Cycling training data mistakes podcast episode

5 Common Mistakes You’re Likely Making With Your Training Data

Topics Covered In This Episode:

  • What cyclists get wrong about Normalized Power
  • Setting your training zones correctly
  • How Normalized Power and Average Power compare
  • How to use kilojoules to measure your workload and plan out your fueling strategy

Show Links:

Related Content:


Listen to the episode on Apple PodcastsSpotifyStitcherGoogle Podcasts, or on your favorite podcast platform

Training Stress Score®, Intensity Factor®, and Normalized Power® are registered trademark of Peaksware, creators of TrainingPeaks software.

This Week’s Episode Was Fueled By The Feed

Where do you buy your gels, hydration mixes, and energy bars for training and races? If you are not shopping at The, you are missing out. The Feed is the largest online marketplace for your sports nutrition, offering the brands you know and love, from Skratch Labs, PROBAR, to Maurten, plus their athlete customized supplements called Feed Formulas. Feed Formulas are personalized supplements for athletes, developed in part with Dr. Kevin Sprouse (EF Pro Cycling Team Doctor), following the same protocols the top pro athletes use. Shop and save 50% today on your first order.

Episode Transcription:

Please note that this is an automated transcription and may contain errors. Please refer to the episode audio for clarification.

Adam Pulford (00:08):

Welcome back train, right fans. If you’re a new listener to the show, glad to have you on board. I’m coach Adam, Pulford your host for the cycling edition of the show. We’re in a three part series that I call demystifying training metrics. And this episode is the final in that series. Today’s episode will flow a little differently as I’ve identified some common mistakes or misconceptions that athletes are doing or failing to do when it comes to analyzing, organizing and thinking about their training data. In part, I think this reason, the reason for these mistakes is that this field is evolving quickly. And it’s challenging for many to keep up to those changes as you’re doing your training, your life, your family, and things like that. In my opinion, it’s up to us coaches and thought leaders in this space to continually educate athletes and more clearly show you how best to use these tools.

Adam Pulford (01:05):

And thus this three part, uh, series that we’re doing here. Additionally, some of this can be really confusing for new athletes getting into the sport and also for seasoned athletes who want to go deeper in their data. And they realize they have to change their training to get more or different gains. Finally, I, I also think that there’s some athletes who are just lazy about learning and they see bigger numbers and they just want to use those bigger numbers to break to their friends about what they’re doing on the bike. And then they really get things wrong. And because they might be just, you know, tough or fast or something like this, we listen to them and that’s, that’s also wrong. So either way, my goal for you, the listener is to get educated so that you can apply some of these learnings to your training to get faster, or to identify some BS when you hear it from your friends.

Adam Pulford (01:56):

So that said, let’s get right into the show. Common mistake. Number one, using normalized power to set training zones. Don’t do this. Why? Well standard training zones are set mostly around the aerobic glycolytic energy system for accuracy. So doing your 20 minute FTP test that’s designed that is that test is set up in a way so that you take the highest average power from that test. You subtract 5% from it and you use that for your FTP. And we know that is a very accurate way to set the hinge point or to set up your training zones for accurate training. We’ll get into that a little bit more later, but again, you want to take the highest average power from that FTP test to accurately set ranges, not the normalized power. If the normalized power is drastically higher during an FTP test. Well, I I’ll still take the average power and you also might wanna just retest because the normalized power should not be higher than the average power, because, and the reason why we don’t wanna do this is looking at the normalized power can skew your training zones.

Adam Pulford (03:17):

If you set your FTP by it, that’s the primary aspect, why we don’t use it. And so to better understand that and to better understand normalized power, I think it’s best to dig in and explore what it actually is, where it came from. And some of the nuances around it, many people know the term normalized power, but do you really know what it is? Probably not. So let’s look into it, normalized power. What is it? It’s an algorithmic function designed to give better insights to the cost of an effort during a ride compared to the average power alone. So what does this mean? Because cycling is never steady, truly average power does. Doesn’t tell the whole story here. Additionally, physiological responses are quick to intensity changes, but they’re not instantaneous. Meaning if I press on the gas pedal, you know, power goes up muscles starting to get fatigued, but the way that we fatigue it’s, it’s not instantaneous.

