Should I use normalised power to calculate training zones?

[sergen]

Today marks my one week anniversary of training with a powertap (I use it in conjunction with WKO+).

I performed the FTP protocol as per 'Training & Racing with a power meter' and came up with 227w. Living as I do in an area of rolling hills the first thing I noticed was just how hard it is to keep within a desired power range. So what I now do is to set the computer to also show average watts as this seems to be more helpful on very rolling terrain.

Yesterday I performed my first 2x20s, aiming for 100% of FTP in both reps. On both of my laps the average power recorded was 216w and 220w. However, when I loaded the data to WKO+ my normalised power for both laps was actually 226w and 227w, bang on my FTP. So my question is, when I'm doing these intervals over rolling terrain in is the average power that counts or the NP? If it's the AP then I fell short of target wattage in the first interval, but if it's NP then I was bang on.

Another point to note - during the FTP test last week I used a small flat section of road nearby that allows me to pedal on the flat for just about 5 minutes. My AP and NP were both 269 watts (VI = 1) so I have used 269w as my Peak 5 min figure.

But on yesterday's ride it started to belt down with rain as I got closer to home so I really put the hammer down over more rolling hills. It was during this escape from the rain that WKO+ says my Peak 5 min power occurred - AP was 260w but NP was 290w.

So what figure should I use to work out my 5 min w/kg? The 269w AP from last week's test or the 290w NP that occurred during yesterday's Peak 5 min?

In short, I suppose the above is all a very long-winded way of asking whether one should use AP or NP when targetting interval durations and also when calculating w/kg?

Thanks

 

[daveryanwyoming]

Today marks my one week anniversary of training with a powertap...

Do some searches on NP, AP and FTP. Most of these questions have been hashed out here before. Here's one thread that touches on some of your questions:
http://www.cyclingforums.com/t-467280-15-1.html

But in a nutshell:

• Users new to power training often have to learn to pace isopower efforts, even on constant grades or flat terrain the numbers jump around a lot and many newer power riders chase their tails by trying to follow the jumpy numbers. Display averaging can help but not necessarily just viewing average power, I'd just stretch the display averaging in the CPU setup menu to maybe 10 to 15 seconds for TTs but still view real time power. You also learn to mentally average or to integrate your displayed power numbers with RPE to adjust pacing. Hard to explain, but easy to do with practice.
• I don't know the terrain you're training in, but is it possible to use all your gears, hit the uphills a bit easier and apply more pedal pressure on the rolling descents to smooth out your AP? If not, is it possible to find steadier terrain for testing and L4 interval work?
• NP is useless for efforts as short as 5 minutes and questionable at 20 minutes. So definitely don't use your 5 minute NP as your 5 minute MMP estimate and I'd strongly advise against using anything like 0.95*20min NP to set FTP.
• Coggan's training levels are defined in terms of AP not NP. If you must use NP then the targets move upwards a bit. I'm pretty sure this information is on the power411 site: http://home.trainingpeaks.com/power411.aspx where IIRC, the training levels chart shows Tempo as an AP of 76-90% of FTP but the description of NP shows Tempo as NP of ~85-95%. That's about right from my own experience, so if you're going to evaluate training to target specific systems using NP, the targets are higher.

I'd keep riding with the PT, focus on delivering steady power during your long training intervals and maybe seek out different training routes if the rollers or traffic interruptions just don't let you hold steady power. Do your L4 intervals at the best steady power you can sustain and set your levels for future workouts on what you can actually accomplish.

Good luck,
-Dave

 

[sergen]

Dave

Thanks for the comprehensive reply.

Out of interest, if 5 min NP is useless and 20 min NP questionable, then over what time period do NP figures become more realistic as the steady-state power one could have held for the duration had one pedalled constantly?

 

[daveryanwyoming]

...Out of interest, if 5 min NP is useless and 20 min NP questionable, then over what time period do NP figures become more realistic as the steady-state power one could have held for the duration had one pedalled constantly?

I think I discussed that pretty extensively in the linked thread. But the short story is that NP tries to estimate effective metabolic load or how hard you're stressing your metabolic processes. Twenty minutes is roughly the lower bound on usefullness for that metric.

I often see NP much higher than my best recorded MMP for short durations but then they start to converge. IOW, 30-60 minute NP is pretty close to my 30-60 minute MMP (AP). Beyond an hour and NP tends to be a lot higher because I simply don't do any best effort AP trials out past an hour. If I raced 50 mile TTs that might be different, but I don't so I have a lot of 70-90-180+ minute NP values much higher than AP for the same durations.

I was wondering if I could see this in my own data so I exported and plotted NP and AP from 5 minutes out to nearly 10 hours. Here's what it looks like. At least for me and the way I ride, train, race they seem to track pretty well from about 15 minutes to about how long it takes to ride a 40K. But NP is still slightly higher than AP.

I wouldn't generalize too much since this reflects the type of riding I do (including terrain) and it bundles together rides from the road and TT bikes. I do longer sustained isopower efforts on the TT bike, but haven't yet matched my typical road bike sustained power numbers (I just don't do a lot of full hour isopower efforts on the road bike). Anyway, I think the chart is interesting for the case of a single rider. I'd be curious to see how other folks compare between AP and NP for various durations.

