How to Use Data to Make Better Training Decisions, with Tim Cusick

We explore how to use a training philosophy to design your program, then use metrics to guide how much, how often, and how difficult those workouts should be.

Pathways Logo Fast Talk Labs - Basic Performance Data Analysis Pathway Badge
Fast Talk Podcast Q&A
Photo: Polina Rytova

Today we’re taking a good long look at training metrics. We’ve released previous episodes on how to use different numbers, what many of them mean, and how they’re calculated. Today, we tie it together into one package, with a master of data analytics, Tim Cusick, who is not only the product leader for TrainingPeaks’ WKO platform, but also an elite cycling coach of athletes including Amber Neben and Rebecca Rusch.

As Tim likes to say, if each ride you do is a single note, to get the most out of your training, you want to string those rides together in the most elegant way. That is, you want to make music. I love that analogy for the art of training and coaching.

To set the stage for our conversation, it’s helpful to understand that even a data aficionado like Tim fully understands that metrics are not the be-all-end-all—the power of numbers is in their ability to effectively inform the decision-making process. Athletes and coaches should use data to learn more about how best to train, but the data cannot be the solution unto itself.

It’s also helpful to define some terminology. Most of you will have heard of stress, or external load; then there’s strain, the internal load applied to a system; and finally TSS, or training stress score, which we will define and dissect. Likewise, you’re likely familiar with the PMC in TrainingPeaks. The performance management chart shows trends in your season. Finally, Tim often mentions the “content” of the work used to generate these different metrics. What he means by that is the composition of the training rides, whether they’re intense or easy, long or hard, and so forth.

To tie it all together, today’s episode is about utilizing a training philosophy to design the right type of workouts—the content—then using the metrics as a guide to inform how much, how often, and how difficult those rides should be. Voila, you’ve got some Mozart, hopefully. Maybe if you’re Trevor it is more like Celine Dion or Shania Twain. (They’re Canadian.)

On the program today, we also hear from a host of other prominent figures about how they use, or don’t use, all the metrics we have available today. Guests include physiologist Jared Berg, pro mountain biker Payson McElveen, the legend himself Ned Overend, WorldTour veteran Brent Bookwalter, and Xert creator Armando Mastracci.

Time to crunch some numbers, and fill out the all-important comments field.

Let’s make you fast!

References

  • Ebert, T. R., Martin, D. T., Stephens, B., & Withers, R. T. (2006). Power Output During a Professional Men’s Road-Cycling Tour. International Journal of Sports Physiology and Performance, 1(4), 324–335. Retrieved from https://doi.org/10.1123/ijspp.1.4.324
  • Halson, S. L. (2014). Monitoring Training Load to Understand Fatigue in Athletes. Sports Medicine, 44(Suppl 2), 139–147. Retrieved from https://doi.org/10.1007/s40279-014-0253-z
  • Jobson, S. A., Passfield, L., Atkinson, G., Barton, G., & Scarf, P. (2009). The Analysis and Utilization of Cycling Training Data. Sports Medicine, 39(10), 833–844. Retrieved from https://doi.org/10.2165/11317840-000000000-00000
  • Kumyaito, N., Yupapin, P., & Tamee, K. (2018). Planning a sports training program using Adaptive Particle Swarm Optimization with emphasis on physiological constraints. BMC Research Notes, 11(1), 9. Retrieved from https://doi.org/10.1186/s13104-017-3120-9
  • Nimmerichter, A., Eston, R. G., Bachl, N., & Williams, C. (2011). Longitudinal monitoring of power output and heart rate profiles in elite cyclists. Journal of Sports Sciences, 29(8), 831–839. Retrieved from https://doi.org/10.1080/02640414.2011.561869
  • Passfield, L., Hopker, JG., Jobson, S., Friel, D., & Zabala, M. (2016). Knowledge is power: Issues of measuring training and performance in cycling. Journal of Sports Sciences, 35(14), 1–9. Retrieved from https://doi.org/10.1080/02640414.2016.1215504
  • Pind, R., & Mäestu, J. (2018). Monitoring training load: necessity, methods and applications. Acta Kinesiologiae Universitatis Tartuensis, 23, 7–18. Retrieved from https://doi.org/10.12697/akut.2017.23.01
  • Pinot, J., & Grappe, F. (2010). The ‘Power Profile’ for determining the physical capacities of a cyclist. Computer Methods in Biomechanics and Biomedical Engineering, 13(sup1), 103–104. Retrieved from https://doi.org/10.1080/10255842.2010.495580
  • Pinot, J, & Grappe, F. (2011). The Record Power Profile to Assess Performance in Elite Cyclists. International Journal of Sports Medicine, 32(11), 839–844. Retrieved from https://doi.org/10.1055/s-0031-1279773
  • Pinot, Julien, & Grappe, F. (2014). A six-year monitoring case study of a top-10 cycling Grand Tour finisher. Journal of Sports Sciences, 33(9), 907–914. Retrieved from https://doi.org/10.1080/02640414.2014.969296
  • Quod, M. J., Martin, D. T., Martin, J. C., & Laursen, P. B. (2010). The Power Profile Predicts Road Cycling MMP. International Journal of Sports Medicine, 31(06), 397–401. Retrieved from https://doi.org/10.1055/s-0030-1247528
  • Sanders, D., Abt, G., Hesselink, M. K. C., Myers, T., & Akubat, I. (2017). Methods of Monitoring Training Load and Their Relationships to Changes in Fitness and Performance in Competitive Road Cyclists. International Journal of Sports Physiology and Performance, 12(5), 668–675. Retrieved from https://doi.org/10.1123/ijspp.2016-0454
  • Sanders, D., Heijboer, M., Hesselink, M. K. C., Myers, T., & Akubat, I. (2017). Analysing a cycling grand tour: Can we monitor fatigue with intensity or load ratios? Journal of Sports Sciences, 36(12), 1385–1391. Retrieved from https://doi.org/10.1080/02640414.2017.1388669
  • Swart, J., Lamberts, R. P., Derman, W., & Lambert, M. I. (2009). Effects of High-Intensity Training by Heart Rate or Power in Well-Trained Cyclists. Journal of Strength and Conditioning Research, 23(2), 619–625. Retrieved from https://doi.org/10.1519/jsc.0b013e31818cc5f5
  • Taha, T., & Thomas, S. G. (2003). Systems Modelling of the Relationship Between Training and Performance. Sports Medicine, 33(14), 1061–1073. Retrieved from https://doi.org/10.2165/00007256-200333140-00003
  • Wallace, L. K., Slattery, K. M., & Coutts, A. J. (2014). A comparison of methods for quantifying training load: relationships between modelled and actual training responses. European Journal of Applied Physiology, 114(1), 11–20. Retrieved from https://doi.org/10.1007/s00421-013-2745-1
  • Wallace, Lee K, Slattery, K. M., Impellizzeri, F. M., & Coutts, A. J. (2014). Establishing the Criterion Validity and Reliability of Common Methods for Quantifying Training Load. Journal of Strength and Conditioning Research, 28(8), 2330–2337. Retrieved from https://doi.org/10.1519/jsc.0000000000000416

Episode Transcript

Chris Case  00:12

Hello, and welcome to Fast Talk your source for the science of cycling performance. I’m your host, Chris Case. Today we’re taking a good long look at training metrics. We’ve released previous episodes on how to use different numbers, what many of the mean, and how they’re calculated. Today, we tie it together into one package, with a master of data analytics, Tim Cusick, he’s not only the product leader for TrainingPeaks WKO platform, but also an elite cycling coach of athletes including Amber Neven and Rebecca Rusch. As Tim likes to say, “if each ride you do is a single note to get the most out of your training, you want to string those rides together in the most elegant way, that is you want to make music.” I love that analogy for the art of training and coaching, something we talk a bit about today.

 

Chris Case  01:09

This episode of Fast Talk is brought to you by WHOOP. WHOOP is a fitness wearable that provides personalized insights on the performance of your sleep, how recovered your body is, and how much stress you put on your body throughout the day, from your workouts, and the normal stressors of life. Today’s episode is all about training load, Trevor, what about recovery?

 

Trevor Connor  01:28

Yeah, we’re gonna hear a lot from Tim today about training load, which the software is good at measuring, we’re going to go pretty deep into it. There’s a lot of metrics that we can use to figure out how hard you are going, but even in getting ready for this show, the closest thing they have to measuring recovery in WKO and TrainingPeaks is something called TSP, and Tim just said, “No, don’t really use that.” It is hard to measure recovery, and that is what are the benefits one the things I really like about WHOOP, it’s not just another tool to measure how hard you’re going, it is really a tool to measure that balance between recovery and stress, which is really important because, a bet you’re getting really get into and this show is more stress is not always better. Try to get that training load up as high as possible is not always better. The best training is in that balance, that nice range of hitting yourself hard and then recovery, and this is what the WHOOP strap is all about, making sure that recovery matches up with the training.

 

Terminology Used to Understand Data

Chris Case  02:35

To set the stage for our conversation, it’s helpful to understand that even a data aficionado like Tim, fully understands that metrics are not the be all end all. The power of numbers is in their ability to effectively inform the decision making process, athletes and coaches should use data to learn more about how best to train, but the data cannot be the solution unto itself. It’s also helpful to define some terminology here. Most of you will have heard of stress or external load. Then there’s strain, the internal load applied to a system. And finally, TSS something many of you, I’m sure are familiar with thats Training Stress Score, which we will define specifically and dissect. Likewise, you’re likely familiar with the PMC in TrainingPeaks, the Performance Management Chart shows trends in your season. Finally, Tim often mentions the, “content of the work used to generate these different data points,” what he means by that is the composition of the training runs, whether they’re intense or easy, long or hard, and so forth. To tie it all together, today’s episode is about utilizing a training philosophy, to tie it all together, today’s episode is about utilizing a training philosophy to design the right type of workout, the content, then using the metrics as a guide to inform how much, how often, and how difficult those rides should be. Voila, you’ve got some Mozart, hopefully, maybe if you’re Trevor it’s more like Celine Dion or Shania Twain. On the program today, we also hear from a host of other prominent figures about how they use or don’t use all the metrics we have available today. Guests include physiologist Jared Berg, pro mountain biker Payson McElveen, the legend himself Ned Overend, World Tour veteran Brent Bookwalter, and Xert creator Armando Mastracci. All that and much more on today’s great episode. Time to crunch some numbers, and fill out the all important comments field. Let’s make you fast.

