46 Minute Power = what fraction of 1 hour power?

Based on that info, I wouldn't change anything. The AP and NP for that duration don't contradict a FTP of 335 (or, for that matter, one of 330.)

Personally, I don't use NP for anything other than the TSS-related extrapolations. If you don't want to test, but feel you have a few max efforts of different durations in your files, you could plug them into the Monod equation.

btw, why was your FTP of 330w 'obviously' wrong?

Roadie_scum said:

I have just returned to serious training about 4 weeks ago, had my FTP set at 330W, which was obviously wrong after I found my legs in a week and a half. I have set it again at 335W. Raced a crit yesterday, NP for 46:17 was dead on 350W and I didn't find it that hard (about a 10-15 minute period was very easy because teammates were up the road and noone was bridging so nothing to cover, leadout at the end did blow me a bit though and I was a bit stuffed after). AP was 318W. Is it safe to say my FTP has gone up? What number should I be using? I am not going to formally test for a few weeks because it does my head in and I hate it (there is also nowhere good to do it around here and I can't hit numbers on the trainer - when I have time to go do a nice one hour alpine climb in a few weeks I will have a better idea).

By the way, don't get the wrong idea - my numbers aren't that good because (i) I am overweight right now by maybe ~5 kg, (ii) even if I wasn't I'm still a reasonable sized unit and (iii) I have been riding for a while (not always consistently, admittedly, but I'm not like some of you guys who ride for a year and hit 300's).

There is not probably much in it from 45-60 minutes. 1% maybe in an mostly isopower environment.

However, not sure I'd rely on NP from one crit to reset my FTP but I do use it to determine whether I suspect my FTP to have improved (or worsened:p) - perhaps suggest it's time to re-test.

RPE is usually a clue which, as you commented, was not off the scale. What sort of NPs did you used to typically get in these races before and what was your FTP at the time? Any pattern there?

I almost always thump out NPs much higher than FTP for crits of 30-50min, mainly due to my reasonable and highly repeatable NMP.

e.g. some crits over last 6 months:

#: Time: AP; NP; FTP (NP/FTP%)
1: 48:09; AP 276W, NP 350W, FTP 295W (119%)
2: 37:44; AP 309W; NP 348W; FTP 300W (116%)
3: 30:48; AP 295W; NP 338W; FTP 300W (113%)
4: 31:18; AP 310W; NP 338W; FTP 300W (113%)
5: 39:33; AP 289W; NP 306W; FTP 300W (102%)
6: 37:12; AP 278W; NP 328W; FTP 310W (106%)

#1 - suspect FTP needs resetting but still needs caution, check out the variability index of that ride (1.27). One week later I did a TTT with NP right at FTP.

#3 & 4 were the same day, couple of hours apart.

My FTP is determined via data obtained from 16km TT and MAP tests. I don't do 1hr tests.

Alex Simmons said:

My FTP is determined via data obtained from 16km TT and MAP tests. I don't do 1hr tests.

Since neither of these will provide a super-accurate estimate, just how confident are you of your functional threshold power?

EDIT: To be more specific, I think it is clear from the data that you presented that you routinely underestimate your functional threshold power. For example, in several of those criteriums - where your power was highly variable, such that the average underestimates the actual physiological "cost" - your average power was greater than your assumed functional threshold power at that time.

Not to speak for Alex, but AP was greater than FTP only in #2 and #4, both of which were shorter than 40 minutes. On their own, this wouldn't make me change FTP estimation -but #4 would get me thinking, since it was the second crit of the day and still was 10w over FTP estimation.

I really like the concept of normalized power, and (for me) it always seems to be consistent (spot on) with what I can do at a stady output for that duration.

I - however IS NOT much of a sprinter - and it is my feeling that riders with a HIGH peak sprint power may be able to "bust" the algoritm used to calculate normalized power in a crit because of all the standing acceleration out of corners.

ANOTHER thing is the "racing effect" - you could have had a SUPER day where you just had 5% extra?

Well 46 min. power REALLY should be close to FTP wattage - maybe 3-5 watts higher. I would say that your FTP is around 345 watts based on the crit. The AP of 318W is quite high compared to the NP - I think it suggests that your FTP is higher than you think!

