"NP Buster" question



kmavm

New Member
May 16, 2005
332
0
0
46
"NP busters" are those rides that seem to "game" NP, producing an impossibly high estimate of what the rider is capable of. I've never produced one of these rides, but then again, I don't race crits, and I've never gone out of my way to try to create an NP buster. Never having looked closely at an NP buster ride, I would have assumed that they exhibited high "VI" (or NP/AP ratio).

Dr. Coggan collects these files, and he's said that many of them have a VI near 1; i.e., average power and normalized power are pretty close to one another. So:

1. Isn't such a ride just as much an "AP" buster as an "NP" buster? I mean, if average power was pretty close to normalized power, but I couldn't put together a steady effort of the same average power, then why blame NP?

2. What the heck? Are there really riders who can produce, e.g., rapidly alternating 400W and 200W efforts for much longer than they can produce a steady 300W effort? This is extremely hard for me to imagine, but maybe my imagination is overly constrained by how I ride and train.

3. If I were such a rider, and I was attempting a course were straight AP pretty much determines time (e.g., a really steep uphill time trial), should I try to ride a "buster", rather than a steady pace? If so, man, Alex, you can stick that in your "optimization" pipe and smoke it :D.

Thanks,
Keith
 
kmavm said:
"NP busters" are those rides that seem to "game" NP, producing an impossibly high estimate of what the rider is capable of. I've never produced one of these rides, but then again, I don't race crits, and I've never gone out of my way to try to create an NP buster. Never having looked closely at an NP buster ride, I would have assumed that they exhibited high "VI" (or NP/AP ratio).

Dr. Coggan collects these files, and he's said that many of them have a VI near 1; i.e., average power and normalized power are pretty close to one another. So:

1. Isn't such a ride just as much an "AP" buster as an "NP" buster? I mean, if average power was pretty close to normalized power, but I couldn't put together a steady effort of the same average power, then why blame NP?

2. What the heck? Are there really riders who can produce, e.g., rapidly alternating 400W and 200W efforts for much longer than they can produce a steady 300W effort? This is extremely hard for me to imagine, but maybe my imagination is overly constrained by how I ride and train.

3. If I were such a rider, and I was attempting a course were straight AP pretty much determines time (e.g., a really steep uphill time trial), should I try to ride a "buster", rather than a steady pace? If so, man, Alex, you can stick that in your "optimization" pipe and smoke it :D.

Thanks,
Keith
:p
As for 1 & 2 - I don't know what Andy collects, he will have to comment on that.

As for 3. well TT optimisation is not about the highest power (average or normalised) it's about the lowest duration on a given course for a given set of constraints.

On a steady gradient, an isopower effort will always produce a shorter time than a non-isopower effort with the same average power.

However, if you can produce a significantly higher average by varying your power, then perhaps you can go faster. Somehow I doubt it though.

When you've ridden one that way that - post the data.:D


For those wondering about the "smoking my optimisation pipe" reference, I wrote a discussion paper recently on Quantifying Time Trial Optimisation, which I posted to the Wattage forum. You can see a copy of it there along with a discussion, or download a copy from here if you are so inclined:

http://www.thelinkup.com/shared/iiw7dyh2nkh8

It's called POI Discussion~v.Word97.doc

Best discussed in a separate thread though.
 
Holy sh*t Alex that is an awesome document. Also 25km which is my distance at Nationals!

All you need now is an infomercial and a few quotes from any old World Champion and you will be rich!

So glad I have a powermeter or two!!!
 
Alex Simmons said:
However, if you can produce a significantly higher average by varying your power, then perhaps you can go faster. Somehow I doubt it though.

That's the real mystery for me. It sounds like some riders really can do this, but I doubt I'm one of them.

FWIW, the optimization work is really neat...
 
kmavm said:
1. Isn't such a ride just as much an "AP" buster as an "NP" buster? I mean, if average power was pretty close to normalized power, but I couldn't put together a steady effort of the same average power, then why blame NP?