 

Chris Case  05:02

Well, welcome back to the show, Tim, It’s been since way back, Episode 44, where we had you on Fast Talk. Thanks for joining us once again.

 

Tim Cusick  05:12

No worries. Thanks for having me always appreciate being on the show.

 

Training Load

Chris Case  05:15

Last time we spoke quite a bit about data, let me read the title of that episode, “The Data Revolution: How AI and Machine Learning will make you Faster.” That’s a that’s a pretty good title, if I do say so myself. Today, we’re also going to talk a lot about data, math, and terminology, and things that a lot of people have heard before most likely, but might not really understand how they’re formulated, how to use them the best. So we want to talk about both of those things today. It’s probably helpful to start out with a few definitions. Tim, why don’t you help us set the stage here for our discussion by defining some terms. Let’s start with training load, simple one, or so it seems.

 

Tim Cusick  06:02

Yeah, so it seems right. Well, thanks for starting out easy on me. When I think about training load as a definition, I break it into two quick areas, one kind of the singular, and I think that’s the way a lot of people first interpret it, and that’s really, training load is a quantitative measurement of the strenuousness of a workout or exercise. So to make that simple, you go out for a workout, How hard was that workout? Right? And that’s a relationship of intensity and duration, How hard did I go? How long did I go? But I also started talking about training load in the area where it’s really important to measure and manage training loads is more in kind of the plural format and the acute and chronic view or the view over time. When you look at training load over time, you’re really just talking about the similar definition, except it really is that quantitative measurement of the accumulation of training load, of training strenuousness, right? Or maybe the lack of, depends what you’re doing or not doing. But really, you’re talking about it a singular level, how strenuous was the workout? When you’re talking about the plural or over time, you’re talking about the accumulation of that strenuousness over time.

 

Trevor Connor  07:15

So it’s looking at all the rides, or all your workouts over time to get a bigger picture of everything meaning, If you do one absolutely killer ride, that’s all intensity for six hours, you’re gonna be dragging your feet at the end of it, your loads can be really big, but if you’re doing that, and then all your other rides are just cruising and noodling along, don’t expect to have a high fitness level. Where if you spread that load out over time, have some hard workouts, have some easier workouts, the accumulative load is actually going to be higher, without that killer, one killer ride, you’re gonna end up in a better place. That’s it. What you’re saying is, you can’t really just focus on the load of one ride, you have to see how it all fits together.

 

Tim Cusick  08:04

I think when you really talk about training load management, one ride gives you a metric of measurement, a score or something, right? However, it depends on how you track it, the ability to put them together over time, really, that’s how we understand training strategy, training plan, training execution. Opens up the doors to tracking adaptation to that training, I mean, it’s a little more complex for that, I have a feeling we’re going to get into it, but the reality is that, you know, you have a singular workout creates a training load that one day is the load, how they all come together and accumulate over time, that’s really where the magic happens.

 

Trevor Connor  08:40

Yeah, I think that’s a really important point, because a question we’ve gotten a bunch on the show and I think we address somewhat recently is the, Why would you ever do an easy ride? Now think about those athletes to get on their bike to go for an hour they get, so we’re going to get a little later in the TSS, but they see a TSS of 30 and go “Well, that was a waste of my time, why would I do that?” That’s the answer is yes, if you look at just the individual workout, that wasn’t much of a training load, but you have to look at it accumulated over time, that’s how it all fits together.

 

A Workout is Like Making Music

Tim Cusick  09:11

In that basis, right? Here’s a way I like to explain it to new coaches and stuff like that, a workout is like making music, a workout is a note, you play a single note, right? And you can easily figure out you can write a note, anybody can play a note on a piano can play one to make it work just fine, but really, to train you got to be able to take those notes and put them in the right order, right? In that right order, somehow it means something it becomes a song, it becomes music. Training is that it, really is you have to take these individual notes, one workout, one workout, one workout, understand its training load, plus some other things content, and what’s in it, what’s its purpose, but a good coach understanding training load, puts those notes in the right order some high notes, some low notes, right? And puts them together in a way that it makes really great music, that’s the magic. One of the things, you and I kind of have discussed this a little bit that where we see right now when you start talking about training load, people over focus on that single note, and they miss the ability to make music, just like you were kind of hinting at there. My education content, what I’ve been trying to work with coaches and stuff now, is focused on making music, take those notes, understand how to put them in the right order, that’s the measurement you want to be using, you know, that’s the way you should use training load as a measurement, if you can put them in the right order, man, that’s how you make magic happen. No one note makes magic.

 

Trevor Connor  10:36

I love that analogy. That’s a really good one, my only issue with it is I’m a horrible singer, If I just sang one note, you couldn’t prove that I was out of tune.

 

Chris Case  10:44

You’re no Mozart, is that what you’re saying, Trevor?

 

Trevor Connor  10:46

I don’t think he’s saying very much.

 

Chris Case  10:48

Well, yes. But he made good music.

 

Trevor Connor  10:53

We spoke with the head physiologist at CU sports, Jared Berg, about the new metrics popping up in the various training software packages to get his opinion, from a physiological perspective.

 

Jared Berg: New Metrics in Various Training Software Packages

Jared Berg  11:05

Doing that type of data analysis is not something that I’m doing consistently, and doing a lot of, you know, looking at power metrics and you know, something like work prime or help you know how much basically, how much do you have in the tank, you know, when you’re above threshold? Is kind of a my standing work prime, but I think it’s certainly a tool, right? And it’s something that provides some valuable information, to me, it’s a little more useful, some that we wouldn’t be able to see in a physiological test is, you know, how good are you at staying, you know, up and over threshold? And so if you can get some baseline measures, and then you can watch that over time, I think that’s really useful.

 

Trevor Connor  11:42

So let’s flip this around, are there any new metrics that you’re seeing pop up in the training software, that you just say you’re hurting yourself using that you really shouldn’t be looking at that those numbers?

 

Jared Berg  11:54

So I mean, estimates of Vo2 Max, I mean, to me, that’s almost as good as a Vo2 max test, I just don’t because there’s not a lot of use for that, Right? You know, so that’s that’s fine, neither to me I really applicable and can sport in performance and research, they all know the Vo2 Max is useful like I said before, but not so much in trying to predict performance.

 

Trevor Connor  12:18

I know more and more athletes are training by TSS like Target in particular, TSS per day, target a particular TSS per week, so that’s the Training Stress Score from TrainingPeaks, but there’s different versions of it from different software. What’s your feeling about that? I know we’ve had this conversation before the one issue is it doesn’t really say anything about where is that stress coming from?

 

Jared Berg  12:43

That’s definitely a good, good debate, good question. I mean, I like anytime you’re, you know, getting things that are quantifying the amount of work and trying to trying to make inferences, how much stress that’s causing you, I think that’s that’s useful and if you’re you know, if If you’re riding longer at higher high rates that should equal more, more stress, right? And then if you’re doing that continuously on a day to day basis, I mean to be able to track that and see that I think that’s useful information, especially if you are tying that, you know those particular heart rates or workloads in with a physiological test, and that will make it even more accurate. I kind of like I actually kind of like that, I do. Yeah.

 

Difference Between Stress and Strain

Chris Case  13:32

Let’s be specific about the terms we’re using here. Tim, can you give us an explanation of the difference between stress and strain?

 

Tim Cusick  13:39

This was popularized early, particularly in power training, by you know, back in the days with Dean Golich and Allen Lim, and those guys, they have this theory of stress plus strain equals adaptation. So stress is a quantitative measurement of the external, so power is a stress right, you are peddling, you’re applying force or torque to a crank and you’re creating power, that’s a stress measurement, you’re actually measuring an external factor. When you apply stress to a system, it undergoes strain. That’s what’s happening internally under the stress, so I’m doing 300 watts climbing this hill, that’s the stress measurement at that individual second, how bad I feel, how many fibers I haven’t get how much energy I’m using to do that what’s happening internally within the system, that’s the strain. So we apply stress to a system, it undergoes strain, and once we recover, rest, allow it to recover, then we adapt to that strain. So it’s kind of it’s a chain event, we apply stress, external load, the system goes under strain, and we hope for an expected adaptation on that.

 

Chris Case  14:55

One issue that might confuse people as they hear power meter, they think strain gauge, and so they equate external measurement with strain, but that could lead people astray.

 

Trevor Connor  15:10

But it is an important point, because this is something that I’ve said this on the show before, this is I know something you have brought up, But when you are measuring power, you’re not measuring what’s going on in the body, you’re measuring what’s going on in the bike. So the example I always give if you have two athletes climbing the hill beside one another, and they’re the same weight, they might both be putting out, say, 300 watts, you have no idea how hard each of those athletes are working, if one’s a pro and the other one’s a brand new cyclist, the one might be at an absolute physiological limit at about the crack, where the other one’s going, “Yeah, I can do another hour or two like this.”

 

Chris Case  15:53

Mm hmm.

 

Trevor Connor  15:55

And you don’t know, because it’s an external measure, isn’t showing you what’s going on physiologically. So let’s jump there quickly before we get to the specifics of cycling, what are some of the more common internal measures? And what are the some of the more common external measures?