Roadie_scum said:

I have just returned to serious training about 4 weeks ago, had my FTP set at 330W, which was obviously wrong after I found my legs in a week and a half. I have set it again at 335W. Raced a crit yesterday, NP for 46:17 was dead on 350W and I didn't find it that hard (about a 10-15 minute period was very easy because teammates were up the road and noone was bridging so nothing to cover, leadout at the end did blow me a bit though and I was a bit stuffed after). AP was 318W. Is it safe to say my FTP has gone up? What number should I be using? I am not going to formally test for a few weeks because it does my head in and I hate it (there is also nowhere good to do it around here and I can't hit numbers on the trainer - when I have time to go do a nice one hour alpine climb in a few weeks I will have a better idea).
...

well an average power of 318W/350N for 46-min isn't inconsistent with an FTP of 335 but the real question is "just how hard was it"? It's pretty much impossible for the gallery to judge that.

Why does testing do your head in? My power is always somewhat lower (say > 20-min duration) indoors but I've learned to correct for that.

peterpen said:

Not to speak for Alex, but AP was greater than FTP only in #2 and #4, both of which were shorter than 40 minutes. .

True, but we're talking about criteriums here...the fact that Alex's average power is greater than his assigned functional threshold power at all is what jumps out at me. As someone who has rather poor neuromuscular power, my only hope of getting results in races such as he contested is to continuously bang away at the head of the field from start to finish, hoping that somewhere behind me the chain will snap and a breakaway will be created. Despite my use of such "scorched earth" tactics, my average power even in a 30 min + 5 lap criterium has never exceeded ~95% of my functional threshold power.

acoggan said:

Since neither of these will provide a super-accurate estimate, just how confident are you of your functional threshold power?

EDIT: To be more specific, I think it is clear from the data that you presented that you routinely underestimate your functional threshold power. For example, in several of those criteriums - where your power was highly variable, such that the average underestimates the actual physiological "cost" - your average power was greater than your assumed functional threshold power at that time.

You know, it's funny because I've always had in the back of my head FTP being what average power one is capable of in an isopower or near isopower effort. But I suppose that's not necessarily the case? I've always been able to bang out better numbers in highly variable power situations than isopower. Motivation presumably. I don't do long TT races.

I certainly don't have any ~1hr TTs in my schedule (the TTT was the exception and even that is not isopower but it was a good validation that FTP was not wrong).

I should have added that I also have my 2x20 L4s as validation, which in my most recent sets were meant to be completed at 100-103% FTP and I definitely struggled with those managing 100.5% average (or indeed needing to pull the pin on one set).

Still, it is the conundrum for someone not regularly doing longer TTs type efforts. I'm just going with the method:
- use AP from L4 sets and NP from races to provide indication that change may have happened and might be time to re-set FTP
- use tests with a repeatable protocol to decide on scope of change
I figure that way I am at least consistent.

I also use RPE vs "zone" or "level" as an indicator - if RPE tells me that L3 power is not hard enough for L3, then that's another sign.

peterpen said:

Based on that info, I wouldn't change anything. The AP and NP for that duration don't contradict a FTP of 335 (or, for that matter, one of 330.)

Personally, I don't use NP for anything other than the TSS-related extrapolations. If you don't want to test, but feel you have a few max efforts of different durations in your files, you could plug them into the Monod equation.

btw, why was your FTP of 330w 'obviously' wrong?

I don't have any pure maximum isopower efforts in my files in the last couple of weeks. Andy suggests using NP from an ~1 hour race to estimate FTP, surely NP can't be that different in a 46 minute race to a 1 hour race?

BTW, I have low NMP and low AWC so I don't think I am an NP-buster.

acoggan said:

True, but we're talking about criteriums here...the fact that Alex's average power is greater than his assigned functional threshold power at all is what jumps out at me. As someone who has rather poor neuromuscular power, my only hope of getting results in races such as he contested is to continuously bang away at the head of the field from start to finish, hoping that somewhere behind me the chain will snap and a breakaway will be created. Despite my use of such "scorched earth" tactics, my average power even in a 30 min + 5 lap criterium has never exceeded ~95% of my functional threshold power.

I like how Andy says things and I go, wait, that's where I was going, except he goes and says it in a smarter way with more information. Thanks AC.

So, yes, 318W is about 95% of 335W and this is despite riding fairly tactically and easily for the middle part of the race whilst my teammates rolled off the front.

Basically I went for it at the start covering attacks and trying to split it whilst my teammates were sitting on (and not helping cover... grrr, not even the ones who were meant to), AP 346W, NP 370W, segment length 11:30, then chilled for a little while whilst they rolled off in a break, it felt quite easy at this point, AP 293W, NP 317W, segment length 16:00, so definitely not scorched earth approach here. At the end, I got a time check of 60s to the break, so figuring they weren't getting caught, started racing to wear down the other guys so my remaining teammate who is a good points racer and madison rider could win the bunch sprint. AP 321W, NP 356W, segment length 19:00. This felt fairly hard, especially the leadout at the end where I did my best 20s power for the race (he won the bunch kick then said 'I thought you went too early, but you just kept going', so I obivously had a bit left - I led the last 3/4 of the lap then he put his nose in the wind with 150m to go and kicked out, nice).

