Power Mentor Group with Shane & Marc - Closed. (Page 35)

Originally posted by marcag

Let’s talk about Normalized Power.
There are two numbers on a workout summary that you should be looking at : Normalized and Average Power.

Average Power is simple, it’s truly the average during the entire duration. If you climb up a hill at 300w for 10min, come down at 0w for 10min, do this over and over, you average power will be 150w.

More realistically you will go up at 300w for 7min, come down a 0w for 3min and get a average of 210w. Do this over and over for 1hour. You will find that this is much more challenging that cycling for 1 hour at 210w.

Here I did it for you :-) This is a ride this summer, up and down a hill that the pro-tour racers do 19x In Montreal . I did it 6x
http://www.trainingpeaks.com/av/NUNUJUTAHV6I7LRGSOP34RKWX4

My Average power was 202w, just at the beginning of my Z2, a nice aerobic pace. I could hold 202w for many hours.
But by the end of the workout I was gased. Why ?

That’s because average power poorly evaluates the physiological cost of a ride with a lot of variability of intensity. Some very smart people (Skiba, Coggan and others) came up with algorithms to better model this. One is called Normalized Power.

If you look in GC, you will have access to a few versions of such an algorithm. NP was done by Coggan I believe. xPower is similar, slightly different but I believe the algorithm is in the public domain. I have seen yPower as well here on BT. There are a few variations, but the idea is very similar.

Here is some stuff on Skiba’s
http://www.physfarm.com/bikescore.pdf

Here is an important quote from the article explaining NP on TP
http://home.trainingpeaks.com/blog/article/normalized-power,-intens...

“This algorithm is somewhat complicated, but importantly it incorporates two key pieces of information: 1) the physiological responses to rapid changes in exercise intensity are not instantaneous, but follow a predictable time course, and 2) many critical physiological responses (e.g., glycogen utilization, lactate production, stress hormone levels) are curvilinearly, rather than linearly, related to exercise intensity.”


Coming back to that ride I did this summer. I had a NP of 254w and a AP of 202w. The problem is my speed is probably better reflected in that 202w number. My physiological cost is probably better represented by the 254w number. So hitting a high NP during a race is not necessarily a good thing. It just means a high cost, not necessarily a high speed. We can discuss this more later. Terrain, gearing and other factors play a role here.

To simplify it, I like to think that AP will be what will predict my speed. NP will be the cost of it. So I want to maximize AP and minimize NP. Idealy AP=NP. I am not saying AP will reflect speed perfectly, but it is a better proxy than NP.

There’s a cool metric that comes into play : VI.
Variability index which is nothing more than NP/AP. If the two are equal, VI = 1.0. In my ride this summer VI = 1.26 very bad if I was doing a tri.
While racing, you want your VI as close to 1 as possible.


Here are some rides from the guys at Kona
http://cat6.trainingpeaks.com/races/ironman-world-championship-kona...

Luke McEnzie VI=1.05
Faris VI = 1.04
Linsey Korbin VI = 1.03
Meredith Kessler VI = 1.03

The men are slightly higher because of the race dynamics. They do more surges. But they also probably pay a higher price for it on the run.

NP power is an important concept. We will talk more about it but I just wanted you to be aware of the concepts. You can google and poke around more and of course ask questions.

For training, watch your NP, it is a much better indicator of your training intensity.

Originally posted by littlewj
Interesting! So how would you go about trying to get your AP=NP for a hilly triathlon? How would you determine the wattage you want to hit when climbing?

Jim

Originally posted by marcag



Do as I say, don't do as I do :-) I am bad at low VIs.

It will depend a lot on the course. If very steep downhills it will be hard to. Rolling hills are easier to get a low VI.
Your gearing plays a big part. Having a compact crank or 28 cassette will certainly help prevent the spikes.
A bigger ring up front or a cassette with an 11 helps. If on top of that, you can descend at high speed will minimize the low power values. But at one point, when you start hitting close to 40mph, not a lot you can do. But many people coast the downhills. You need to work them

I try never to go over FTP.
I try to work the peaks to well over the top. I used to climb really hard, gas myself and stop pedalling at the top to recover.
I try to work the valleys as early as possible. Working the downhills at higher cadence is something to work on.

There was discussion previously on cadence. Sometimes very low cadence helps cap the climbs and high cadence allows to generate power on the downhills.

