How steady should we be?
From a thread on ST:
--------------------------
For added amazement, take a ride two years from today on a similar course where you don't really pay attention to your powermeter too much. It'll be astounding how smooth it is.[/reply]
Having seen one of Jordan's PT files before...he does a pretty good imitation of an electric motor ;-)
His PT trace (which by definition of the PT power calculation algorithm should be "spikey") is smoother than my Quarq files on the same course. The Quarq should be "smoother" since it's power calculations are done on an integer crank revolution basis instead of a fixed time basis like the PT.
Some people are more "constant torque" than others ;-)
---------------------------
Regarding Tom A's post on ST about one of your power files...
--------------------------
For added amazement, take a ride two years from today on a similar course where you don't really pay attention to your powermeter too much. It'll be astounding how smooth it is.[/reply]
Having seen one of Jordan's PT files before...he does a pretty good imitation of an electric motor ;-)
His PT trace (which by definition of the PT power calculation algorithm should be "spikey") is smoother than my Quarq files on the same course. The Quarq should be "smoother" since it's power calculations are done on an integer crank revolution basis instead of a fixed time basis like the PT.
Some people are more "constant torque" than others ;-)
---------------------------
Interesting comment because this is something I've been contemplatively lately.
I personally have no issues with the jumpiness associated with a 1 sec display rate for real-time power. I believe this is reflective of this "constant torque" or "electric motor" characteristic that Tom A refers to above. So, my power files obviously look like this too. Now, I've seen plenty of Jordan's power files as well as many other athletes (both pros and AGers) and this level of steadiness is really not that common even among athletes who have been using PMs for at least 2 or more years. However, it does appear to be a common characteristic among those who execute well on the bike (assuming an appropriate IF). I would go as far as saying that this is a common difference I see between tier-1 and tier-2 pros.
This leads me to the following questions:
How significant do you believe the benefit is to being this steady? Assuming it's significant, how would you quantify it?
I'm asking because I think this is an indicator that is a bit more granular than something like VI. Although, it's only a visual one. Meaning, there could be a fair amount of subjectivity associated with the interpretation of it.
Thanks, Chris
Comments
Well, my very strong suspicion is that the correlation is due to the fact that you have a well-ingrained style of riding with a relative amount of variability so you return to that style of riding after being outside for a while. It's not like you have a choice on a CT. When I train on a CT, which I also do every winter, nothing changes between June and October when most of my riding is outside. However, I know that I have a well-ingrained style of riding very steady.
Still, you'd need to give me a sense of the typical terrain you ride and the difference in VIs between early summer vs late summer. It's hard to assess someone based on subjective words like low, high, etc.
Thanks, Chris
Not claiming to be anything close to an electric motor but I have never had much of a problem with the jumpiness of the power display and have never used any sort of averages or smoothing. Also have no idea how much of benefit that is but your hypothesis makes sense to me. It is sort of taking the low VI to the extreme.
I think we need to distinguish between variability in training vs in racing.
Training
I train very non-tri-like most of the time. I can ride much harder, make the ride harder, bank more TSS, higher VI if I...ride harder...and riding very surgey is just more fun for me. As I sees it, everytime I go over FTP, or way over, I'm recruiting more fast and slow twitch muscles and forcing them to adapt.
Racing
However, the ability to ride very, very steady is a skill I have that I can turn on and off. I 100% confident that this is the way to ride the IM distance. I, and I hope you, have a global perspective on the day that I'm applying my power to the bike and to the course that is just smarter (I may end up with a faster bike split because I'm managing myself while you're not, or staying on the gas while you're not) from the standpoint of a bike split but also with setting up the runs. I've raced enough, and dog knows I've seen enough races, to know that a 5-10' "disadvantage" off the bike can be more than made up for on the run. This global perspective (it's not about the bike) keeps me disciplined and riding the way I know is best which, as you described it, is probably similar to an electric motor.
To the rest of the Team: make a point to absorb discussions like this and use them to increase your confidence on race day -- that YOU are doing the right thing and everyone else is screwing up their day. Chris and I have been bouncing around on internet forums together for probably...6-7yrs? He's easily the most consistent IM execution age grouper I know. Pay attention, he's knows his shit and guys like him, Patrick, Mancona, Al Truscott and many, many others can litterally save you years and several thousands of dollars in wasted race registrations but doing it the right way, their way, the first time...not the 5th or never like 90% of the field.
