read one paper by Swain and was not impressed.

In short, you're probably better off just trying to concentrate on breathing well and dipping your head slightly against the headwind than worrying about power variations. Pushing it on hills is definitely the way to go though.

There was a time trial in colorado last weekend. 6 Mile out/ 6 mile back with a screaming tailwind out/ headwind back. It was a crosswind, so I'm not sure of the portion parallel to the course. Out was nearly -1% gradient, back +1%. The time splits were typically in the ratio of about 1.5:1 between back and out.


Intuitively, it would make sense to push it with the tailwind and conserve slightly with the headwind. I ran some simulations with the kreuzotter.de calculator and they all indicated that it was slightly advantageous to push it with the headwind. This is with power -5% out and +3% back (to make average power constant). A greater power variation will have greater results.
However, depending on the wind, the difference isn't huge. About a 2 s advantage for 5mph wind and about 10s for a 15 mph wind. I don't trust the calculations that much as they predict a speed that's much too high on the outward leg (can't spin that fast).

A power variation of +/- 10% is probably too big, unless you've got great recovery powers.

I've read one paper by Swain and was not impressed.

In short, you're probably better off just trying to concentrate on breathing well and dipping your head slightly against the headwind than worrying about power variations. Pushing it on hills is definitely the way to go though.geez this ain't rocket science guys: when you start to slow down: push harder ... :)

seriously.

Now on our trad. local course here we have a long false flat ~10k @1.6% going out , then a cloverleaf turnround with grades +/- 8-9% and the fast -1.6% leg home (total ~21k).

Ignoring the prevailing winds, it's definitely faster to ride this course with higher power going out vs. coming home. The turnround is best ridden with pretty variable power and normally even standing up for a portion.

Running the course thru the NP pacing algorithm suggests the overall best times (in calm conditions) will be done with IIRC around 20% power delta outbound vs. return. Practically speaking I find the limiter on that course is how much I can average for 15-20min going out around 40-42-43 kph going out vs. the 55-60kph return leg.

just ran some numbers and the model suggests the following outleg/retleg power ratios vs. outleg headwind:

Calm ........ 1.17
+1 m/s ..... 1.22
+3 m/s ..... 1.31
+5 m/s ..... 1.43

-1 m/s ...... 1.13
-3 m/s ...... 1.06
-5 m/s ...... 0.99


anyhow it's just an example where the overall duration may be 26-30 min but a big limiter is 15-20MP. It definitely doesn't teach one good pacing and it's not a route I like for TT training either. but for racing, it is what it is ...

Thank you for those very great links!

These abstracts might prove an interesting read:

http://www.ncbi.nlm.nih.gov/pubmed/17497402?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA

http://www.ncbi.nlm.nih.gov/pubmed/11083127?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA

http://www.ncbi.nlm.nih.gov/pubmed/17133287?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA

I would love some pacing analysis on this in the spirit of this thread.

Here's a file from Stage 1 of last year's San Dimas Stage Race, which is the infamous 3.8 mile uphill GMR TT. It starts with a brief, flatter section then goes into a fairly constant 5.5 to 6.0% grade with plenty of switchbacks.

At the time and by RPE, I thought I rode the course great. But in looking at the file, I made the classic error of going out too hard and gradually weakening as I progressed. My plan was to go steady and punch it hard from the 1k mark to the finish.

All the stats in the file were accurate at the time. My numbers have improved since last year. I'd therefore like to see an improvement in my time commensurate with the work I've done. The TT is in 2 weeks.

327 watts for 17:30 at 75kg netted me 33rd place in the Cat IVs. The winning time was 14:29 for our group. The 'mortal' IVs were coming in around 16:00.

Here is a link to the Power File (http://images.promax.com/gmr_time_trial.wko) and a Gradient Map (http://www.socalvelo.com/maps/GMRgrademap600.jpg) of the course of which we ride the first 3.8 miles.

What would you do differently were you me on this course? If any other info would be helpful, I'll be happy to provide.

Thanks for reading,

wattsup?