2. What the heck? Are there really riders who can produce, e.g., rapidly alternating 400W and 200W efforts for much longer than they can produce a steady 300W effort? This is extremely hard for me to imagine, but maybe my imagination is overly constrained by how I ride and train.
/QUOTE]

1. I don't think NP buster rides tend to have similar AP and NP. I expect Andy will chime in at some point to clarify what he actually said about VI of these files.
2. IME, having produced one or two NP buster rides, the efforts that are the culprit are not "rapidly alternating." Often they are due to a course with a short (<1min) very steep hill that is ridden all-out every time. Two of mine came on the Infineon race track, which includes a 45s hill ~17 percent but has lap times ~6 minutes. One race had AP of 235w and NP of 310w for 57min. I am certain I could not have ridden a steady state effort at 310w then - but give me a couple of local pros to chase up a steep-@ss hill once per lap and I could get pretty motivated.
:p
 
peterpen said:
I don't think NP buster rides tend to have similar AP and NP. I expect Andy will chime in at some point to clarify what he actually said about VI of these files.

It was probably something along the lines of "NP busters are not necessarily characterized by a very high VI", or something along those lines. To be specific, about a third of the files that I have collected have a VI of <1.25.

peterpen said:
2. IME, having produced one or two NP buster rides, the efforts that are the culprit are not "rapidly alternating." Often they are due to a course with a short (<1min) very steep hill that is ridden all-out every time.

Now this is a pretty common scenario, which raises an interesting question: does the normalized power algorithm fail in such cases because the math needs to be tweaked, or does it fall because people are comparing rides that entail frequent out-of-the-saddle efforts - which means that the upper body can contribute much more to power production - w/ rides that are done using the legs almost exclusively?
 
acoggan said:
Now this is a pretty common scenario, which raises an interesting question: does the normalized power algorithm fail in such cases because the math needs to be tweaked, or does it fall because people are comparing rides that entail frequent out-of-the-saddle efforts - which means that the upper body can contribute much more to power production - w/ rides that are done using the legs almost exclusively?
Well the latter would account for my one "as close as I've come to an NP buster ride", which was 28 out of the saddle smacks up a hill after a near u-turn during a 45-min crit championship.

I say "as close as I've come" since 45-min is just that bit short of nearly an hour.

VI was 1.28. Each hit up the hill had a 30-sec rolling avg power ~> 150% FTP
 
acoggan said:
Now this is a pretty common scenario, which raises an interesting question: does the normalized power algorithm fail in such cases because the math needs to be tweaked, or does it fall because people are comparing rides that entail frequent out-of-the-saddle efforts - which means that the upper body can contribute much more to power production - w/ rides that are done using the legs almost exclusively?

I know I sent you a file that was either an NP buster or close to it that was a RR with two or three big hills. I know that during that race there was no way I could get out of the saddle without being spit out the back a few seconds later! IIRC the hills were about 2 to 3 min long.
 
All my NP buster rides are 1m interval workouts. The first ~10s of each interval is out of the saddle. Six or seven 1m intervals with around 5m between them. The rest of the hour is warm up and cool down.

My last crit was heading for NP-buster, but it ended early to get the day's remaining races back on schedule.
 
waterrockets said:
All my NP buster rides are 1m interval workouts. The first ~10s of each interval is out of the saddle. Six or seven 1m intervals with around 5m between them. The rest of the hour is warm up and cool down.
So it's not a real NP buster because the hard effort wasn't ~1hour.

My understanding was that to really be under the conditions of "NP validity" (for lack of a better term), efforts need to be something more than 30 minutes and I believe that Andy likes to study files that are ~1h because, after all, NP is defined as an estimate of what one could do as a steady effort for an hour. This makes the IF from an NP-buster effort of ~1h more "relevant" or perhaps "of interest" is the term I want here.
 
Steve_B said:
So it's not a real NP buster because the hard effort wasn't ~1hour.

