dsmyth12

What is actually a better measure of your power, normalised or average? For example, my best 20 minute power done during a ride out on the road is 303 watts but the normalised power for that segment was 335 watts. The terrain could probably be described as rolling hills.

rubiksoval

Average, because it tells you what you actually did. Normalized power is a mathematical formula that can be useful, but can also be gamed. For longer intervals like 20 minutes, it's even possible to have higher average power than normalized power because normalized power is calculated with a 30 second rolling average, thus lags by 30 secs.

It's also possible to have significantly higher normalized power simply by going at a max effort for 30 or more seconds. I can do 30 minutes of riding and have an average power of 220 watts but a normalized power of 350 or so with only 2-3 minutes of max efforts (I've even done 1 hour rides with normalized power 40+ watts over threshold by doing ~5 mins worth of max 30 second efforts). Playing around with workouts like that will show you the wide discrepancies in normalized power.

As you train more, you'll be able to keep your power more steady, especially on the downhills. I'd focus on that more than normalized. And I definitely wouldn't ever use normalized as a number in training.

topflightpro

Rubik is correct. In fact, in many of my sprint training rides, it's not uncommon for me to post a normalized power that is 3-4x my average.

Troul 

True current health (age, weight, height, gender) as well as real time heartrate with the average of speed, distance, & measurable time. That data & those averages are going to provide your output of power (cal) . Should be able to compare that to the power (watt) generated.

rubiksoval

100% no.

burnthesheep

I wish some of the platforms would make NP less prominent of a metric. Like, you can see it if you want to dig. Folks see it and get all giddy about how high it is. All it is to me is an item that goes into the TSS formula.

Folks might see a huge NP and think they had a stellar workout, but not if the TSS for the workout doesn't contribute towards your goals.

You can go do a 20min workout with mega NP that doesn't really contribute to your fitness at all.

So, take NP with a grain of salt.

chaadster

Normalized Power is not a measure of power, it is an expression of power variability in a workout, used for the assessment of training load and intensity.

It is the exact purpose of NP to reveal how a workout was done, i.e whether it was steady-state efforts or included peaky, high power, by comparison to average power. You could go in and look at a ride file or use graphs to see the same thing and get an understanding of the physiological cost of a ride, but having numbers like AP and NP make that really easy to do.

burnthesheep

To an extent, doing a solid warmup and cooldown can toss the NP and that revelation. In an effort to burn some more KJ's, I've been adding some lower Z2 work at the end of my harder workouts. I'm already suited up and on the trainer, another 15-20min won't kill me. And is often a good cooldown.

Doing that though will give a lot lower NP than if I did a 2min warmup, slammed out intervals, then hit "stop" the second I finish the last interval. And honestly, I see riders do this for outdoor rides a LOT because the data "looks" cooler to have a faster average speed and higher NP than logging that warmup and cooldown.

This is why I say NP is more of an input to the things that really matter.

The ONE thing even Zwift gets right out of all the things it gets wrong is giving you the little "stars" for each rep you hit on target. Compliance. Compliance against the time in each zone the workout was designed to make you do. If you don't get many stars, you set your targets too high or something.

It's all about workout compliance and hitting the prescribed time in the prescribed zones. Not about the NP. NP is just a math way of simply aggregating that data into the TSS.

Best way I can put this is an example, which of the below will provide a gain in fitness and illustrates "compliance"?
Goal: 1 hour of VO2 work in the week
Athlete A: athlete hits 250w NP and 200 AP for three 30min workouts in a week........but only accumulates 15min of VO2 work
Athlete B: athlete hits 220w NP and 200 AP for three 75min workouts in a week........accumulates 60min of VO2 work, 20min per workout

So, athlete A demonstrates a pattern that may lead you to think they nailed some VO2. Solid NP. But didn't. Athlete B did a good long warmup and cooldown while hitting their 20min VO2 per workout. But had a lower NP.

Which was compliant in the plan to work on VO2?

Clearly athlete B.

The ONE time I would really pay attention to NP is for time trial or a triathlete. Your AP and NP should be pretty damn close for a good race. It makes for good data analyzing your race there. Or it could reveal how efficient you are in a road race or crit. NP is good for post race analysis, not so much for workouts.

IMO.

chaadster

I think we’re on the same page with this, although you seem to have more focus on the actual value of NP as expressed in watts. I honestly don’t think about NP at all— I have a trainer to handle the planning— but it could just as easily be expressed in letters for me, like A means NP and AP are the same, moving right through to Z, expressing extreme power swings. I’m not suggesting that would actually be better, I’m only saying that NP doesn’t tell me anything more about anything other than how a workout was done in terms of effort variability.