 

Common Internal and External Measures in Training

Tim Cusick  16:17

So when we start talking about external measures power is king, Right? So let’s just start there. Yeah, people ask me all the time about, you know, training, “how can I train better? I want to use my, my heart rate monitor to get faster, I’m going to use my speed sensor get faster.” Here’s my answer, right? invest in the power meter. It really is that simple because it is the most accurate measurement of external stress being applied, right, and what you’re doing out there, everything else is secondary to that. But the reality is look that not everybody can afford or wants to buy a power meter, and that’s true, So what are other external factors? Well, you’re kind of limited to a specific external, but you you can begin to talk about things like duration, speed, how fast you get from point A to B and this and that, but externally speaking, you don’t have a lot of other choices really, power is the core way to do it. I think for a while there, when we start talking about strain, they’re really talking about heart rate, right? So one of the ways you can always different powers of stress for heart rate is a strain event, meaning it’s showing the reaction to the system going under stress. We are seeing some new metrics coming out on heart rate, some utilizations of heart rate data, some HRV, some estimated HRVs, things like that, that are having some impact and are learning. Some ability to put the external power together with internal strain, heart rate, and HRV, I think we’re seeing beginning to get some traction and are probably going to be some of the more interesting metrics of the future, short term future. But I think you know, it’s one of these things right? I can answer that question with a whole bunch of different things, but If the idea of your training, if you’re listening to this podcast, you’re probably trying to get faster, the two things you need stress, you need a power meter, strain you need a modern heart rate monitoring system, which will give you something beyond just beats per minute. So there you’re talking about availability of HRV, in different formats of HRV. Those when you’re really talking about on the bike training are the two elements, and you see a lot of debate about this, and I, you know, and I’m just being pointed to save everybody some time, you’re eventually going to buy the power meter. If you keep training, you’re going to get there and make that investment, you might as well just start there and start your learning curve right now, and matching it with a quality modern heart rate, that combination of stress and strain measurement, really give you all the tools you need for success.

 

Trevor Connor  18:47

It actually is quite amazing how ubiquitous power meters have become. I remember when I started out my career, you’d heard of them, nobody had one, and then there there was a point where you bought it, but you had to be a pretty serious cyclist to have one. Now you look at things like how many people are on Zwift? Well to be on Zwift, you have to have a power meter, or have a built into your trainer, and it’s really, we’re getting to that point where you rarely see a cyclist who does not have power. The other common measure, is probably the original common measure for that internal load for internal stress, strain, whatever you want to call it would be RP.

 

Tim Cusick  19:30

Absolutely, and still should be better utilized today. I think that’s a great point. I mean, we tend to think data and we forget the subjective data venues. You know, I think one of the more interesting things you’re seeing in the evolution of software and area, you know, and online programs and software, obviously my area, the ability to capture the subjective data and apply it to the objective data to create uniquely blended analytics, I can tell you absolutely, that’s a very powerful technique and for people to invest in collecting that data from their athletes or from themselves, and then learning how to utilize it is a dramatic improvement in your ability to understand training load.

 

Trevor Connor  20:14

But I would say we’ve been talking about this a bit on the show lately, but a theme that we have been seeing is, the more advanced the athlete, the higher level the athlete, the more important you actually see that simple metric of rate of perceived exertion becomes. Meaning top level cyclists really have this innate ability to understand exactly how hard a workout should feel, they can feel those sensations of their body and know exactly where they should be at, where a less experienced athlete is going to have to rely more on numbers because they don’t have the feel yet.

 

Chris Case  20:52

And that comment section becomes one of the most critical components here of understanding where somebody is at in their training, and you’ve mentioned that many times before on the show, Trevor. Tim, what are your thoughts?

 

Quantitative Approach Towards Training: Athletes Should be Writing Notes

Tim Cusick  21:06

I couldn’t agree more. I require, literally require, all my athletes to write notes, in every workout, because in the absence of that subjective data, and I always tell them, I have one rule, “You can’t write your power numbers or your heart rate in the notes,” because I can read that. I’ve got the objective data. I don’t need your opinion there. I want to know, how did you feel, and then if you had a feeling or something was going on, that you needed to tell me about, go deeper, don’t just say I was tired, what was tired, what was going wrong? Don’t just say I felt great, what felt great? I think the more the athlete is able to communicate that level of understanding and perceived exertion and their feelings during the effort, the coach gets much better. It really improves the ability to apply that subjective, that color commentary to the quantitative approach torwards training. And you know, I think it’s a super important tool for people to be using.

 

Trevor Connor  22:00

We’ll have to get you back on the show the near future, because we have an episode plan that’s tentatively titled, “The Most Important Field in your Training Software is the Notes Section.”

 

Tim Cusick  22:10

I teach my athletes how to write notes.

 

Trevor Connor  22:13

I had an athlete who I coached last year who first refused to write notes, Yeah, whatever I’d asked you to write notes is like, “well you have all my data, what else do you want?” So then i’m like, “I have to get the notes.” So I think the longest note I ever got from him was nine words.

 

Trevor Connor  22:36

Well, I had to start getting on the phone, whenever I’d get him on the phone, and I was like, he talked to me for about half a minute, He’s like, “Okay, I gotta go now.”

 

Chris Case  22:44

Wow.

 

Trevor Connor  22:45

And it was actually really hard to coach him, because even though I had, he was huge, in collecting all the data, I discovered that without the context, it was actually really hard to interpret the data.

 

Chris Case  22:58

Well, it’d be kind of like putting the notes symbol, but not putting it on the, Oh no, here’s my lack of music terminology. What’s the thing? There’s like the G clef, and all the lines, Every Good Boy Deserves Fudge, and you put the note somewhere on that, right? But if you just have the note with none of those lines, you don’t know what it is.

 

Trevor Connor  23:20

That’s a good point. Consider it an analogy. I like it.

 

Chris Case  23:23

Did I just come up with that?

 

Tim Cusick  23:26

Making music.

 

Chris Case  23:27

Making music.

 

Tim Cusick  23:28

It is really important, you guys, I mean, just as a sidebar, Bravo, Kudos to you guys for doing that, because the data thing gets everybody sucked into just quantitative, quantitative, quantitative, and they missed the whole color commentary.

 

Trevor Connor  23:43

Yeah, no, it’s both. I’m still trying remember who it was, but I was talking to a coach who said, I could always detect overtraining in their notes before I could detect it in any data.

 

Trevor Connor  24:00

Cycling legend Ned Overend, really needs no introduction. Obviously he knows how to train very effectively, but for Ned feel is the biggest metric. He’s just had several decades to learn the feel, when you train, what do you listen to? Or what do you look for to know, Okay, I’m going to hard I’m going to easy I’m going the right pace.

 

Ned Overend: Feel is the Biggest Metric

Ned Overend  24:21

I can plan to go on a hard day, and I know the segments right on Strava, So I, I’ll pace myself for those certain segments to try and get a fast time on. When I go out and do that, it’s not set in stone, because it’s really just kind of a feeling I have in my legs I’ll warm up, I’ll start doing the segments and my legs will feel heavy, and I won’t be able to turn over the gear that I’m looking for, and I know the speeds and just the feeling my body has when it’s rested. Then I’ll skip doing intervals, do a recovery ride, and wait to do intervals on a day when I’m when I’m better recovered.

 

Trevor Connor  25:08

Are there particular fields in your body that you say, okay, that’s a dangerous sign or that’s a red flag.

 

Ned Overend  25:14

It’s a fatigue that I’m feeling in my legs, and I’m not gonna call it necessarily a burning, because it’s almost, because I don’t get to the point where my legs are burning, It’s where I’m attempting to put in an effort, And, you know, I feel like my respiration increases, right? I’m breathing harder, and I can’t actually get to that point where I can make my legs burn, sometimes your body has to wake up, right? So if you’re just starting one effort, and you don’t, you’re feeling fatigued, sometimes you need more of an opener before you can get to the point, you know, where you can do a quality interval.

 

Chris Case  25:54

So when you’re out, It sounds like you know you you don’t use heart rate, you don’t use power to give yourself cues as to how your effort compares to previous efforts or what you’re trying to target, but you are using Strava, which is numbers. I’m curious if you use the live segments?

 

Ned Overend  26:16

It’s after the fact, I haven’t been using live segments.

 

Trevor Connor  26:20

So that begs the question when you’re doing some sort of structured work, like you’re doing, let’s say, threshold on some of the climbs, how do you pace yourself? Do you just know about how hard you want to go? Is it just years of learning the pace?

 

Ned Overend  26:37

Yeah, it’s, it’s from knowing the segments, and then basically feeling in my legs, not going too hard, until I can know that the end of the segment is coming. So it’s kind of just pace learned over you know, 20-30 years of training.

 

Trevor Connor  26:59

Sounds like you just have this innate sense of,

 

Ned Overend  27:03

Yeah, I would say it’s trial and error, you know, the feel in my leg and the building up of fatigue as far as how long can you last? over this specific climb, you can bury yourself a little harder, if you know that the climb is going to flatten out and you’ll get a little bit of recovery. It’s specific to knowing that climb and not so much just based on time.

 

Trevor Connor  27:25

So let’s flip this around and ask it a slightly different way. Do you think there is a danger with some of these younger athletes who have all these numbers? If they’re going out and doing their workouts and time trialing, staring at wattage, staring at heart rate, Do you think there’s a danger in not learning the feel and affecting their performance?

 

Is There a Danger in Not Learning the Feel, and Focusing On the Numbers?

Ned Overend  27:48

In the same way, that when I’m doing a group ride, it pushes me harder than when I’m doing intervals myself or it pushes me in a different way, and I think that when you’re in a race, if somebody, especially in a race, if somebody is looking at a watt meter or a heart rate monitor, they’re restricting themselves when they may be able to push themselves beyond kind of the numbers they’ve seen in training, to whether it’s make a break or or stay away from somebody, or I think they if they put those kind of parameters on themselves, that that it may hinder the performance or capable.

 

Chris Case  28:35

The question, why measure training load?

 

Why Measure Training Load?