-Dave

 

[Watoni]

Dave

Thanks for the comprehensive reply.

Out of interest, if 5 min NP is useless and 20 min NP questionable, then over what time period do NP figures become more realistic as the steady-state power one could have held for the duration had one pedalled constantly?

Dave will have a better answer, but I find it useful starting when there have been multiple efforts with recovery in between (such as when doing climbing repeats). For example, I often do an SST workout with one climb of just under 30 minutes, followed by a 15 minute descent, followed by a steepish climb (9% for 2.5 miles) where I ride L4 and then a longer false flat to the summit where I tend to ride tempo, and then descend home. Average power is generally around 200 due to the technical descending, with NP around 260-270. Just AP does not tell the story of that workout, for sure!

 

[daveryanwyoming]

Dave will have a better answer, but I find it useful starting when there have been multiple efforts with recovery in between (such as when doing climbing repeats). For example, I often do an SST workout with one climb of just under 30 minutes, followed by a 15 minute descent, followed by a steepish climb (9% for 2.5 miles) where I ride L4 and then a longer false flat to the summit where I tend to ride tempo, and then descend home. Average power is generally around 200 due to the technical descending, with NP around 260-270. Just AP does not tell the story of that workout, for sure!

FWIW, glancing at NP for an entire ride helps tell you things about the overall ride. But for what you've described I'd either use the interval function on the PT CPU or roll through the file in WKO+ and look at AP and duration for each of the on sections instead of using NP for the entire ride to evaluate how well I met my training goals or what systems I stressed during a group ride or race.

Holding AP in the L4 range but only holding it for five or six minutes before getting some rest is very different than holding AP or NP in the L4 range but broken up by rest periods. If the rests are shorter than 30 seconds you can basically ignore them and look at AP like you would during HOP style microinterval work. But four 5 minute "L4" efforts broken by 3 minute recoveries is very different in terms of training stress than 20 minutes sustained in L4.

I don't think you're suggesting that the on intervals should be short (you talk about 30 minute climbs and 2.5 mile steep grades) but some folks use NP to connect the segments in a way that can be misleading. NP is a useful tool, but Andy's training levels are AP based and assume minimum sustained durations or rests short enough that the body doesn't really recover during them.

-Dave

 

[Fightin Boba]

Today marks my one week anniversary of training with a powertap (I use it in conjunction with WKO+).

I performed the FTP protocol as per 'Training & Racing with a power meter' and came up with 227w. Living as I do in an area of rolling hills the first thing I noticed was just how hard it is to keep within a desired power range. So what I now do is to set the computer to also show average watts as this seems to be more helpful on very rolling terrain.

Yesterday I performed my first 2x20s, aiming for 100% of FTP in both reps. On both of my laps the average power recorded was 216w and 220w. However, when I loaded the data to WKO+ my normalised power for both laps was actually 226w and 227w, bang on my FTP. So my question is, when I'm doing these intervals over rolling terrain in is the average power that counts or the NP? If it's the AP then I fell short of target wattage in the first interval, but if it's NP then I was bang on.

Don't sweat about which counts. Your NP on your 2x20s confirms that your FTP estimation is pretty good. More importantly, your AP on a rolling course for this interval set now serves as a baseline for development on the same course / same interval set. To me power levels are nice when ballparking a first time workout session. After the workout is done, the training levels are largely out the window and the prior workout becomes the key reference point.

Another point to note - during the FTP test last week I used a small flat section of road nearby that allows me to pedal on the flat for just about 5 minutes. My AP and NP were both 269 watts (VI = 1) so I have used 269w as my Peak 5 min figure.

But on yesterday's ride it started to belt down with rain as I got closer to home so I really put the hammer down over more rolling hills. It was during this escape from the rain that WKO+ says my Peak 5 min power occurred - AP was 260w but NP was 290w.

Count 5 min NP as practically meaningless in this situation.

 

[acoggan]

To me power levels are nice when ballparking a first time workout session. After the workout is done, the training levels are largely out the window and the prior workout becomes the key reference point.

It's not just you.

(I made that exact point when I first described the training levels. Unfortunately, it is often "lost in translation", due in part - I think - to the fact that heart rate monitors have made everyone excessively "zone" oriented.)

 

[Watoni]

FWIW, glancing at NP for an entire ride helps tell you things about the overall ride. But for what you've described I'd either use the interval function on the PT CPU or roll through the file in WKO+ and look at AP and duration for each of the on sections instead of using NP for the entire ride to evaluate how well I met my training goals or what systems I stressed during a group ride or race.

-Dave

Dave,

Agreed.

I am looking at did I hold L4 for 28-30 minutes on climb 1, and did I maintain my wattage goal for the 2.5 mile climb (often repeated 2-3 times) as the main workout indicators. I do look at NP, however, as a more effective measure of the overall "cost" of the workout than AP, when evaluating the whole ride.

NP over very long rides is much more useful than AP for me ...