Converting Anaerobic Power to Aerobic Endurance



Cycling Pigeon

New Member
Apr 30, 2013
1
0
0
Hi All,

I've recently started training with power, and its confirmed a reality for me. My power curve is heavily weight toward Anaerobic efforts (CP1 470, CP5 365, CP20 280, CP60 275).

I'm a long course triathlete and really need to be maximizing my training for increased endurance and am really looking to convert some of the anaerobic strength to FTP.

What I'm wondering is if anyone has suggestions for particular sessions that would be maximal for making that conversion. Obviously 2x20 sweet spot is a no brainer for raising your FTP, but are there other specific sessions people would recommend?

Best,

Pigeon
 
I think you want to look at longer time periods.

Triathlons are what 5-6 hours of bicycling typically at 80%FTP.

I think it is a stretch to extrapolate your 1 hour power out to 5-6 hours. Or to believe that 80% is the correct effort for you.

I think you should work on measuring and improving your 2-3 hour power.
 
Have you looked at these on a "power profile" chart such as this? When I assume a weight for you, it shows a sharply inverted V, and a very low 1 minute power in comparison to your others, (Assuming 75kg your FTP puts you in the 'moderate'/'good' area, your 5 min the 'good' and your 1 minute in the 'untrained') This leads me to believe that you perhaps don't have a full image of your power profile yet (which is quite possible as you said you are new to power training.

That said, you don't need to pigeon hole yourself as a specific "type" of rider. Too many people decide early on that they are a specific type of rider and give up on other types of riding. While it is important to recognize your strengths and weaknesses, you do not want to force yourself into being a certain type of rider. If you really want to focus on long course triathlon, then you need to focus on riding your bike a long ways. The best way to get better at something is to do it. For your training that means more longer tempo rides than most roadies do, more steady state work. That isn't to say there isn't value in doing the group rides and high intensity type work, but it shouldn't be your focus. You can and should do things like 2x20's, 3x20's, 2x30's, 1x40's to improve FTP, but even if you have an FTP as high as spartucus, it won't help you much if you can't hold tempo at hour 4 of a long course bike leg.
 
Originally Posted by An old Guy .

I think it is a stretch to extrapolate your 1 hour power out to 5-6 hours.
You think wrong - FTP can be used to predict someone's power to w/in (on average) +/- 5% from 2 minutes onward:

 
  • Like
Reactions: dhk2
Originally Posted by joroshiba .

Have you looked at these on a "power profile" chart such as this? When I assume a weight for you, it shows a sharply inverted V, and a very low 1 minute power in comparison to your others
+1. 171% of FTP for 1 min is actually below average, not above.
 
Originally Posted by acoggan .

You think wrong - FTP can be used to predict someone's power to w/in (on average) +/- 5% from 2 minutes onward:
It is poor form to make such a claim. The claim is untestable. The claim does not even make sense.

---

It seems that on average the predicted error is zero. But since no one is average the deviation from zero is not bounded by 5%.
 
Interesting way to visualize power variation. Is this chart and the error bars based on Monod modeling, athlete sampling or some combo of the two?

-Dave

Originally Posted by acoggan .




You think wrong - FTP can be used to predict someone's power to w/in (on average) +/- 5% from 2 minutes onward:

 
Originally Posted by An old Guy .

It is poor form to make such a claim. The claim is untestable. The claim does not even make sense.
I don't think it's poor form, the claim makes sense to me, and I can think of ways that it could be testable.

Originally Posted by daveryanwyoming .

Interesting way to visualize power variation. Is this chart and the error bars based on Monod modeling, athlete sampling or some combo of the two?
I don't know how Andy did this but I can think of ways to construct this empirically. All it would take is a huge data base of power data across many athletes across many durations. For example, if you had access to, say, the entirety of the data uploaded to TrainingPeaks I think it would be nearly trivial to produce a chart like that. [Edit:] As I said, I don't know how Andy did this. I'm just saying that *if* one had access to data like that it would be possible to construct a chart like that.
 
Originally Posted by RChung .

...I don't know how Andy did this but I can think of ways to construct this empirically. All it would take is a huge data base of power data across many athletes across many durations. For example, if you had access to, say, the entirety of the data uploaded to TrainingPeaks I think it would be nearly trivial to produce a chart like that. [Edit:] As I said, I don't know how Andy did this. I'm just saying that *if* one had access to data like that it would be possible to construct a chart like that.
Yeah, I know how the data could be reduced from a large enough sampling of athletes and durations but I assumed he took a different approach especially as smoothly as the means trend and as balanced and smoothly trending the error bars appear to be.

-Dave
 
Originally Posted by RChung .


I don't think it's poor form, the claim makes sense to me, and I can think of ways that it could be testable.
As I said on average everyone is within 0% of the average. So his claim makes no sense.

As for testing: Good luck. The goal of testing to to disprove the claim. Any test that disproves the claim leads to aacogan and others saying the FTP is wrong.
 