Chris- I want to be sure I understand the question you are posing. Are you saying that the small bouncing around that one sees on the PT even when riding "steady" isn't going to be significant enough to show up in the calculated VI, but may still be posing some level of greater effort by the body vs if one were truly riding steady?
Not exactly. Also see:
http://forum.slowtwitch.com/gforum.cgi?do=post_view_flat;post=3028015
(jump to my first post which is about 30 deep into the thread)
First, it's probably only significant enough to change your VI by .01 over something like 112 miles. At least that's my observation so far. Also, it's not an issue of physiological cost. That's basically a wash, imho. The issue I'm contemplating is whether it saves time to ride that steady. Think of it like running a model where power is maintained within a 5% window vs a 10% window for a given terrain. Which one is faster or does it really matter? Remember, 2min over 112miles matters to some of us. I lost a podium spot at IMC by that amount.
Thanks, Chris
Another way to ask the question(s):
At what point does it make a difference? Does VI give us a good enough indication of variability?
But shouldn't we state it differently? I think there's an assumption in the statement above about variability and physiological cost. It's really IF/NP/TSS that determines the physiological cost. So, high variability =/ high phys cost. What it really means is that we're time inefficient. Yes, high variability often leads to high phys cost but certainly not always. I can ride IF = .70/TSS = 260 with high variability (eg. VI = 1.07 at Kona) but that clearly doesn't indicate my phys cost was high. However, it's certainly an indication that I probably left a lot of time on the course.
Here's the real-life example:
IMC Course -- Two athletes all equal except
Athlete A -- Power file has this electric motor or constant torque characteristic and follows our pacing guidelines
Athlete B -- Power file has a bit more spikiness to it but still follows our pacing guidelnes
Now my suspicion (from experience) is that athlete A's VI will probably be 1.04 or 1.05 and athlete B's VI will probably be 1.05 or 1.06. However, what if they're both 1.05? Yes, obviously NP and AP are equal in this case but is it possible given how the power was applied in athlete B's case that it yielded a slightly slower time?
I don't know but I believe all of the current models don't or can't perform this granular level of analysis.
So, is it worth the effort to teach an athlete to be this steady? It certainly can't hurt.
Thanks, Chris
Are you suggesting that VI as it is currently calculated ay not be a sensitive enough metric?
If I understand you correctly, I agree in some respect. In fact, I just had Quarq remove my Rotor rings when it was in getting serviced and bought some E-motion rollers to train on and get my spin back during the winter. I rode them yesterday and looked at the data file and it was tighter at a higher cadence than normal. I usually churn 80-85 rpms on the trainer but tried 90+.
If we set our computers to sample at 1 sec rates then we can, at 95+ rpms, be quite spikey when we ride and still have a pretty low VI. I have a tendency to just hammer or accentuate the top and downstroke of the pedaling circle while adducting the thigh towards the top tube. This maximizes the rectus femorus, vastus groups and adductors, which are big muscles and apply big bonus watts with the knee pulled in. I can just twitch those muscles from one leg to the other at a high enough rpm that my watts are very stready....but my application is very "choppy" and it beats the hell out of my muscles and joints. This started which riding with some roadies as I was trying anything to keep up with the Cat 1 woman and the faster men. I figured the high twitchy load allowed some "rest" for my leg during the scrap and upstroke to give the muscles lactate clearance time and offload my cardio for a split second. It seemed to work but it beat my legs up!
My files looked much cleaner than I expected on the flats. Very low VI....but there would be no brick run cause the quads were crap after.
My N=1 experience tells me you are onto something Chris.
Vince
After those rides
Yeah, that's at least one question I'm throwing out there.
Let me back up by saying that if you haven't noticed by now I'm a research nut. It's just my personality. I enjoy doing it. However, there are at least two questions I always ask myself after researching something. They are:
1. Am seeing ONLY what I want to see?
I think it's interesting how many people aren't willing to admit that they have this problem. Everyone has this problem. Nobody is excluded. We all have a tendency to see only what we want to see and I believe we constantly have to remind ourselves of this fact.