Anyway, I am going to do a 1 hour effort in a few weeks on a real long climb, which is the only place I really test well (or in actual TTs), my 2X20 wattage (not formal 2X20 workout, just going and finding climbs) is sneaking up quickly. With my training history I would expect to see fairly fast improvement at this point and I want my numbers to be about right so I can track my training stress accurately (as right as they can be without riding a trainer for an hour).

That said, I don't get why you guys are so down on NP for working out threshold.

I like it because:

(i) I've used it before and it seems to give me a fairly good idea what I can do steady state AND
(ii) Andy says its fairly good and he is smarter than me AND
(iii) Although I'm not sure if people's conception of an NP buster rider is correct, if it was, it would be the absolute opposite of me AND
(iv) It seems to me that the entire point of NP is to be able to do this (compare variable and isopower efforts)

By the way, Andy, I think I would enjoy racing with you. I'd probably have to shed some kegs to climb with you, but windy flat road races and crits, sounds like we'd line up quite well trying to split things up.

Roadie_scum said:

That said, I don't get why you guys are so down on NP for working out threshold.

I'm not - indeed I'm a great "defender" of the NP concept/algorithm (not that it needs defending) despite being one of those that seems to have some capacity to "bust" it.

My ride #1 is a case in point (I can't post the file from here but it was a beauty for testing NP to the limit). Even at the +/-5 % limit of "accuracy" for NP in a hard hour race, that would put my FTP near 332W - but I haven't been able to do a 16km TT test at that power (best ever is 328W) and that is some way short of an hour. I do acknowledge my isopower abilities are not as well developed as say my crit abilities (I just don't do many TTs), so maybe there is more in me than I give myself credit for or have been able to draw out. What "scares" me even more about that race was it was the day after a hard 3hr road race where I set some sizeable aerobic PBs. So that would also indicate an understated FTP.

But Andy's comments do make me think (he's good at that!). The crits where AP was high would have been handicap crits, where basically I am riding it like a mini TTT hence pretty much banging it out the whole time and not the typical attack/cover/recover style.

What's good I think is that for you, NP appears to be a really good indicator of FTP.

For me I tend to lean to it being closer to 105% of FTP, how much I'm not sure but certainly >100%. But it is consistently that way and is a really good marker for me.

Season before last I would routinely over-estimate FTP. When I back crunched PMC data with a revised FTP, it told some interesting stories. High CTL ramps rates preceding sickness for example were not evident with an over estimated FTP.

Alex Simmons said:

I'm not - indeed I'm a great "defender" of the NP concept/algorithm (not that it needs defending) despite being one of those that seems to have some capacity to "bust" it.

My ride #1 is a case in point (I can't post the file from here but it was a beauty for testing NP to the limit). Even at the +/-5 % limit of "accuracy" for NP in a hard hour race, that would put my FTP near 332W - but I haven't been able to do a 16km TT test at that power (best ever is 328W) and that is some way short of an hour. I do acknowledge my isopower abilities are not as well developed as say my crit abilities (I just don't do many TTs), so maybe there is more in me than I give myself credit for or have been able to draw out. What "scares" me even more about that race was it was the day after a hard 3hr road race where I set some sizeable aerobic PBs. So that would also indicate an understated FTP.

But Andy's comments do make me think (he's good at that!). The crits where AP was high would have been handicap crits, where basically I am riding it like a mini TTT hence pretty much banging it out the whole time and not the typical attack/cover/recover style.

What's good I think is that for you, NP appears to be a really good indicator of FTP.

For me I tend to lean to it being closer to 105% of FTP, how much I'm not sure but certainly >100%. But it is consistently that way and is a really good marker for me.

Season before last I would routinely over-estimate FTP. When I back crunched PMC data with a revised FTP, it told some interesting stories. High CTL ramps rates preceding sickness for example were not evident with an over estimated FTP.

Cool, thanks Alex. That makes a lot of sense.

What kind of consensus is there about a certain kind of rider (a certain power profile?) being more able to produce NP busters than another? Or is there a certain kind of ride that does it?

These are the comments I don't understand:

peterpen said:

Based on that info, I wouldn't change anything. The AP and NP for that duration don't contradict a FTP of 335 (or, for that matter, one of 330.)

Personally, I don't use NP for anything other than the TSS-related extrapolations. If you don't want to test, but feel you have a few max efforts of different durations in your files, you could plug them into the Monod equation.

btw, why was your FTP of 330w 'obviously' wrong?