Adam Pulford (04:21):

You don’t do a one second sprint, get tired and then fall off your bike. Okay? But these physiological changes, they are predictable, normalized power considers these rapid accelerations, deceleration, the anaerobic and the aerobic effort, as well as these physiological changes. The physiological changes I’m talking about are things like glycogen, depletion, lactate, production, and accumulation, stress, hormone activities, things like this. It’s, it’s more, it’s pretty complicated, but it, it was designed by very smart people and it works very accurately. Okay. And all of those things taken into consideration into normalized power. This gives us the better estimate of the cost to an effort given while you are riding or racing your bike without getting too deep in the weeds when calculating normalized power, each data point in the selected range that you’re looking at, meaning if you’re on training peaks or, or, uh, WK oh five or some sort of analytical tool, if you go to your group ride or your race or your, your interval session and you select however long those intervals are, or the race was that’s the range that we’re looking at here.

Adam Pulford (05:44):

And so when you’re looking at that range and you’re looking at the average power and the normalized power that normalized power is calculated using the aggregate of the 30 seconds before that particular data point for every single data point, okay. Essentially using a rolling average of 30 seconds to analyze each data point. Okay, now that can be kind of confusing. And most people don’t really need to know how it works only that it works and it works really, really well. And I’ve really simp, um, simplified how normalized power worked, even glossed over a few of the nuances for those in order who really want to learn how normalized power is calculated. I’ve linked to that in the show notes, um, on our landing page at train, pack slash podcast or train right podcast. So you can learn more about it there because there is some math involved, but in, in summary, a normalized power and how it is calculated, it’s it is pretty complicated, but it really is based on this 32nd kind of like rolling average with fancy math that goes in and it’s taking into account how you fatigue and these physiological changes that are occurring during, uh, aerobic and anaerobic efforts.

Adam Pulford (07:02):

Okay. So it’s taking in the highs and lows into the account to get that better estimate or cost that we’re talking about. So going back to the question then, if normalized power is so good at doing this, why can’t you use it to determine FTP? Well, as I said before, most training zones are mostly based on aerobic glycolytic energy production. You will have a more accurate training zone or zones if you base it on the more steady state, highest average power from an FTP test that is evenly paced, normalized power. Isn’t perfect. Okay. And depending on the race, the course, the group ride the effort given it could take that anaerobic effort into too much consideration, which would then overinflate your FTP, which would then overinflate your training zones, which would then cause you to kind of train harder than you think you’re actually doing, which will lead to a lot more fatigue, which will lead to a lot more frustration.

Adam Pulford (08:08):

And relative to the recovery, you’re not gonna have very accurate and effective training. So essentially you’re always gonna be reaching higher than you probably should for the goal of the training session. And that could lead to just burnout, fatigue and training too. Mu train too much fatigue relative to recovery, even though, uh, I’m not gonna say the word overtraining because, uh, <laugh> philosophically, I don’t believe in overtraining. So that said you can do too much overreaching in a certain time period and, and that’ll cause some problems. Again, I, I do look at normalized power for sure. And in particular, I look at normalized power for 40 to 60 minutes in particular of say a race or a group ride or something like this. And that’ll in that 40 to 60 minute normalized power. That’ll give you some really good indication of how hard the effort was and perhaps where FTP is headed.

Adam Pulford (09:08):

Okay. And what I mean by that is it’s gonna give you, like, if you are doing a bunch of say, anaerobic hard training going pretty hard, you go and you do a race criterium 60 minutes, and we see that your FTP or sorry, your normalized power for that effort is maybe like 20, 30 Watts above your FTP. It’s a good indicator that it’s like, oh yeah, you know, we got the, the engine going in the right direction. And, and perhaps eventually we could retest to check in and see if that FTP has increased or it is increasing along the way, but no need to change anything. So, um, I use that as is again, good indicators of how the athlete is going and how hard it was, but I’m never gonna set training zones by it.