Tim Cusick  28:39

Well, personally, I think both the acute and tronic training, acute and chronic training, load measurement is crucial to all your successful training strategies. It’s something that has to be a part of it, remember, I’m talking about measuring training load at the moment, I know we’re going to get deeper into some definitions. But to answer that question, right for me, I always want to start with one overarching goal, as we brought more and more data, to endurance coaching to being an endurance athlete, you got to understand what is the role of data? What is the role of data science, right? You look at a bunch of data and that now you’re a data scientist, because you got power, you got heart rate, you got all these things, so now you’re a data scientist. People think that all this data and the data science we apply is meant to give us some definitive answer, “oh, go train for 56 minutes, at 282 watts,” right? That’s not what it really is, data science is decision science. So you’re collecting all this data so that you the person making the decision, has more knowledge, and you improve the odds of the success of the decisions you make. In this particular case, we’re talking about athletes success, their ability to achieve their goal, you know, to be on peak form when they want to be on peak form to to win they’re big race, whatever that might be. But you’re really talking about data, all this measurement of training load and other factors that are involved in that. It’s not a magic answer, there’s not one, I opened a can of data and what popped out like, you know, spring-born snakes, and magic answers of what I should do. What happens when you look at all the data, it makes you more knowledgeable, and it allows you to make better decisions which increase the odds of success. You know, when you apply exercise stimuli to an athlete, the response, physiology, isn’t this neat, linear thing, right? Responses can be different, sometimes predictable, sometimes not, right? So all you’re doing is using all this data to make better decisions to improve the odds of success. So therefore, when you’re talking about training load, remember my definitions had measurement and quantitative within there, you’re talking about the ability to use this measurment of training load to improve your odds of success. So that’s always my overarching thing with all data, something that all data should be utilized. When you start talking about training load, specifically, you really are talking about the measurement of training stress, and the prediction of the resulting strain and adaptation. So, really, when you think about the most quantitative training load metrics or measurements that are out there, they are a stress measurement because we can best and most easily understand that stress, I have a power meter, I’m doing 300 watts, I get it, I get that 300 watts, that’s the stress and putting a hard number, that’s very trackable. When you get to strain, we don’t have a great system of measurement of strain right now, it isn’t out there. With heart rate, you have HRV you have some other factors, but no matter what, you’re still somewhat in the dark of the true strain the athlete is going under, we’re getting better at that we’re rounding out information and data there more and more.

 

Tim Cusick  32:11

The trouble that brings with training load, the challenge, and where the art of coaching using data to make better decisions is important, is we can measure the stress pretty quantitatively, like there’s an objective set of data. We’re starting to get our hands around strain a little bit better, but reality is we don’t have like that hard quantitative piece of data, therefore, if stress plus strain equals adaptation, we really can’t nail adaptation. So when you start thinking about training load, we’re really trying to improve the odds by understanding what we’re applying, we’re applying a certain amount of stress, that stress is going to have some relationship to fitness, some relationship to performance specificity, some relationship to how the athlete performs in a given event, right? But we don’t have that same hard measurement for strain, and we really, and when you think about the adaptation we’re best guess, aren’t we? We don’t have that way of predicting the adaptation. Now the athlete goes to their event and does well, you said, you know, your thumbs up, you did a great job of measuring stress and strain, but if they fail, you have no diagnostic because you apply to stress that you thought would result in a strain, but yet the adaptation wasn’t there. So what did you do wrong, or wrong as harsh word, right? Where were areas for improvement? What changes should you have done before you created that scenario? So if you think about that, in the sense of training load, it really is a stress load number that is the hard quantitative starting point. And then really closing on that that gives you the real drill down, so some of the key physiological principles that you should be looking at in training, progression, overload, recovery, you know, and super compensation within, thats the idea of, you know, we get stronger when we recover. And then eventually, you know, you using those I applied those metrics, well, I’ve got the right training load, I’ve done some of the right content and what’s in there, you’re hoping for adaptation. It’s really important to say this, right? We have stress score, we’re gaining in some strain numbers, and there’s been some progress there. What we really, really want to focus on in the future as we see data metrics playing a bigger role is adaptation scoring, adaptation understanding, and how we each individually adapt. That’s when you’ll see the next break trip through that will replace stress scoring, when you get adaptation. I don’t think you’ll ever actually get it in strength, It’ll be the jump over strain and right to adaptation scoring that will be the next frontier.

 

The Art of Coaching: The Data Doesn’t Make Decisions, The Coach Does

Trevor Connor  34:48

So an example using the software, at the beginning of every week with my athletes, we back out the week. So we’ll figure out what the workouts are, I’ll look at everything, I’ll look at the TSS, where they end up with their their CTL at the end of the week, and I will definitely use that as a guide to map out the week. But that second decision is, all my athletes have learned, call me. It’s quite frequent on Wednesday or Thursday, I get a call and they go, I was struggling with my interval session, or I didn’t sleep well last night, and I seem to be dragging my feet, whatever, and then we adjust the plan, and that’s what you can’t see on the numbers.

 

Tim Cusick  35:30

The art of coaching.

 

Chris Case  35:32

There it is.

 

Trevor Connor  35:33

Yep.

 

Tim Cusick  35:34

And with data, you’re better informed, right? You’re looking at the history, you understand where they’re at a point in training load or training content and this and that, in response, you have some numbers, but at the end of the day, you’re talking one to one to that human being and making a human decision. You’re smarter, you’re wiser, you’re better, at the decision making process, because you have data, but the data doesn’t make the decision the coach does.

 

Trevor Connor  35:55

The data really helps. So I had an example two weeks ago, where I had an athlete who was starting to overtrain, and I told him it was coming. And I said, “yo, we need to take a rest.” He’s like, “but I’ve gone so well right now.” So we finally hit this, like a Thursday, he did an interval workout, and he got through the first one, And then the second one he did two minutes and it stopped, waited a while then did it again, got through another two minutes then stopped, then forced out the full interval, and then got through a third one and quit. And I didn’t hear anything from them, so I just sent them an email going, “Are you going to tell me how bad yesterday felt?”

 

Chris Case  36:39

Yeah, yep. Predictable, right? So now that we’ve explained why you want to measure training load, it’s important, what should the measurement actually do here?

 

What Should Training Load Actually Measure?

Tim Cusick  36:50

Great question. So I think really, when we’re talking about measuring training load, and there’s different ways to do it, I know we’re going to get into some of them here today. One It needs to be quantitative, so when we look at training load, we aren’t having, you know, an impact bias of different factors, so we talked about training, what impacts training? Well, you have intensity, right? how hard you’re riding every day or how hard you’re not riding, you have duration, length of time, and you actually have some other factors, you know, going into that as far as specificity, whenever you say specificity, we tend to think about, am I climbing? Am I time trialing all night learning something? Really, you can talk about scales and other factors that will fall into training at times, which you might be on a bike doing something but have different impacts on training. The training load has to take all that and do what I like to call normalize the number, meaning give us something that gives us a quantitative number that allows us to look at the singular workout and understand it under a fair quantitative process, normalize it, and then be able to take those series of one day numbers, right? And put them into some factor of accumulating them, and looking at that over time, a single workout in itself is, you know, you’re never going to perform great after a single workout, you could do one workout in a vacuum and you get some fatigue, and eventually you’ll get a little more fitness, but the reality is, it’s the ability to take that normalized quantitative measurement of training load over time, utilize that, you know, in the right rhythm, you make your music, you build a certain amount of related fitness and performance capabilities, and then you use that quantitative measurement to predict or maybe even to periodized, I don’t like using the word periodized, because periodized is a it’s a great term, I think periodization works well for a lot of athletes because it puts things into perspective, we tend to think about training mode in that relationship to periodization. So, when we talk about what should training load do, it should allow us to track in quantitative normal fashion, we tend to think periodized, when we think that our training load goes up, up, up, up, up, as we go through a periodized plan, and then at some point, we back off that training load, again, using that quantitative number into some form of a taper and performance. That, you know, I blended the idea the answer of a little bit what it’s meant to do, but the reality that’s usually the easiest way for people to visualize it as some type of periodized plan.

 

Trevor Connor  39:33

So there’s a word you brought up that I think is a really important word, which is normalize, because I’ve got to go out on a limb here but if there is one thing that has made this whole load measurement difficult, it is the stochastic nature of cycling. So I’ve got a bit to this, Chris and I were talking before we we got on the phone with you, and here I am the heart rate guy, apparently that’s a reputation I’ve got, I pointed out to Chris, if you go and do a criterium for an hour, and you go out and do a time trial for an hour, you look at your heart rate profile, and they’re actually gonna be somewhat similar. They’re both gonna be pretty much right up at threshold, but nobody is going to say those two races are having the same effect on your body. So it’s really hard to identify the true effect of workouts have in your body, certainly just averaging power isn’t going to do it, I know that you use a normalized power, which tries to get at some of the physiological effect. Other other ways, which I’m sure you get to talk more about is this putting power into an ordinal format, so it’s basically put it into zones, some sort of zones and saying, okay, you spent this much time at this zone this much time at that zone, that much time at that zone and at having a multiplication factor that helps give you a better sense of how hard was that workout? Where were you spending your time? What sort of effect did it have on you?

 

Tim Cusick  41:13

That’s why I think stress measurements are the best quantitative direction to go, because they’re, they’re least susceptible to noise. Meaning Look, there’s going to be no perfect metric for some of the reasons which you just very clearly brought up and very accurately, so right, because we, we really would love to know the adaptation score, we’d really know, we really want to know what’s happening downstream but we can’t right? Too many variables, If you think about data, too many variables in the strain score, the one place where we have the ability to limit the variables to some degree is in stress scoring, and in scoring the external load that we’re putting on something. So this is where we probably back to why it’s Training Stress Score, and not training strain scoring, I know we’re gonna probably get into that a little bit later. But, you know, obviously, TrainingPeaks and WKO, we use Training Stress Score which was created by Dr. Andrew Coggan, It is based on normalized power, but the premise was by using stress that external measurement, we had the most normalized number, meaning if we were going to be looking at a load measurement, we had the one that would have the least amount of bias, the least amount of noise, and it was only ever meant to be a measurement of external load. The reason I say normalized is what’s happened is as we look at Training Stress Score, people want to use it as a training strain score or a training adaptation score all the time, and that leads into the factor of why are they you know, two workouts being different, is Training Stress Score or you know the resulting metrics directly related to my fitness? The Training Stress Score or the measurement of stress is the best quantitative measurement of one of those three areas, stress strain adaptation that we’re applying to the system, it gives us the most normalized output of data with the least amount of noise.

 

Trevor Connor  43:15

That’d be worth mentioning, so what exactly is normalized power? And I admit, I used to not be a fan at all, but I have certainly come around what I learned more about how Dr. Coggan came up with it, which was an attempt to correlate it to physiology. So I’m looking right here at his definition and it starts with a 30 second smoothing, which is based on Vo2 and heart rate response that it’s raised to the fourth power which is based on a regression of blood lactate concentrations. So it is an attempt to try to match up stress and strain.