I know others will have other ideas to smooth things out.

The site bestbikesplit may also provide "plans" for optimal power pacing

Originally posted by littlewj

Thanks for the explanation Marc. This really helps. I'm doing Lake Stevens 70.3 in August which is somewhat hilly. I know in my past triathlons, climbing the hills is where I would burn all of my matches.

Jim

Originally posted by marcag



Try bestbike split. Get an idea of what the spikes will be like.
Do you know what the steepest pitches are ? You can pretty easily figure out what type of cassette you need to not run out of gears.
Very hilly I use an 11-28. Flat 11-23, rolling 12-25. But I am "heavy"

Originally posted by marcag



Do as I say, don't do as I do :-) I am bad at low VIs.

It will depend a lot on the course. If very steep downhills it will be hard to. Rolling hills are easier to get a low VI.
Your gearing plays a big part. Having a compact crank or 28 cassette will certainly help prevent the spikes.
A bigger ring up front or a cassette with an 11 helps. If on top of that, you can descend at high speed will minimize the low power values. But at one point, when you start hitting close to 40mph, not a lot you can do. But many people coast the downhills. You need to work them

I try never to go over FTP.
I try to work the peaks to well over the top. I used to climb really hard, gas myself and stop pedalling at the top to recover.
I try to work the valleys as early as possible. Working the downhills at higher cadence is something to work on.

There was discussion previously on cadence. Sometimes very low cadence helps cap the climbs and high cadence allows to generate power on the downhills.

I know others will have other ideas to smooth things out.

The site bestbikesplit may also provide "plans" for optimal power pacing

Originally posted by gsmacleod

I would suggest that in a triathlon (or TT for that matter) time above FTP should be minimal. With an IM, this gives an athlete a rather large window as their target wattage will be well below FTP so it is easy to make a plan where an athlete rides a little harder uphill or into the wind and a little easier on the descents (and tucks after about 30mph) and with the wind.

For HIM events, there will be a little less room between target wattage and FTP but there still should be a large enough gap that you can apply power in a manner that isn't an iso-effort but still keeps wattage below FTP in almost all cases.

For short course where an athlete is likely targeting somewhere between 90-100% of FTP, then more care must be used but even in a sprint where an athlete might target FTP, their goal will be to keep VI low and attempt to spend little time above FTP.

Shane

Originally posted by dtoce
what %FTP do you recommend for HIM and OLY race distances?

Originally posted by dtoce
so, during triathlon races, and racing with a PM, it is best to aim for a goal % of FTP (say 90% for a sprint,) and try to distribute your power/energy uniformly with no matches burned. Later on, see how the legs felt for the run then adjust the number for the next race?

Originally posted by marcag



The easy answer is 80-85% for HIM and 90-95% for Oly. But that answer is IMO misleading.
There are many variables that make up a good target. Confidence in your true CP, ride data, bike volume going into the race, run volume going in....Many factors.

I would say 80% is a rough number to TRY IN A RACE REHEARSAL to see what happens.



yes

There are ways of converting run speed to watts. For hoots the other day I calculated that my good HIM runs are at about 85% of my threshold run power. I bike conservatively. I hit 80% max, sometimes less. 77-78 is fine for me. But I do make it up on the run. Maybe I should try 82%, 82%.

I then started to play with bestbikesplit and biked 10watts less, get a time, add that 10watts to the run see the improvement....and it would be very close. A little less on the bike, a little more on the run

I would say that if you underbike 10-20watts, you can make that time on the run by going a bit harder
Overcook the bike, you will walk and be significantly slower

I would say race on the conservative side of the bike, do the first half of the run conservative and if there is gas in the tank, you can spend it fully and not be any slower.

But again, this is only part of the equation. Fitness, volume and training data play into this. As does swim and run pacing. Also, time on the course (3hr vs 2h20) also is a factor.

FYI, when someone says 80-85% ask them to see a good run that follows that up. Few people can do it. Some can, TomsPharmacy for example, Bryan I bet could. But these guys have HUGE biking engines.

We can and should talk about race pacing more.

Originally posted by marcag

FYI, when someone says 80-85% ask them to see a good run that follows that up. Few people can do it. Some can, TomsPharmacy for example, Bryan I bet could. But these guys have HUGE biking engines.