Adam Pulford (09:58):

All right. So now that we know what normalized power is, let’s go into common mistake. Number two, looking at normalized power for durations, less than 20 minutes, kind of a pet peeve of mine carrying on with that normalized power trend here. I find too many athletes look at normalized powers on group rides or intervals. And, and they’re looking at, you know, short hill climbs of 5, 7, 10 minute efforts or something like this. And they look at normalized power and they’re like, oh, my normalized power for seven minutes was 4 0 5. And it don’t do that. Why? Because the, a, it wasn’t built for that. The algorithm wasn’t designed for a use that is shorter than 20 minutes. So you’re not gonna get accuracy or you’re not gonna get this better cost or more insights on what the normalized power should be for that 7 5, 10 minute effort, whatever it is.

Adam Pulford (11:01):

The algorithm was created to look more broadly over long periods of data, 20 minutes and beyond for more accuracy of the cost associated with those efforts. Okay. There’s just too much individuality and too many variables to consider when looking at squiggly stuff, that’s short. So, you know, one minute up to seven or 10 minutes, like I said, uh, where we can’t yet have a tool that gives us more insight, but I don’t think that we need it because again, when you’re looking at average power, um, relative to normalized power, or when you’re looking at average power for some of these short efforts, that’s good enough relative to heart rate response. And, and, um, and then you look at kind of a more sweeping of what happened during the ride itself, you know, 20 minutes onward looking at the normalized power, that’s gonna give you more insight, the insight that you need.

Adam Pulford (11:58):

So normalized power. Isn’t perfect, but it’s darn good. And just make sure that you’re using it for its intended use 20 minutes and beyond not shorter. And this is where, uh, this is where that big number comes in. Again, you can look at shorter durations and sometimes get a bigger normalized power, and then, you know, go spout off to your friends, spout it, but it’s not accurate. So don’t listen to ’em. All right. So common mistake, number three, not looking at average power at all. So I see some athletes out there just talking in terms of normalized power all of the time, and that’s not good either cuz some people it, so the reason why that’s bad and in some mistake is that normalized power was not created to forego other metrics and other data altogether looking at average power as well as normal as power.

Adam Pulford (12:50):

Just like I talked about before will bring more insights to how the effort was for the athlete. And so that self coached athlete or the coach looking at the athlete’s data needs to use these metrics and data points together to really get a good idea of what happened out there on the road, especially since, you know, a coach probably is not out there riding with you or for the self coach athlete to really have a better account of what happened out there rather than try to re you know, remember or remember the feel of what happened out there. So the key thing I look at when, you know, especially for average power is on solo long rides over time. Okay. And over the course of a season, I want those solo long rides, uh, the average power to be higher or pro kind of progressively building as we go.

Adam Pulford (13:47):

And why solo long rides? Well, if we’re talking like a zone two, um, endurance smiles sort of long ride, long ride, meaning just whatever’s longer than your normal ride could be three hours. For some people could be five hours for some people it’s you versus you in more of a steady manner. And the average power is way better. You know, the best thing to look at there because normalized power probably shouldn’t be all that drastically different because again, it’s, it’s a fairly steady, uh, effort out there on the road. And this I find is a better marker for aerobic development on those long rides rather than normalized power, but I’m still looking at normalized power on those rides. Okay. And common mistake, number four, not looking at kilo jewel expenditure during your rides. So first what the heck is a kilo jewel. You probably heard about it, but we don’t really understand it all the time.

Adam Pulford (14:42):

So first it’s the standard base unit to quantify energy work and heat. And it’s set by the international system of units or an SI unit, how it’s measured from your power meter. A Juul is defined as Watts times, seconds, and ACHI Juul is 1000 of those. So there’s some fun, ironic math that goes on to make all of this work. But the simple fact that you need to know as an athlete is one kilo Juul equals one calorie. Again for math nerds. I’ve linked to how we get there in my show notes in terms of the math that’s, uh, laid out for that. But the main takeaway is if you’re looking for the most accurate way of measuring calories burned via cycling, you’ve had it all along. It’s in kilojules. And when I roll up on somebody and say, you know, I’d say we’re three hours deep on a five hour ride and it’s kind of a lumpy course.