My understanding was that to really be under the conditions of "NP validity" (for lack of a better term), efforts need to be something more than 30 minutes and I believe that Andy likes to study files that are ~1h because, after all, NP is defined as an estimate of what one could do as a steady effort for an hour. This makes the IF from an NP-buster effort of ~1h more "relevant" or perhaps "of interest" is the term I want here.
Actually, I'm talking about an hour ride. My first solid example of an NP Buster was AP of 179W, NP of 349. My FTP at the time was around 320W. Duration was 59:53. So, even with warm up and cool down factored in, I had a 349W NP for an hour.

It is interesting though (as pointed out by Andy) that the NP is still much closer to my FTP than AP is, even though it quite a bit higher.
 
waterrockets said:
Actually, I'm talking about an hour ride. My first solid example of an NP Buster was AP of 179W, NP of 349. My FTP at the time was around 320W. Duration was 59:53. So, even with warm up and cool down factored in, I had a 349W NP for an hour.
Got it. So you must be pretty high up there in AWC capability vs. your FTP.
 
waterrockets said:
All my NP buster rides are 1m interval workouts. The first ~10s of each interval is out of the saddle. Six or seven 1m intervals with around 5m between them. The rest of the hour is warm up and cool down.

My last crit was heading for NP-buster, but it ended early to get the day's remaining races back on schedule.
I did a ride this week in which I had an IF of .98 for the first hour which was done on my local crit loop (AP was only 4 w lower than NP). I then rode easy for 10' until I reached my hill loop which consists of 3, 40" to 1.5' climbs, spread over ~5 km.

I rode the hill loop for 50' at an IF of 1.03, with an AP of 77% FTP.

For the 2 hr ride, my IF was .98 and my AP was 83% FTP.

Anyway, because of the nature of the second half of the ride (13, L5-L6 efforts spread over 50'), I was able to do ~2 hrs at an NP of ~FTP.
 
Steve_B said:
Got it. So you must be pretty high up there in AWC capability vs. your FTP.
Yeah, I'm either going to explode off the front at 750m, or I'm leading someone out. It's handy for bridging breaks too :D

postgj0.png
 
waterrockets said:
Actually, I'm talking about an hour ride. My first solid example of an NP Buster was AP of 179W, NP of 349. My FTP at the time was around 320W. Duration was 59:53. So, even with warm up and cool down factored in, I had a 349W NP for an hour.

It is interesting though (as pointed out by Andy) that the NP is still much closer to my FTP than AP is, even though it quite a bit higher.
How well nailed was that FTP?
 
Alex Simmons said:
How well nailed was that FTP?
It was from 5m and 20m tests using Monod to predict. I have since found that Monod estimates a little high for me over 30 minutes. I've never done a 60m test, but I just don't have the ability to maintain power for some reason. I need the variability.
 
holli said:
Pretty good power. In which cat you ride in?
Thanks, I've been trying to bring that FTP up, and it keeps coming, but it's pushing everything else up too.

I only race a 5-10 times/year, mostly M35+ on the road here in TX. I hold a Cat 3 license, but I've won against the 2s in M35+. I don't think I've got the base to be more than pack fodder in a p/1/2 race... well, maybe a 90m crit...
 
waterrockets said:
I've never done a 60m test, but I just don't have the ability to maintain power for some reason. I need the variability.
Just curious, what NP Watts/kg do you get for a hard 60minute criterium (or peak 60 minutes in a 90-minute crit)? I bet it is better than 4.18 W/kg.
 
Steve_B said:
Just curious, what NP Watts/kg do you get for a hard 60minute criterium (or peak 60 minutes in a 90-minute crit)? I bet it is better than 4.18 W/kg.
I won a 4-corner downtown crit with an attack at the bell. Before the attack, my AP was 3.77 and NP 4.16 W/kg.

Following that 1-minute effort, it was up to 3.88 and 4.32 W/kg, respectively.

This race was supposed to be 60m, but they cut it short at the beginning so they could set up more bales because of the rain/crashes. So it was only 40 minutes.