2. Am I "overthinking" the problem/issue?
Clearly the best answer is often the most simple one. Of course, simplicity and complexity are relative terms too. Enough said... Back to the subject at hand.
I have noticed this "electric motor" type profile in a number of athletes who perform well at IM. It's something I've been thinking about for some time now. Now, the smoothness of your profile is also highly infuenced by how many bikes are around you so I have to factor that in too. Obviously this is difficult to know when you're lookin at someone else's power file. This is why I always want to see someone's training data too. Typically I like to see long ride data when riding alone.
Anyway, I think I'll do a write up because it does require some visual aid. Unfortunately I'm way behind on the number of things I want to write about lately.
Thanks for listening!!
Chris
Bias? Absolutely we all have it. I think there might be some validity to what you are seeing especially over longer distances as the toll is greater.
Look forward to your write up.
Vince
I agree, but at the end of the day, I only care which of the athletes ran better...or at least ran closer to their potential. What is your metric for measuring "better" per the above? Bike finish? Run Time? Overall time?
I assumed it was a given that we ALWAYS consider whether the athlete ran better or close to their true potential (eg ~80% of T pace) when doing this analysis. It's never just about their bike time.
Obviously it's even more complicated than that because we have to consider terrain, conditions, etc. For example, running at 80% of T pace is not necessarily realistic on a course like IMC.
All I'm saying is that this electric motor characteristic is common in a # of IM power files where I've felt the athlete has performed well. However, let's be honest with each other, the evaluation of determining a good/great IM performance is a realtively subjective process. It's a bit less subjective for guys like us due to our experience but it's still subjective nonetheless. There is just no exact #, set of #s, indicator(s), etc that tells us whether this is the truth or not. I'm going down this path because I think I'm better at doing this analysis and therefore making this determination than 99% of the people out there.
I find it hard to believe that anyone would run slower just because they were MORE smooth (or smoother) in the application of their power on the bike. Wouldn't you agree?
Last thing... I asked Jordan his thoughts about the fact that his files have this characteristic. His response was that he had no idea this characteristic was unique since he doesn't really look at other athlete's power files. He assumed this was something that all athletes achieve after a couple of years of training and racing with a PM. Clearly that was a bad assumption on his part but it gives you an idea of how much knowledge and insight we develop due to the amount of time we spend on analyzing other people's power data.
EDIT: Big mistake... I said "less smoth in the application of their power" when I clearly meant "more smooth (smoother)...."
Thanks, Chris
I wonder how much it's a trainable vs innate thing, too. I know that on the trainer, my power output looks ruler flat. On the road, it seems to look more like that when I first get off the trainer and less like that later in the season.
Mike
I don't think it necessarily equates to a smoother pedal stroke. For example, my Spinscan figure isn't near as round as they say it should be (although I'm not necessarily a believer that a round pedal stroke = a more efficient pedal stroke either) yet I have the characteristic in question.
Here's another observation: This electric motor characteristic is not something I have ever seen in a MOP cyclist. In fact, I could easily say that the height or amount of spikiness I'm talking about is consistently high, relatively speaking, in cyclist that qualify as MOP and below. However, it's not consistently low in FOP cyclist. Now I have seen a correlation between FOP cyclist and those who have a roadie background. IOW, the combination of those two always seems to yield a relatively high level of spikiness. Of course, one of the biggest challenges I have in the sport is trying to teach a former (male) roadie how to ride steady. ;-) I came into the sport of cycling as a triathlete so I can honestly say that well over 90% of the miles I've put on a bike in my entire life have been on a tri bike and most of those have been in the aero position. Hence, my ability to ride steady is 2nd nature.
Is it trainable? I'm fairly certain it is. The question is: Is there any value in training to achieve it? If I was to look at certain individuals, I think the benefit of trying to achieve it could be huge even if you never get there. Anyone ever see Dev Paul race? The dude is like a freakin' yoyo. No there's a guy who I'd immediately put on a training plan where his primary goal was to learn how to ride that steady. Of course, I'd be shocked if he ever learned how to ride as steady as I currently do today but if he even improved by 50% from where he is today, it would be huge, imho.
Thanks, Chris