Adam Pulford (15:42):

I’m always kind of curious, Hey, how many kilojules are you running? And, and so many athletes are like, how do I, how do I find that? It’s like, well, you can have it directly, you know, Mount on your, your, your wahoo or your cycling computer. And they’re like, oh, I never, never knew that. And then what’s a kilo Juul, and then we’d start talking about it. And so again, it’s, it’s a very accurate way of measuring calories burn during cycling. Well, why is that important for so many reasons? Okay. Uh, a lot of people, you know, they wanna know how many calories you burn so that you know, how many calories you can eat. Right? Um, <laugh> but that’s not the only reason. Okay. Knowing how many calories you burn total and exercise can help you determine an ideal nutrition and hydration plan for the ride itself.

Adam Pulford (16:30):

So this, this could be a whole episode by itself, but bear with me here, because I’m just gonna do a quick kind of rule of thumb that I use with my athletes when looking at total K jewel expenditure and setting up, um, an appropriate, uh, fueling plan for the day. So for, for O rides that are two hours and more, and let’s just say, I want performance out of ’em, meaning we’re gonna do some intensity. It’s either gonna be a group ride or hard interval session, but again, probably two hours and more, I’m gonna take a look at that total kilo ju expenditure, and then I’m gonna multiply it by 30 to 40, per 30 to 45% of that total kilo ju expenditure to get a, to a total calorie, then I’m gonna break it down per hour for that fueling basis. Okay. I kind of hacked through that.

Adam Pulford (17:23):

So let’s, let’s use an example. Let’s say that an athlete has done a three hour ride and they burn 2000 kilo drills just for, um, nice round numbers. We’re gonna take 35% of that and it’s gonna come out to 700 calories total. So then if you divide that 700 calories by three hours, we come up with around 230 calories per hour of intake that the athletes should do to fuel properly for that ride. Okay. This works for most people in terms of a 30 to 45% intake of output. And that output again, is a kilo jewel measurement based on your power meter. And some, some athletes can do more. I wouldn’t go much less than 30%, especially if we’re talking again two hours and more, and we want performance, meaning you’re probably gonna be going hard on this ride. So don’t go below 30% and over 50%, you might have some GI issue, but again, start to explore with those, those percentages or those ratios.

Adam Pulford (18:32):

And I think you’ll find some, some really good success. And for those who have been having some problems, either with GI issues or under fueling issues during a bike ride, start with, uh, start with these ratios and also start with knowing how much kilo jewel expenditure you have during a total ride. And that’s gonna really help you out. Okay. Knowing total calorie burn during exercise also can help you eat for your goals. Okay? So whether you wanna lose weight, gain, weight, maintain weight, or maintain performance over time, again, knowing how much you’re burning is gonna help you to know how much you can eat. Not only for the celebratory, uh, meal post, you know, post-race, but also just kind of a, more of a longitudinal approach on a weekly, monthly and yearly basis.

Adam Pulford (19:20):

Total work done on the bike is also a really good way to measure workload. I’m gonna expand on this, like right now, and then we’re actually gonna come back to it on the, on the final mistake. But, uh, I I’m gonna set it up for you right now. Okay. So if your biggest ride has been around 3000 kilojules and then all of a sudden you do a ride that’s 4,000 kilojules okay. I’m just using kind of, you know, round kind of some of these bigger numbers too. Okay. But I’m using round numbers to just make my point. So if you did a 3000 kilo jewel ride, then all of a sudden you did 4,000, you did more total work than you’ve ever done before. Okay. We’re talking like the biggest rides you’ve ever had before. And if you do more total work, you’re gonna cause more fatigue than you rest. And then you get better. Remember from all the past episodes, I’ve talked about training stress plus rest equals adaptation. Okay. And when you’re stressing the system, you’re doing work, you’re doing aerobic work. Okay. And pumping out more kilo jewels is a really, really, really good measurement of work done. Again, your as long as the power meter is accurate and cons or accurate calibrate and consistent, we can, we can hinge upon. We can, we can make great strategies based on accurate numbers.