I would think it would be very hard/impossible for someone with FTP at 330W could do AP 318W with variable power in a crit, or NP 350W for more than 20-25 minutes, let alone do it with a longish section of the ride that felt 'easy'.

LarsEjaas said:

I really like the concept of normalized power, and (for me) it always seems to be consistent (spot on) with what I can do at a stady output for that duration.

I - however IS NOT much of a sprinter - and it is my feeling that riders with a HIGH peak sprint power may be able to "bust" the algoritm used to calculate normalized power in a crit because of all the standing acceleration out of corners.

You sound like me!

Quoted post said:

ANOTHER thing is the "racing effect" - you could have had a SUPER day where you just had 5% extra?

I was reasonably motivated, but not tapered, TSB < -20, I don't think I was abnormally motivated.

Quoted post said:

Well 46 min. power REALLY should be close to FTP wattage - maybe 3-5 watts higher. I would say that your FTP is around 345 watts based on the crit. The AP of 318W is quite high compared to the NP - I think it suggests that your FTP is higher than you think!

That was kind of what I was thinking. Thanks for your post.

Hmmm, perhaps I am just suffering from confirmation bias? Andy/Alex do you think I should be resetting FTP at 340W/345W? Is it worth taking time out for CP testing when I am unlikely to stay at the same FTP for long if my training is going OK?

Roadie_scum said:

...

These are the comments I don't understand:

I would think it would be very hard/impossible for someone with FTP at 330W could do AP 318W with variable power in a crit, or NP 350W for more than 20-25 minutes, let alone do it with a longish section of the ride that felt 'easy'.

I would say it depends on the crit and how the race was ridden. Having that high NP suggests it was highly variable, but certain types of efforts can skew NP - all it takes is a repeated, hard effort that is *just* long enough to trigger the NP algorithm, something that is totally conceivable in a crit. I guess I also have the somewhat contradictory view that, while I find tracking TSS/CTL etc somewhat useful (and therefore buy into the NP concept) I wouldn't use NP to set my FTP, particularly from just one event.

IMO, riding 95% of FTP for 46 minutes should be hard, but not impossible, even with variable power. As to having a section of the ride that felt easy, RPE in races can very strange, IME. I've had races where I never felt stressed, only to look at the file and see that I was repeatedly hitting powers for certain durations that would wreck me in training.

Alex Simmons said:

You know, it's funny because I've always had in the back of my head FTP being what average power one is capable of in an isopower or near isopower effort. But I suppose that's not necessarily the case? I've always been able to bang out better numbers in highly variable power situations than isopower. Motivation presumably. I don't do long TT races.

Motivation would have to be it, because I can't think of any good reason from a physiological perspective that would explain why an isopower effort doesn't result in the highest power.

acoggan said:

Motivation would have to be it, because I can't think of any good reason from a physiological perspective that would explain why an isopower effort doesn't result in the highest power.

Alex's crit numbers are undoubtedly from his road bike while his ITT numbers (and possibly 2x20's?) are from his TT bike. There could be a substantial loss in power on the TT bike which may exaggerate the NP/AP differences.

Or not .... 🙂

rmur17 said:

Alex's crit numbers are undoubtedly from his road bike while his ITT numbers (and possibly 2x20's?) are from his TT bike. There could be a substantial loss in power on the TT bike which may exaggerate the NP/AP differences.

Or not .... 🙂

Not.🙂

My crits and 16km TT tests are on the same bike. Only time I rode with aerobars on the road in last six months was the TTT and that was using borrowed clip-ons on same bike.

same can't be said for my track bike though....

I definitely become a different animal with a number on my back and prey in my sights....:p

BTW - here is a pic of the offending race file for Crit #1 - pic inside a pdf (I found it from an old post "kilowatt club"😉. I think this would be classified as making the NP algorithm work hard!

Alex Simmons said:

BTW - here is a pic of the offending race file for Crit #1 - pic inside a pdf (I found it from an old post "kilowatt club"😉. I think this would be classified as making the NP algorithm work hard!

What's it look like with 30s smoothing applied?

Alex Simmons said:

<snip>
- but I haven't been able to do a 16km TT test at that power (best ever is 328W) and that is some way short of an hour. I do acknowledge my isopower abilities are not as well developed as say my crit abilities (I just don't do many TTs), so maybe there is more in me than I give myself credit for or have been able to draw out. <snip>

Alex,

Maybe you should start your TT effort a little easier.

Say under FTP for the first couple of minutes...?

😉

Jim