 

What is Normalized Power?

Tim Cusick  43:55

Correct, and that’s exactly why right? When Andy had set out to come up with some way of quantifying this, you bump up against that problem we already had work, which is average power, right? So that we had kilojoules, the ability to measure kilojoules, but that didn’t deal with variability very well. So normalized power was an outgrowth of power, because we needed some better understanding to understand the physiological strain of the variability of power, the stochastic nature of it, we then applied physiological thinking, physiology, right? And look, there’s no perfect answer here, It wasn’t like we wanted to get a score, there was some direction of a perfect score, the goal was to give us something that was normalized that did a very good job, it was based in in physiology, it represented, you know, external stress, understanding some of the impact of strain and it gave us a good basic tracking number that was applicable to looking at different workouts, and determining the strenuousness of them, because it’s based on normalized power, not average power.

 

Trevor Connor  45:09

So here’s the question I have for you, and I actually spent some time looking for this before this podcast, It is based on physiological concepts. Has there been any validation to say, yes, this normalized power is a bridge between stress and strain, between what’s going on inside the body and what’s going into the bike?

 

Tim Cusick  45:31

No, not specifically. I mean, there has there’s been some studies that have come back and you know, like anything that people have looked at, and suggested some relationships but to say a hard study that really sits here and says normalized power is a physiological fact, there’s none that I’m aware of.

 

Chris Case  45:51

So we’ve thrown around the the terminology TSS, that acronym people have, most likely a lot of people out there have heard that term, and we’ve defined it in relative terms in other episodes, But Tim, I’d like to get your perspective on what exactly is TSS?

 

What is a Training Stress Score?

Tim Cusick  46:11

Well, Training Stress Score, TSS, is an external measurement of the stress of an effort or a workout. It is based on your individual FTP, so it is something where it’s super important, you’re doing a proper FTP test, your Functional Threshold Power Test, so that you know that because the score is normalized to what you can do. The score is based on a combination of ride duration, normalized power, and intensity factor. And the reality is, in its most simplest definition, right? If you ride as hard as you can for an hour, that’s 100 points, and that makes it pretty easy, goes back to the point of normalized, if I was to go out and do a 40k time trial and let’s say it takes me exactly 60 minutes, that’s 100 TSS score. That’s a way to quantify the strenuousness of that.

 

Trevor Connor  47:07

So I get to throw this out here, because I have added a lot of our listeners have never seen this, the actual formula here is TSS equals, in brackets, open brackets, times, times normalized power, times intensity factor, closed brackets, divided by open bracket, FTP times 3600, close bracket, and then all of that times 100. So a couple of important things to point out there is it’s based on normalized power, It’s based on the intensity, how hard you are going, but I think it’s also really important to look at that denominator, which is FTP times 3600. There’s 3600 seconds in an hour, so It is the one thing that’s really important to this is your one hour power.

 

Tim Cusick  48:05

Absolutely correct. And it is on the basis of an hour, that has led to some confusion about FTP, but the basis of TSS is one hour of power.

 

Trevor Connor  48:14

So I actually found a study, 2017 study, that I just read, it talks more about TSS but as we’re going to get to TSS is based on normalized power. It starts by saying, look, there’s actually never been any sort of study to see if TSS has a physiological basis, so they use 15 professional cyclists, they’ve tracked them for 10 weeks and get this as you probably read this study.

 

Tim Cusick  48:43

I know the study, yeah.

 

Trevor Connor  48:44

The results of the study looked at all these different measures of load, the one that seemed to correlate the best was iTRIMP, which is based on heart rate and lactate, or matching up heart rate to lactate, but second best and not too far off was TSS.

 

Chris Case  49:01

This episode of Fast Talk is brought to you by WHOOP.

 

Trevor Connor  49:04

WHOOP is offering 50% off with the code “Fast Talk,” go to WHOOP.com, and enter “Fast Talk” at checkout to save 15%. Sleep better, recover faster, and train smarter. Optimize your performance with WHOOP.

 

Chris Case  49:27

Let’s dig a little deeper on TSS, there are a lot of advantages to using TSS there are some negatives to using TSS, maybe those are the advantages, negatives. Let’s not qualify them as such, what are you see Tim, as the pros, first of TSS?

 

Tim Cusik: Pros and Cons of a Training Stress Score

Tim Cusick  49:50

One of the greatest features of TSS is, it’s easy, It’s functional. When you look at a lot of the metrics that are created through the WKR software, that functional thing comes up a lot, right? And the reality is, I know it was Andy’s vision to be able to give people this ability to quantify, not only training load, but training content in different ways, right? Through the use of a power meter, which then is also trickled down into heartrate, and some other metrics at this point, so that it’s functional, that so anybody with a power meter or you know, a device, a heart rate monitor, can use this information, but yet it’s based on physiology. The goal was, to some degree, at that given point, when power meter started getting popular in the early 2000s was, we had labs and you can go in and you could do Vo2 Max Testing and Lactate Testing and get a lot of stuff, but they were expensive and hard to access. We wanted to be able to take this new power data that was really kind of coming to the surface and use it for more. So Training Stress Score was one of the first thing developed, in using power data to create metrics. So it’s easy functionability really is one of its biggest advantages. You have a power meter, you have a heart rate monitor, you can have TSS, you can create that score you can begin to normalize or better understand the strenuousness of your workouts.

 

Tim Cusick  51:16

Second, it feeds into some form of a load management system. Now in TrainingPeaks that load management system, is called the Performance Management Chart, or the PMC, as it’s known. But why that’s important to note is, since you have an easy and functional system to look at the daily scoring right, the daily strenuousness of system of strenuousness of the workout, you also have the ability to then take those daily scores and put it into something bigger, that gives you a greater understanding of training load over time, which is the Performance Management Chart. It is easy to utilize and predict, I think that a lot of people by undersell that as a pro as one of the advantages of using it, because the reality is when you build workouts and stuff like that, I can only speak for myself, I don’t build workouts by TSS, I build the workout by the purpose, here’s what I want the athlete to accomplish. But because it’s such a common quantitative number, when I’m done building the purpose and the workout that goes around it, I could be like, “okay, yeah, that’s gonna be 165 TSS,” and be pretty much right. Now I can use structured workouts in today’s world to get it exactly right, but the reality is, I’ll build that purpose first, I’ll understand TSS.

 

Tim Cusick  52:39

The cons of TSS though, they’re out there, and I think we’ve talked about it a little bit already in this podcast is it is not an absolute, TSS is a construct, a model an algorithm built off of other algorithms, it’s meant to be an estimate of that stress, It’s not a perfect number, and it does not give you everything you need to paint the whole picture of a training strategy or a training plan. One and why is it really is a measurement of your muscular metabolic system, meaning it really was taking a look at the energy you were using to produce the efforts that it’s scoring, It does not, though, take into account the specific content of that effort. So if you were to go out and ride you know, and Trevor had just said it a little bit ago, if you’ve gotten right, a one hour time trial as hard as you can, and a one hour crit as hard as you can, right? They’re both going to be 100 TSS points, but from a muscular demand, the way you recruited fiber, the fatigue, the lactate, all the different systems happening underneath that they’re not going to have the same impact, they’re actually have different strain on the system. So, TSS is an overall training load score, it really isn’t content specific in any way. It is a general score, meaning it’s a good average, the basis of TSS came from observation of a pool of riders and data, so it kind of fits in a bell curve. There is some individualization that could be applied there, that could make it more specific, But uh, at this time, it’s a generalized kind of number. It is reliant on a single metric, if your FTP your Functional Threshold Power is not set correctly, your resulting TSS is going to be incorrect. So it does require you to test to maintain that FTP.

 

Tim Cusick  54:43

So I think in there the summary is the pros are, it’s pretty accurate to stress, It’s pretty easy to use, you don’t need to get in the lab, you don’t need to do a lot of other things, you can just track your Training Stress Score every day understand the impact of or understand how strenuous that workout is, and then you can apply that to a system like a Performance Management Chart and look at it over time. The cons are, really can be summarized in, it’s not content specific, meaning it’s only showing the overall training load, it’s not saying here’s a specific way you generated that training load.

 

Trevor Connor  55:21

I would say something you said at the beginning of this episode really rings true, which is, it’s far more useful when you look at it over time.

 

Tim Cusick  55:31

Yeah, I think that’s an excellent point, and it’s why it’s almost hard not to, you can’t even separate the two to some degree, I can’t my mind, right? My mind when I start training  load and Training Stress Score really goes right to the PMC.

 

Trevor Connor  55:45

Yeah.

 

Tim Cusick  55:45

More than any one training stress workout, any one workout, any one note, just, it isn’t relevant enough, it really isn’t in the big picture of creating performance excellence. It really is the overall accumulation of those scores and how they play out that is more relevant to train.

 

Trevor Connor  56:07

We recently spoke with Red Bull Athlete and host of the Adventure Stash, Payson McElveen, and asked him his thoughts about the pros and cons of TSS.

 

Payson McElveen: Pros and Cons of a Training Stress Score

Payson McElveen  56:15

For one, I don’t think the numbers are a particularly good guide when tracking workload for mountain biking, and I’ve had to come to accept that, like if I’m tuning up for a mountain bike race, and I’m going to do four mountain bike rides in a week, and only three road rides, rather than six road rides and one long bike ride, I have to just know that my CTL is going to drop more than it actually is, and I simply have not found a way around that. Oftentimes what I’ll do is I’ll switch the TSS, know how you describe it, I will, instead of using power to gauge TSS, I’ll use heart rate and that seems to get it a little bit closer. That’s not to say that I won’t use both power and heart rate to track my training, but in actual TrainingPeaks, I’ll go to the little toggle thing and select heart rate instead of power as the thing that that calculates the TSS, and the majority of the time TSS will go up, that TSS reading will go will go up, that is really never the case for me on the road, power always yields a higher TSS score. So that’s one of the reasons that I know that that algorithm just isn’t quite right for mountain biking at this point, and it’s sort of where knowing your body comes into and then also, at the end of the day, remembering what the point of all of this is, going faster. Like the reason we split hairs about wattage and heart rate, and durations of interval, training volume, and CTL and TSS, all of these things are so that we cover terrain faster. And so if all else fails, pull out a dang stopwatch. Time yourself on a climb, time yourself on the on a descent, keep track of your time on a loop, and that’s where it kind of goes back to Strava also. I think we can get so into the weeds of power and heart rate and all that sort of thing, and just take them as these definite numbers, despite the fact that we know that heart rate is so impacted by heat, and so are temperature, so impacted by fatigue levels, power, I mean, even the most accurate power meters out there are going to vary a little bit for mountain bike training specifically, there are so many extra variables going on, that sometimes I think it’s it’s good to just almost take all of the data out of it except for what the clock says, now and then.