Adam Pulford (20:44):

Additionally, we can start to monitor work rates overtime. So going back to this 3000 kilo ju example, if you did that over the course of five hours, you simply take 3000 divided by five and we get 600 kilojules per hour. If you start doing five hour rides that are, say 3,500 kilojules total, that’s 35 divided by five equals 700 kilojules per hour. And this is more work per hour, which is going to give more gains over time. So you say, well, wouldn’t your average power. Maybe even like your normalized power, be higher on these rides that you’re doing more work on. And yeah, it, it will be slightly okay. But again, what I’m, what I’m saying is it’s not only, it’s not one metric that we’re, that we’re looking at all the time in order to get these insights on what’s actually going on, because you can do a 3000 kilo jewel ride in zone two.

Adam Pulford (21:47):

You can also do a 3000 kilo jewel ride with zone two to six mixtures. Okay. So how you’re actually producing that power and the total work done makes all the difference when it comes down to where the gains are being made for you as an athlete and for a coach looking at your data. Okay. So just we’ll put that total work concept to bed for a second. And I’m gonna go onto my fourth point here in terms of, uh, looking at K jewels. So you can also estimate your long ride kilo ju expenditure for nutritional planning. Meaning if you say on this, my theoretical, uh, numbers and examples here, if five hours has always been your long ride and you’re planning on doing a six hour ride here’s, here’s how I work through again, with my athlete of, um, about how many kilojules total, or how many calories total we’re gonna need on the day and BA, and that’s all based on the projected kilojules that they might do on this long ride.

Adam Pulford (22:57):

Okay. We do this by taking the total duration in seconds multiplied by the projected watage that the athlete is gonna do, and then simply divide by a thousand. So let’s take this as a real life example, we’re gonna do a six hour ride and six hours is 21,600 seconds gonna multiply that by say, let’s just use 200 Watts for a nice round number. And, and the way I get at this average power is I just look at what the athlete has done for, you know, four and a half or five hours prior to, or whatever their long ride is, what their average power. And I’m gonna take that, maybe bring it down just, just a bit, or take that number in particular. So I’m gonna multiply the total duration in seconds, by the average power that they are projected to do 200 Watts, and then I’m gonna divide by 1000.

Adam Pulford (23:52):

And if you do that math, you get 4,000, 320 kilojules or calories burned on that long ride. If you don’t believe me, that it’s that easy to quantify, how many kilos you’re gonna burn out there. Uh, go ahead and try it. Let’s go back, look at some of your long ride data, uh, break everything down into seconds, multiply it by the average power that you did, then divide by 1000 and you’ll find that this works every single time. Okay. Then if you got a long ride coming up, go ahead and project it, and you can then base a nutritional plan based on what I was talking about earlier on a per hour intake of output that you’re gonna do. So in this example that I just used for six hours, you can dial in your calories per hour, based on the fact that you’re gonna be burning 720 calories per hour in that six hour example.

Adam Pulford (24:48):

And if you take 35% of that, it’s gonna be roughly 250 calories of intake. That would be your goal for a performance goal. So some pretty, pretty, not, it’s pretty simple math. You just gotta have to work with the numbers a little bit, but it’s really actually fun to do that once you have a, an accurate grounded basis to base the numbers off of. Okay. So I’m just giving you some of these, uh, little fun tips that I use in my coaching that I think will, will really apply to your training in racing to help just simplify what you’re probably doing already. All right. So our fifth and final common mistake is only using TSS as a training load and not looking at kilojules or normalized power over these longer term, um, training time periods. Okay. So what I mean by that is in past episodes, I’ve mentioned some things to consider around training stress score and CTL.