Well it's idiocy to say the following, and proves that you really don't have a clue about long course triathlon.
Originally Posted by An old Guy .
Triathlons are what 5-6 hours of bicycling typically at 80%FTP.
But that doesn't stop you from posting it. Extrapolating from that, most of what you post is equally nonsense about road racing too.
 
Originally Posted by joroshiba .

He isn't THAT far off for an Ironman bike split. 80% of FTP is a bit high, 70-75% is more realistic.


-- http://blog.trainingpeaks.com/posts/2012/10/30/how-to-cheat-by-using-a-power-meter-in-an-ironman-by-joe-fri.html
The difference between 70-75% and 80% FTP is the difference between running vs walking the marathon though... so actually a pretty huge difference. If you've ever ridden 5 hours at 70% FTP and 5 hours at 80% you'll know what I mean.

Obviously the faster you go, the harder you can go because the time spent riding is less, but very few top end AGers will do 80% FTP on the bike and still manage a decent marathon. Ironman is about the run anyway, so a gain of 10-15 min on the bike from going too hard is lost with the first 2 miles you walk rather than run.
 
Originally Posted by daveryanwyoming .

Interesting way to visualize power variation. Is this chart and the error bars based on Monod modeling, athlete sampling or some combo of the two?

-Dave
Obviously can't be Monod, as the errors bars would go to infinity at the shorter and longer durations. But, yeah, something like what you say above.
 
Originally Posted by smaryka .

Well it's idiocy to say the following, and proves that you really don't have a clue about long course triathlon.
But that doesn't stop you from posting it. Extrapolating from that, most of what you post is equally nonsense about road racing too.
At least I read the original post.

The poster seems to have long course experience and seems to want to improve his power for that event.

I simply suggested that he monitor his 2 or 3 hour power to see how his power is improving. I also suggested that he use 80%FTP as a goal for his triathlon ambitions.

I make no claim that he can improve his long term power or that he can reach the goal. I guess you have problems with reasonable advice.
.
 
Originally Posted by acoggan .


Obviously can't be Monod, as the errors bars would go to infinity at the shorter and longer durations. But, yeah, something like what you say above.
Yeah, Monod doesn't describe the real short and long stuff very well.

Really nice way to illustrate things.
 
I looked at the graph aacogan provided and at the table joroshiba provided. Together they seem to be less than helpful.

Both cover 5-6 hour bicycling events. aacogan's graph indicates 67% FTP for that time period. Since joroshiba's bicycling event follows a swim and precedes a run, I would expect his table to provide lower percentages. But the percentages for the "good range for age group athletes" goes as high as 76%.

My math is not as good as it used to be but 76%-67% is a 9% difference. (1 - 76%/67% is a 13% difference. I don't want to argue which was aacogan's intent.) and 9% error is a lot more than a 5% error range. But more important Friel's table has the bicycling followed by a run.

It appears one of the two is in error by a lot. Perhaps both are in error.
 
Originally Posted by An old Guy .


I looked at the graph aacogan provided and at the table joroshiba provided. Together they seem to be less than helpful.

Both cover 5-6 hour bicycling events. aacogan's graph indicates 67% FTP for that time period. Since joroshiba's bicycling event follows a swim and precedes a run, I would expect his table to provide lower percentages. But the percentages for the "good range for age group athletes" goes as high as 76%.

My math is not as good as it used to be but 76%-67% is a 9% difference. (1 - 76%/67% is a 13% difference. I don't want to argue which was aacogan's intent.) and 9% error is a lot more than a 5% error range. But more important Friel's table has the bicycling followed by a run.

It appears one of the two is in error by a lot. Perhaps both are in error.
1. Not my table.

2. Pacing strategies for triathlons and/or longer TTs are typically based on normalized power, and for good reason.
 
Originally Posted by Cycling Pigeon .

Hi All,

I've recently started training with power, and its confirmed a reality for me. My power curve is heavily weight toward Anaerobic efforts (CP1 470, CP5 365, CP20 280, CP60 275).
How long have you been riding, and how much do you weigh? The first variable--your experience level--would be good to know for prescribing any workouts, and it'd be better to have an idea of how much you ride presently and how much you've been riding lately. You may just need to do more volume overall, and it's conceivable that the additional miles you ride need to be performed at a place slower than what you'd anticipate. Again, this depends on how much work you've done and are doing.
 
Originally Posted by acoggan .


1. Not my table.

2. Pacing strategies for triathlons and/or longer TTs are typically based on normalized power, and for good reason.
1) I made it clear that the table was not yours. But since you bring it up, perhaps you could explain your graph. In particular, why there are "error bars." There might be a variation among individuals, but there should not be errors of that magnitude.

2) "[normalized power] is an estimate of the power that you could have maintained for the same physiological "cost" if your power output had been perfectly constant." (This is from a paper by Andy Coggan. Perhaps you have heard of him.)

For triathlons NP and AP are very close. So any claim that one should use one or the other seems foolish.

---

It confuses me why you would put up a graph based on average power and then claim normalized power is more appropriate for the matter at hand.

It also confuses me why you would dismiss the differences between the AP graph and the NP table. If I was interested in using power to guide my training, I would be all over the differences.