Originally posted by marcag Do as I say, don't do as I do :-) I am bad at low VIs. It will depend a lot on the course. If very steep downhills it will be hard to. Rolling hills are easier to get a low VI. Your gearing plays a big part. Having a compact crank or 28 cassette will certainly help prevent the spikes. A bigger ring up front or a cassette with an 11 helps. If on top of that, you can descend at high speed will minimize the low power values. But at one point, when you start hitting close to 40mph, not a lot you can do. But many people coast the downhills. You need to work them I try never to go over FTP. I try to work the peaks to well over the top. I used to climb really hard, gas myself and stop pedalling at the top to recover. I try to work the valleys as early as possible. Working the downhills at higher cadence is something to work on. There was discussion previously on cadence. Sometimes very low cadence helps cap the climbs and high cadence allows to generate power on the downhills. I know others will have other ideas to smooth things out. The site bestbikesplit may also provide "plans" for optimal power pacing

Originally posted by GoFaster
Marc - since I know you had a fantastic result at the Muskoka 70.3, can you share how well you managed to keep your wattage below FTP on that course?  I don't have access to my ride file at the moment, but I know I was over FTP, and that was simply dictated by the terrain.  I also wonder about trying to keep the VI as close to 1.0 on a course like that.  In my mind that would have you really spinning a tiny gear to manage the steep uphills and cranking a very large gear in the downhills, and the 40mph governor doesn't really kick in too often on that course.

Very curious since I'm taking another run at the course this year.

Originally posted by RunningJoke
Can I also ask what goal power is in general for a sprint tri?

Originally posted by marcag
average power poorly evaluates the physiological cost of a ride with a lot of variability of intensity. Some very smart people (Skiba, Coggan and others) came up with algorithms to better model this.

Originally posted by marcag Muskoka was probably my best run ever. I have run faster at Tremblant and Miami but as you know Muskoka was a brutal course. BTW, I think they changed the run couse. It was over 600feet of climbing on the run. In hindsight that run was probably more about my run pacing than my bike pacing. There were several "emotional" events that made for a well paced run. A successful HIM is as much about pacing the run as the bike IMO. This is very much a personal opinion but I think people undertthink the run. For the bike I rode at about 78% of FTP with a high VI (1.08). I went back and looked at the ride. 8 times I went over FTP but only for about 1minute. Once for 2 minutes. Highest was 114% of FTP. On that course, it was inevitable. It is also a long course with an extra 4km which means you are out there for longer I was on the bike course 30min more than Miami for example on similar power. While surges at or above FTP are not ideal, for short periods of time I don't think is the end of the world. Avoid them if you can, but sometimes you can't. But it's better to have done a few of these in training.

Originally posted by Andrew Coggan



Normalized power is my original idea, and mine alone. Everything subsequent (e.g., GOVSS, xPower, rTSS, etc.) are merely variations on my original concept of smoothing/weighting/unweighting the raw data to estimate physiological cost.

Originally posted by marcag Muskoka was probably my best run ever. I have run faster at Tremblant and Miami but as you know Muskoka was a brutal course. BTW, I think they changed the run couse. It was over 600feet of climbing on the run. In hindsight that run was probably more about my run pacing than my bike pacing. There were several "emotional" events that made for a well paced run. A successful HIM is as much about pacing the run as the bike IMO. This is very much a personal opinion but I think people undertthink the run. For the bike I rode at about 78% of FTP with a high VI (1.08). I went back and looked at the ride. 8 times I went over FTP but only for about 1minute. Once for 2 minutes. Highest was 114% of FTP. On that course, it was inevitable. It is also a long course with an extra 4km which means you are out there for longer I was on the bike course 30min more than Miami for example on similar power. While surges at or above FTP are not ideal, for short periods of time I don't think is the end of the world. Avoid them if you can, but sometimes you can't. But it's better to have done a few of these in training.

Originally posted by ligersandtions
I'm not currently sitting in front of GC and I haven't tried to figure it out in the past -- but I believe there's a way to analyze on there the number of times you go over FTP, right?  Can you explain how you do that analysis (on a Mac if it's different from the PC version)? 

Originally posted by Andrew Coggan



Normalized power is my original idea, and mine alone. Everything subsequent (e.g., GOVSS, xPower, rTSS, etc.) are merely variations on my original concept of smoothing/weighting/unweighting the raw data to estimate physiological cost.