Adam Pulford (25:54):

If you miss those past two episodes on this, uh, this little series, go back and listen to it. Cause this point will make a lot more sense if you listen to those first. Okay. What I said in those past episodes is at some point you’ll run out of time and TSS and CTL will plateau, but no need to freak out because you can still make gains during those time periods. And one of those ways to make gains is to look at improving your Kage jewel total per week, your Kage jewel, total per ride, your Kage jewel, total, or, uh, per hour on these long rides. And also looking at normalized power on the rides. Additionally, if you’re, so if you’re doing group rides or you’re doing like hard interval sets, and we’re, we’re really talking about some like anaerobic, uh, training here, check that normalized power out because if your normalized power is growing during these time periods, yet you just kind of run out of time.

Adam Pulford (26:53):

You’re not gonna see your fitness or your CTL improve, okay. TSS will stagnate, but if you’re doing more work, kill the jewel power, or you’re doing more normalized power for 40, 60, 90 minutes, those are the metrics that you wanna look at to see if you’re actually gonna be moving the needle during these time periods where other stuff may be plateauing. Okay. Often these things are going up in the background, but we’re so focused on these other common metrics like TSS, CTL, or FTP, uh, that might not be changing over the course of three, four months. And so I want you to start looking at some of these other metrics to get a good handle on how you’re making some improvements or where the improvements could be, could be coming from when you see a change in your FTP, or you see a change in your VO, two max down the road.

Adam Pulford (27:49):

All right. So in summary, I threw a lot. I threw a lot at you, um, over the course of the series. And today, today was pretty heavy on even some of the math or how I get to, uh, some of the solutions in, in, in my coaching practice. But hopefully you’re starting to see how all these metrics, algorithms and terms need to be looked at and analyzed together to get the most out of your training and to get the most outta your data. Over the past three episodes, I’ve taught you more about TSS, CTL, FTP, TTE, normalized power, and a jewels. I also taught you how to use these metrics to apply to your own training and to apply this coaching knowledge, to get faster today. I hope you at least learn that normalized power doesn’t apply to efforts that are 20 minutes or less.

Adam Pulford (28:42):

And that kilojules is not this like mystifying, like weird number. It’s actually one of the most accurate ways to measure caloric expenditure during cycling. And there’s a way that you can use that Kew expenditure to map out a very successful nutrition hydration plan. And it’s all based on your power meter that you probably have on your bike. If you wanna use some of the same tools that I and fellow CTS coaches use to analyze and organize training data, I suggest getting the WKO five software. I reached out to Tim Cusick. Who’s a good friend fellow coach, and he’s a w K oh five product leader. He’s been on this podcast several times too. And he provided 20% off code. That’s good until September 1st of this year, 2022, that code is WKO five CTS again, it’s w K number five, CTS, and a link to where you can go and get this, uh, analytical tool for your computer to look at your ride files and your data to help you organize and make better, uh, decisions and strategies when it comes to setting up your own training.

Adam Pulford (29:56):

If, uh, if you’ve enjoyed this series on demystifying metrics and you want to go deeper, I highly suggest getting that, that analytical tool. But if you’re like, no, all this is kind of confusing. I just want to know like the basis or my coach takes care of that, then I don’t suggest that you get it. Okay. You gotta take some time. I mean, the working with data becomes a hobby in itself. I mean, it’s something that I do full time for a job with my athletes. So again, it’s not recommended to all, but I wanted to make that offer to anybody. Who’s really interested about this data and understand what I’ve been talking about on this podcast a little bit more. Okay. So that’s a wrap for today’s show. Um, I’m, I’m trying to make these a little bit more short and concise so that you can, uh, hammer one or two of them out on, on a training ride, or you can get through the podcast while you’re on a, a quick commute.

Adam Pulford (30:46):

So I hope that the shorter episodes are going a little bit better. And also I’ve been solo episode for, for the past three. We’re gonna get some guests back on, so you don’t have to always listen to my voice in your ear all the time. So look forward to some new shows coming up with, uh, different guests and different topics. Uh, but that is it. Thank you again for listening. Don’t forget to, uh, rate and review us on apple podcasts and share it with your friends. Okay. Just spread the word of the train right podcast. Thanks to your listeners. Appreciate y’all.

Leave a Reply

Your email address will not be published. Required fields are marked *