 

Trevor Connor  58:47

I remember having an argument with a friend because I mentioned to him I did a ride where I was trying to keep my average speed up, and he just goes, “you’re still looking at speed? That’s such an outdated number. Why do you even look at that?” I just thought about it for a minute said to him, “There’s never been a bike race where the winner wasn’t the person with the highest average speed.” It’s not outdated at all.

 

Chris Case  59:10

Yeah, the only one that ultimately matters in this in some ways.

 

Payson McElveen  59:15

Absolutely, absolutely. I mean, that’s been this is sort of specific, but over the last couple of years, that’s one of the ways that I’ve known my aerobic engine is getting better, and I think gravel racing is a lot to blame for this, but no matter what type of ride I have, whether it’s an endurance ride, an interval ride, arecovery ride, my average speed is just higher, and the powers going up, also, my weight is continuing to go up, so you know that automatically you’ve got two variables there, so things get a little muddier. Yeah, you have watts per kilo, but at the end of the day, I’m just going faster, from an average speed standpoint, regardless of the type of ride, and that’s just about all I need.

 

Trevor Connor  1:00:02

Because I think about my own training, I think about with my athletes, when I look at individual workouts, the only time I will look at TSS and go, “Oh boy, there’s something going on here.” So let’s say I give one of my athletes a recovery ride, and I see a TSS of 120, I go, “okay, you don’t quite get what a recovery ride is.” But if I go out and do a six and a half hour ride, with 15,000 feet of climbing, I don’t need to see a TSS score to go that was a hard day.

 

Chris Case  1:00:32

Yeah.

 

Trevor Connor  1:00:33

But I do when I’m mapping out my athletes plans when I look at the whole week, I like to put in those approximate TSS numbers, to get a sense of here’s about how hard this week is, because I actually do find that over time, and again, I bet you this that that q&a episode, it’s as long as you are following a training philosophy. So if you have an approach where you go, Okay, here’s the interval work we’re doing. Here’s the You know how we want to balance out easy and hard and long rides. I do think you get into a little bit of trouble when you are just designing a week to try to get as big a TSS as you can, but if I build out a week based on my coaching or training approach, I do like to look at the end of that and what is the TSS for this week? So I find that really correlates with about how hard a week they find it.

 

How to Effectively Use Training Stress Scores When Training Athletes

Chris Case  1:01:26

So you just touched upon it, Trevor, how to use TSS, how to integrate TSS into the art of coaching, Tim, I’d like to hear your your perspective on effectively using this metric in coaching athletes.

 

Tim Cusick  1:01:42

Well, the best use for Training Stress Score is training stress over time. So, yes, Training Stress Score, you apply it to a single workout, it’s an excellent quantitative score of the strenuousness of that exercise and workout, but the reality is it’s more important to use it as a cumulative score over time, or apply it in our system to a Performance Management Chart. Once you start talking about Performance Management Chart,  you often want to just go into like training load, like how much and how little. But I think before you, we enter into that conversation, because I’m certainly going to go there, you also have to understand that Training Stress Score does give us an understanding of what I call training rhythm, which is how hard and how easy. So if I’m looking at a Training Stress Score for a day, I’m looking at kind of high low right and understanding and making sure that the athlete is following the training rhythm that I’m prescribing, but then I really want to escalate that rhythm up into the Performance Management Chart. So when we look at Training Stress Score, in a Performance Management Chart, we really have two key metrics. One is Chronic Training Load, or CTL, and the other is Acute Training Load or ATL.

 

Chronic Training Load

Tim Cusick  1:03:01

Let’s start with chronic training load. Chronic training load is really the accumulation of training stress over time, it’s based on the half life of exercise, and most people look at it, it has a default setting, it really is looking heavily at the last 42 days, and then people always say, “Well, why the last 42 days?” Well, the reality is most human beings adapt to exercise or getting maximum adaptations from exercise stimuli and about four to eight weeks, so we take six weeks as the default setting. But in reality, chronic training load exercise has a half life of about 90 days, so all of your older workouts count, but CTL was really focused on the last 42 days.

 

Acute Training Load

Tim Cusick  1:03:48

Um, ATL was the acute training load, which a lot of people you know, like to look at it the seven day and that is our default, and that gives us an understanding of how much short-term training load the athlete has in the system, and there’s both simply a moving average of those numbers. Now, I have to stop when I was trying to define it specifically because I’m going to tell you what it’s not, and I think this is an important discussion that should be a game changer for a lot of new people beginning to learn how to use Performance Management Chart. So when I say chronic training load or CTL, that is not fitness. Right? And you see this all the time and even, you know, you see name changes even within some of our own systems here that we’ve named it fitness. I think, in some ways, it’s correlated to fitness, and there’s a correlation, right? But it’s not specifically fitness. It is simply a quantitative measurement of the external stress you have applied to the system over time, or load, right? You’re Chronic training load. Now let’s say you wanted to go out and you had two ideas, I just want to get my CTL really high because that’s fitness. And you went out and did seven two hour time trials every day, you’d accumulate 1400 points of TSS if you could do a two hour time trial, so let’s say you’d accumulate 1200 points of TSS, and your chronic training load would go up fast, and if you could keep doing that for 42 days, it would get pretty high. But I guarantee you wouldn’t have fitness, you wouldn’t have the ability to perform with fitness, you’ve done a lot of training, but that isn’t fitness. You could also go out and ride 18 hours a day, at eight miles an hour on a road biker, five miles an hour and a road bike and generate the same kind of number, and again, you wouldn’t necessarily have fitness. So all chronic training load is doing you is telling you how much training load you’ve put into the system, that is a quantitative metric of external stress of the system over the time. Fitness is a responsibility and the role of the coach or the self-coached athlete, by designing content under or as part of that training load. Training stress gives us the ability to normalize it and quantify it, but it doesn’t give us enough insight into the training load, or I’m sorry enough insight into the training content to really say we have fitness. ATL is often called fatigue, and that’s not true either, it’s actually more true than fitness maybe, for CTL, but the reality is ATL isn’t necessarily fatigue, it’s just an excellent score of the short term metric of training.

 

Tim Cusick  1:06:48

Now we also have training stress balance, which really is your CTL minus your or your ATL minus your CTL, and offset by day, just perfectly tracking the math. And that’s your Training Stress Balance which people often call form. So now how do I use those three? Well, I never really use TSB to be honest with you. So form and that balance between ATL and CTL, when you go back into my own experience, when I correlate that to performance, there’s no heavy correlation, I find better correlations in chronic training load and acute training load than I do with TSB. Don’t get me wrong, I track it and I look at it more maybe as a training readiness and an overarching look at the athlete, but most of my focus is on the CTL progression. How are they accumulating over time? Specifically, I have a system of kinda plateau and overload, you could you know, people look at periodization of these numbers, you could put that idea into periodization. I think when we’re looking at CTL growing, there’s only so long you can grow CTL, you can keep accumulating training load and expect improved performance out of an athlete, and there’s only so long you can sit at a CTL plateau and expect an athlete to hold a level of performance. So I think when you start thinking about this idea of Training Stress Score of scoring these is some external training load process, It really is about understanding the progression of that training load. That’s the science that’s giving me ability to make better decisions, but I have to color in the content underneath that we’re gaining.

 

Tim Cusick  1:08:38

That is an art form, I mean, I have some specific techniques, I bet Trevor has some specific techniques to that. For me, you know, when I’m thinking about it, I guess like, so CTL was not a prescription, you know, and people need to really wrap their heads around long. You can put out some generalized numbers, and I can give you some right? An athlete, once they get in the I don’t know, 70 to 80 range tend to be getting and assuming the training content is good, not perfect doesn’t have to be perfect, but good, they’re probably going to have a performance improvement in that range, they’ll have another one, and I don’t know, 100 to 120 range of CTL, It really depends on the content, and what they’re doing. And then for the elite athlete, you might see another around 141-145 and above. So you could put some generalized thinking to that, but that’s not a prescription, that doesn’t mean let me just put an athlete on the bike gain to those levels, just ride, don’t worry about what you’re doing, just gather TSS, and you’re going to be great, that’s just some kind of numbers to shoot for. For me, it really is about building a purposeful training strategy, understanding the ability of the rider, and the demands of the event, building content based off that, right? So first up build the workouts, then you understand the weeks, then you understand the months, the four week cycles, the three week cycles, whatever you’re using, then once you get a good grip of all that, then for me what I do in planning, then I back the win TSS, and then I’ll tweak that plan to make sure that the training load is plateaued or overloading in the timeframes is that I want them to work in, then when I have the training actually occurs, I’m measuring that CTL, ATL, growth to what I had put within the trend plan. So I know that’s a long answer, but that’s how I use it.

 

Brent Bookwalter: How To Determine When Your Fitness Is Where It Needs to Be

Trevor Connor  1:10:43

We know we can always look to Brent Bookwalter, Pro with Mitchelton Scott, for some good training wisdom. Brent talked with us about how he determines when his fitness is where it needs to be.