Training tools only have value when used with intellectual honesty for their intended purpose. Someone using the NP number to signal how cool they are compared to other riders is the exact opposite of that.

rydabent

I think you are over thinking the enjoyment of cycling.

tomato coupe

I think you're under thinking that different people ride for different reasons.

burnthesheep

Move along....street's closed recumbent boy.....----> https://www.bikeforums.net/recumbent/

aclinjury

normalized power is VERY IMPORTANT, and I'd argue even more so than average power.

So given 2 different rides:

Ride 1:
steady state for 1 hour with 200w Avg and 200w NP

Ride 2:
hill repeats for 1 hour, with 190w Avg, 240w NP

Ride 1 could be feeling like a "sweet spot" ride and you would probably be able to do it again next day without problem.
Ride 2 on the other hand will feel much harder and cost more physiologically, and you probably won't be able to do it again next day (unless you want to feel dead for the remainder of the week)

You need to understand both numbers if you wanna understand exercise physiology.
Avg power is easier to observe in realtime during training and can be used in realtime to gauge your effort, especially if it's a steady state effort.
Normalized power is a metric you would analyze after your training session, it's not something you'd want to use during trainining.

https://www.trainingpeaks.com/blog/w...malized-power/

burnthesheep

Again, my example above. Your comment may work if analyzing a race. Doesn't matter for workouts. In a race autopsy it might be great to see as a gauge of how a rider raced.

For "exercise physiology" it's possible the rider with the lower NP did the proper warmup and cool down for the workout. I don't log three different workouts for one session.

Even if you do, all you need to do is zoom in on the sets you're concerned with anyway. Most tools out there give you the data for the current time selection.

Workouts should be planned around the time you want an athlete to spend in the zones necessary to cause the adaptations you want. The targets are set by benchmarking through testing or looking at recent PRs for given time durations. Looking at NP and going "hmmm, that looks nice" or "hmmm, that looks low" doesn't given any credibility to the effectiveness of a workout.

I've logged 240w NP for an hour workout just chugging along at some tempo. I've also logged the exact same NP for VO2 workouts. So, that would say the NP for the VO2 workout is low and bad, right? Wrong. I did a full 12min warmup and 12min cooldown and maybe tossed some Z2 at the tail end of the workout.

What if I cut the VO2 workout shorter and don't do any cooldown or Z2 at the end? The NP jumps from 240w to 280w. Whoa, that must be great right?!!!! No. Not necessarily.

mr_pedro

To estimate your FTP you need to take average power. Normalized power is a measure for how much stress you incurred.
If Normalized Power is much higher than average power, it does mean you could have produced a higher average power by being more constant.

tomato coupe

Actually, there are better ways to measure power. The first is known in the scientific world as "typical" power. Basically, you use the highest power you saw while randomly glancing at your Garmin, and round up to the nearest multiple of 10, e.g. 340 W, 350 W, etc. The other really good measure of power is called "expected" power. This is the power you think you could produce if you trained more, multiplied by 120%.

aclinjury

I'm not talking about stats padding strava.
I'm talking about using NP when assessing an intervals.
The OP is asking about his best 20 min effort, producing 303w avg and 335w NP, within this 20 min effort. I assume that when someone is asking this sort of question, he is doing a hard effort, and would like to understand what 303w indicate, and what 335w indicates. There is a discrepancy of 32w in his Avg and NP, that means that there are some higher highs and lower lows (and indeed he said rolling terrains). And since 335w is NP, then there are times that he would need to go way harder than 335W, perhaps with many short bursts of 500w efforts.

Is this what the OP wants to do? Is this what he should be doing? Maybe he want to train his out of saddle attacks and then still be able to sustain high effort after such attacks. Whatever is his goal, I'm saying that he needs to look at both numbers to fully assess his effort. If he's looking to train to simulate hill attacks and still be able to "tempo" away after the attacks, then NP is definitely an important metric to study, because perhaps he would better off aiming for 280w Avg but with an even higher 350w NP for the 20 min, which would probably cost him more physiologically, but he won't know this until he actually does the exercise at different intensities and do a post-analysis. And depending on his goals, he may put more weight on looking at Avg power over NP. But to say NP is not important or even not as important as Avg is to not understand its application beyond Strava padding.

I'm not talking looking at NP on strava and stats padding or warmps/cooldowns. I'm talking about using NP as a postworkout assessment tool for the intervals of interest.