 

Brent Bookwalter  1:10:54

Gosh, these days we have so much with so many metrics, so much data at our disposal, it’s easy to get too caught up in that and to let those sort of numbers become too much of a fixation. I think without being too boring, once again, just preach the fundamentals of consistency, and that sort of just systematic progression as I come into the season, you know, hopefully having the time and the ability and the environment and the conditions to just have sort of a sort of, you know, as clean of a build as possible, sort of applying the stress systematically more and more and more as the weeks go, So, I’m taxing and stressing my system more and more, but I’m also recovering enough each week that I’m also getting stronger and able to load more, so whether I’m just doing that, looking at my my weekly hours or my energy expenditure, or whether I’m using some of the more sophisticated charts and graphs that helped me I think those can be good tools, but they’re never absolutes, and I think ultimately, it needs to come down to the feel and the feedback you’re getting out in the road, and also in your life in general, how you’re sleeping and how you’re, how you’re just rolling through the day to day life. And then ultimately, I’m also relying heavily on my coach during that phase, you know, he’s the one that’s kind of looking back historically comparing my progression to other years, and considering what you know, what the goals were early season in past years versus what they are this year, and then, you know, modifying and tweaking a little bit as needed.

 

Brent Bookwalter  1:12:44

I’ve been doing this long enough and using training and racing my power long enough, you know, I have a pretty good idea like, you know, I can step back into my training and doing my first intervals of the season and I have a good sort of idea like, well, I’m, you know, I’m way off, or like “Like, well, I’m actually fitter coming into this first week than I thought I would be.” So yeah, I am looking at my power, I’m looking at power relative to perceived exertion just on a sort of like average ride, weekly basis, and I’m comparing that it sort of comparing that to my perception in history and sort of all the zones, you know, whether it’s just base endurance, up to my early threshold efforts. You know, I’m using I’m definitely using wattage numbers. Definitely keeping tabs on the heart rate early on in the season, I think, especially when I’m not as fit, I think the heart rate can be a great place to start. Yeah, generally in those first few weeks back, I started to focus a little more on heart rate than power, let that sort of stabilize and equalize, It’s usually pretty high when I’m not as fit. And also in terms of like climbs or 20 minute tests, I don’t, I can’t say I always do before the first race of the season, I don’t always go and hit you know, the same 20 minutes test or the same climb, but I do sort of have one of my training buddies back in Asheville, Johnny Clarke and I, we have this loop we do called the Form Finder, and it’s just a nice around five hour loop, depending on how we do it, and it has four or five solid like 15 to 20 minute climbs up, and that’s, that’s a Form Finder, you know you’re out there, whether you’re just riding that loop at at base endurance or whether you’re going out and doing a workout on the climb. You know how I roll back into Asheville, like when I come back in through West Asheville, I’m just rolling over those rollers, if I’m totally on my hands and knees versus feeling pretty fresh and rockin over him, that’s a good indication on my form. I kind of try to do that ride, you know, once a week or you know, every other week during the offseason, and that’s a nice way to sort of keep checking back in and that form, see how I’m fine.

 

Trevor Connor  1:14:42

So that’s interesting. It’s more how you handle the overall ride, versus what a lot of people do of saying, “Okay, I’ve got five Strava segments here. Let’s see if I can take everyone,” it’s more, how does the whole five hours affect you?

 

Brent Bookwalter  1:14:54

Definitely, yeah, for sure. Because you know, so much of I think what I’m training for, what I’m training for is not very Strava segment hunting appropriate, you know, we’re the repeatability of the demands of the races that I’m doing, It’s, I have to be able to do the load deep in the race and the fourth, fifth and six hour, day after day and day, day, seven day, fourteen day, twenty in a Grand Tour. So yeah, it’s for me, it’s really about how that how I’m able to tolerate that load. Hour through hour, and ride day after day, and then week after week, that I’m kind of monitoring.

 

High CTL Is Not Always Better

Trevor Connor  1:15:32

Too many people look at it as the higher the CTL the better, an example I’ll give you is an athlete who I’m coaching right now, he a few weeks ago, went out with a couple buddies, and was ripping them apart. He was out he was at a peak, he was feeling pretty good, and at the end of it, “they’re just like, what’s your CTL,” he goes, “70,” they go, “mine’s a 110, how are you beating us?” And he asked me that question I’m like, “probably because they’re fatigued out of their minds and you’re fresh and ready to race.” So it’s certainly not a case of the higher the CTL, the better, now there is a degree if your CTL is 30, and you get it up to 70, well, they will bet a lot of money that you’re, you’re a little fitter. But what I have noticed is every athlete has an ideal range, an example I’ll give you is, remember that that CTL is based on FTP. Now you take a sprinter versus a time trial stage racer, that time trial stage racer tends to have a very high FTP, it’s very close to their Vo2 Max. Often, they don’t have much of an anaerobic capacity, where that pure sprinter is going to have big anaerobic capacity, their FTP is actually going to be quite low. So if you took that sprinter and tried to train them to a really high CTL, you’re gonna probably kill them. Where that that time trial stage racer, they can generate a lot of TSS on their training ride, so you’re going to see them get much higher, so at that type of rider, I get to target a much higher CTL.

 

Tim Cusick  1:17:21

I think that’s an excellent insight. This is one of the challenges of all the data that we have in today’s world, people might be they might find it too easy to just make the data the solution, and not take the time to learn how to really bring it to life as you just did a great job of explaining individualization, really more in a phenotyping way, right and grouping similar people together. But the reality is that stuff, you still need to invest in the learning. If you’re like, wow, I have a Performance Management Chart, right? It’s got a blue line that’s going up, that’s important and yes, with enough change, you will get  faster and faster, but you want to learn to maximize that you do invest in the education of what’s the proper way to make that line go up and to what number and what’s the right thing. Just because we have data, it doesn’t, that doesn’t mean we don’t have to learn. As a matter of fact, it should be one of the tools that enhances us to learn more, to test more on ourselves and our athletes or other things, and then continually improve.

 

Chris Case  1:18:26

Well, let’s talk about the art of managing training load, and Tim, do you have a system here?

 

The Art of Managing Training Load

Tim Cusick  1:18:31

I do. I aave a process, you know, and understand one of the biggest challenges in defining your processes, people think you’re going to be absolute, not every athlete fits neatly into this process, but it is the one in general and managing training load that I try to apply. I think one of the things when you’re measuring training load and this should be something you know that everybody and every coach should be thinking about when they’re thinking about CTL and ATL specifically is, you really need to think about progression and trend. I think people underestimate trend, and we’ve seen some newer metrics in the performance management system like Ramp Rate and other stuff, which is helping people think about trend. Let me define trend, trend is really the change over some set amount of time in one of these metrics like CTL. So when it comes to managing training load, that trend is very important to my desire to bring an athlete to peak. So in what we would traditionally call base training, I’m looking for low trend, variable content, a multi discipline approach, meaning we’ll do some cross training, I might want a little more strength out of some athletes and maybe a little more base for aerobic fitness, it really depends on the athlete and their athletic maturity and other factors. But that trend is a slow burn trend, we’re not trying to create a fast ramp up of chronic training load, we’re not trying to add training load at a fast rate until we’re ready for that rate. And for me, probably more than most coaches I interact with, I’ll extend that longer. I think there’s a point where every athlete needs to accept that they’re gonna benefit from more and more based training. I mean, until they don’t, but the reality is they often get wait limited by time then before the actual benefits of base training will stop giving them you know, simple underlying aerobic gifts.

 

Tim Cusick  1:20:42

But then, to me, what becomes really important is when it okay so we’ve got that early trend, we were building training load slowly, I’m trying to build that athlete to where we move in what people might call to build and taper kind of event, which I just call performance as a whole. Right? In that performance phase, I am an overload believer, I think it’s really important at that point that once we have a certain amount of TSS in the bank, right and the PMC, we’re going to accelerate overload, so I’m looking for more of a change in the trend, a shift that ramp rate than the absolute number. I’m less concerned if that number is going from 70 to 90 versus 100 to 120. I want the ramp rate to be very specific. So for me, I do a lot of custom analytics, you know, and, and this is just a WTO thing for me, because I can and it’s flexible enough to do it. I’m really looking at 28 day CTL trend, now I’m not talking about changing the default metrics, what’s driving the PMC and CTL, I’m literally talking about the daily numbers. So I’m looking at the trend over 28 days and I’m actually looking at 56 days. So I’m looking at four to eight weeks, right? And the reason being for me is once we get in all that great bass training, we’ve created the basis that we’re going to perform on, I don’t want to train the athlete to peak, I want it to be hard, Fast and Furious, right? If you’re training an athlete to peak for 12 weeks or 16 weeks, that’s way too long. Once you turn open the faucets and you’re doing your heart or training, you’re really going for peak most of the gains is going to come in three to five weeks.

 

Trevor Connor  1:22:30

Yep.

 

Tim Cusick  1:22:30

The rest is just polishing off some speed and really getting them ready. Then I’ll use trend to taper but to be honest, I’ve become more and more, this is why I don’t use TSB, tapering to me is a dangerous science, right? It’s a dangerous art, better way to say it. You one of the places where I see some of the most performance individualization on the athlete is what they need to taper. So I have athletes who will perform better on the forward moving trend who actually need to load into peak performance, and some other athletes need to taper. One of the things that I will use CTL for that is, you know, there’s a point of diminishing returns loading into an event if their CTL was already very high, I might say, “look, we were not going to load quite like we would.” But that’s the whole point of measuring it correctly over time. So I’m more of a trend person, I see greater performance not coming from some training load absolute number, ie CTL. But the trend of that number, again, you have to assume good content, CTL doesn’t measure content. But if you’re measuring that good content that really is the best way measuring or management of that trend and ensuring that trend is occurring correctly, is the best way to bring an athlete to peak on the timeframe or when they need to peak. So my system really is that, plateau and trend, plateau and trend, are more important than absolutes. Don’t get me wrong, I have the luxury of working with pretty much all professional athletes at this stage, and they have time to train and we have, we can execute very, you know, it’s their job, right? They’re training every day, so it’s easy to live that it gets harder with people who are working for a living and stuff like that. But I find that’s the best way for me to use the tracking of training stress or training load, in this case for performance. But all that being said, I have to say it one more again, in that the content trumps, you could do all that training load with the wrong content, you’re still gonna fail. You have to have the right content in there, but using it and plateau and trend management, that really does set up the best environment for success for the athlete.

 

Trevor Connor  1:24:47

You know, I see the same thing with Masters athletes I coach. I’ve had several athletes who have come to me for coaching, after a couple years, they feeling like they kind of plateaued and so I’ll upload all their data into WKO, and lo and behold, you look at their CTL line, it’s a flatline for the last three years, they’re just kind of sitting at a 70-80 CTL in the Winter and the offseason and the season, that it’s it almost becomes this thing of, if that number goes down, I lose my fitness, I have to keep that number up. And because they’re never letting it vary, they also don’t get it very high either. And they just got to sit at this plateaus level, so I would say, which is what I think you’re describing one of the things I really look for as a particular shape to the CCL lines, it’s like I said, it’s not about how high can we get it, it’s getting it to the right point but more importantly getting the right shape and being at the right point at the right time.

 

Tim Cusick  1:25:53

Yeah, you know, you made an excellent point about the flat CTL over the time, you know, that’s training stagnation. You know, at some point, the athlete has to give up some load and rebuild, just to revitalize the system. If you’re constantly within, if you’re spending a year within that same CTL range of 10 to 20 points, because you don’t want to give up any fitness, but you don’t have more time to train and kind of take it over the top, you’re not gonna get faster, you’ll get faster for the first 8-10 weeks from that range, but after that, you’re not, you’re just gonna stagnate and eventually decline.

 

Trevor Connor  1:26:28

You know, I experimented that with 2018, because I had some people who said, “Trevor, you gotta try this,” or they had that conversation with me about you know, “why do you always let your CTL drop?” I let mine dropped 30-40 in October and November, and it rebuilded back up over the base season. So I finally just said, “Okay, let’s give this a try.” And so fall of 2017, all through the winter of 2018, I’d never dropped below 80, and I hit March and I was doing the best I’d ever done and races March. Problem was my target was Nationals in June, and I flatted out in Nationals, but let me tell you, I was never so happy to flat out of a race. Because I have no legs left, I was done.

 

Chris Case  1:27:14

Looking for excuses.

 

Trevor Connor  1:27:16

Yeah, I hit June, and basically my season was done.

 

Chris Case  1:27:21

Mm hmm.

 

Working Backwards to Understand How to Improve Going Forwards

Tim Cusick  1:27:22

And that’s, you know, it’s common. And it’s so funny to the reality, there’s a lot of individualism to that. That’s where having one of the things we’ve been having is a lot of discussion about current training load, right? One of the things when I say that data science is decision science, one of the things that should help you make decisions on your current training load and your predicted going forward training load is go back and look at your history. There is going to be some correlation. Right? There should be some correlation between performance and training load, content really matters, so it’s probably going to not be clean and neat and linear. But don’t forget to work backwards, to understand how to improve going forwards. And that’s the beauty of all the power of data analytics now, we have the end the fact that so many of us have had a power meter now for 5-10, and I think I’ve had one for 15 years now, right? You can go back through all these years of data, and you can say, “what am I learning from history? What am I learning from my historical performances?” You now have that lesson ingrained in your head, but you actually have the data to show it right? The language, here’s what I did, and used it, you know, and it’s language. I kept my CTL in a certain range, I did this, this and this and it burned me out by a certain point. I think it’s super important that that people don’t just take this kind of discussion we’re having to think about it that’s just go forward, don’t be afraid to look back and look at your data and look what’s occurred and try to use that to make better go forward decisions.

 

Trevor Connor  1:28:50

Yep, I agree 100% Great point.

 

Tim Cusick  1:28:52

Here’s a great CTL thing, right? I am living the ultimate CTL science project right now, you could take this verbatim, um, I coach these two athletes to pretty well known professional level athletes. One I have Amber Neben, and you know multi-time World Champ Time Trialist, driving for the Olympics. On the other hand, I have Rebecca Rusch, you know, adventure racer World Champ, multi-time Leadville Dirty Kansa winner, you know, everybody kind of knows her story. Amber is the ultimate machine time trialist, phenotype right in there, Trevor, some of the points were making before training load is, I don’t want to say it’s King, but it’s up there, and she can train to such a high load will drive high peaks, but we can’t do it every year. So you take an athlete like that, and you have a three, four year thinking in your CTL cycle. Right? And that’s important with athletes at that level, particularly her, looking for Olympics, so you have an Olympic cycle going. So in Olympic years, over three years, she you know, three years ago she was lower CTL than she’s been in the previous three years and higher higher, right? So, and she will train, she’s made the long team Olympics, we will train pretty deeply when the time comes to a pretty darn high CTL, considering she’s a time trialist, she’ll be on course about 40 minutes, that’s a big deal. Then you have Rebecca, so she’s an amazing athlete, just like Amber is total natural grain. If I tried to get her anywhere close to those same CTLs, we’d probably kill her. Trend is way more important, she could be to 30-40 CTL, trend accurately for five to six weeks, and go out and do amazing performance and event. There’s events that I’ve sent her off to with a CTL, where I was afraid we were going to kill her. Like “Wow, she’ll never come back from that, it’s going to be too hard, She’s not fit enough.” But as long as her trend is good, she performs amazingly well, she’s not as dependent, Amber needs the big load to respond to training. Still the same kind of you looked at the trend, same trend coming into peak, Amber takes a little more load into the final performance, Rebecca needs a little more taper, but the reality is it’s shaped sort of the same, and they’re probably about 70 CTL points apart, but they both go in and perform really well in their events. The absolute number isn’t only and that’s extremes, right with that much different, but the absolute number isn’t always the driver, that understanding the individual, understanding the trend it takes to get them on peak form and perform is key. You can look back over their histories, I have the luxury of looking back over the histories, and I can see that, so I can use it going forward. But it just shows you that peak performance, now would Rebecca do well at Amber’s CTL? No, I literally don’t think she could get there. Would she do better maybe with a chunk more? Maybe, and maybe we’ll find out, but the reality is they can perform pretty darn well, as long as that pattern brings them to peak form.

 

Chris Case  1:32:07

So Tim, one final question before we close out the episode, and that is, what’s the future hold here? What’s your view of data science and training load measurement going forward?

 

Tim Cusick  1:32:20

I think what’s really cool in this field is we have more and more wearables and data collection devices, so other sports are really now getting into the game of measuring training load, you’re looking at professional football and ice hockey and, and all types of different sports now collecting some of the same data we’ve had in cyclists for years. So this is bringing real horsepower to the idea of analytics, I think what you’re going to see the next evolution is this ability to get to adaptation scoring, I think, I truly believe you know, the stress plus strain equals adaptation. We have a good stress score, I think training stress is better stress score, then, you know, some people might think initially when you begin to quantify it, it works. We’re going to get to an adaptation score about five years from now, because what stops us now is we don’t have enough data. I know that’s shocking, Like, we have all these devices, and, you know, you have your Garmin with eight screens at ten data fields and all those things go on, right? And the reality is that we don’t have enough to the variables of data, we need to better understand data like HRV, and get that continually improved, we need to understand sleep, we need to understand nutrition, we need to understand what’s happening within the body. Now during these training loads, I think we’re getting there we’re able to collect more and more this data, and we’ll see analytics in the next couple of years, assuming we keep enhancing the robustness of the variables data that we’re collecting, I think you’ll see things come down more and more to predictive analytics, which is exactly what an adaptation score is, if I do this, that is going to happen. So or actually a better way to say it might be, if I want this to happen, I should do this. I think we’re getting there, I think three to five years, you’re going to see the ability to begin to take this data science all the way to that extreme.

 

Tim Cusick: The Future of Data Science and Training Load Measurement

Chris Case  1:34:22

That means Trevor’s gonna be out of a job soon?

 

Tim Cusick  1:34:24

Yeah, no.

 

Tim Cusick  1:34:28

At the of the day, and maybe that’s a great ad, I mean, I will tell you right now, in the end of the day, you will never replace the coach. But if the coach says, “I need to do this, I need to get this.” They’ll have more data to make a better decision.

 

Chris Case  1:34:42

That roadmap just becomes a little bit clearer.

 

Tim Cusick  1:34:45

Yeah, right. But you’re never going to replace the coach because you’re not going to do in the absence of knowledge and understanding. It’s never going to happen, because even when we have more variables to put a scored adaptation, there’s two still too many other variables, you know, way too much, you know, Trevor, give an example, phone starts ringing on Wednesday and Thursday right? People need to tweak they need to change, those type of things AI all this analytics will never replace that, you’ll just have better coaching tools, I don’t think you’ll ever replace coach.

 

Trevor Connor  1:35:19

Glad to hear that, you did a good job.

 

Chris Case  1:35:22

Very good.

 

Trevor Connor  1:35:27

While back, I spoke with Armando Mastracci, the developer of the Xert training software platform, that takes a bit of a different approach then TrainingPeaks. I asked her Armando his thoughts on the future of software. Let’s see how his thoughts compare to Tim’s.

 

Trevor Connor  1:35:42

So where do you see the the next revolution or the future of the software, what’s going what’s going to happen in the next 5-10 years?

 

Armando Mastracci: The Future of Software

Armando Mastracci  1:35:51

I believe that the software is going to start to incorporate a broader range of information. So as we instrument more of our lives, Right? That this information is now going to be incorporated as part of our overall training. So one of the one of the biggest challenges with our software, I think, for a lot of software to be able to really prescribe training, is that there’s just so many variables involved. So you’re only really looking at it through one lens, Right? So all you have is power data, you only have power data for indoor rides and not outdoor rides, and we only have power on one bike and on the other bike, it’s really difficult to get a real clear picture of what this athletes, what kind of training they’ve been doing. And then when you start to incorporate other types of training, whether they’re in they’re in the gym, and you know, they may have other types of sports that they want to they want to train as well, Maybe they’re a triathlete doing multiple sports, in which case you need the instrumentation across these other sports as well. I think that’s where the really the tension is going to be is how do you provide a greater holistic view of training beyond just cycling, I think cycling is great, it really created a way to kind of quantify things that you’re unable to quantify in other sports, because of the availability of the power meter. But we can apply the same principles across other sports and start to quantify and use it use the same ideas for different types.

 

Trevor Connor  1:37:12

Now do you also see that incorporating recovery metrics? So sleep?

 

Armando Mastracci  1:37:17

Absolutely.

 

Trevor Connor  1:37:18

Yeah, this this new, you know, obviously people, you can think of the Fitbit switch track, how active you are through the day.

 

Trevor Connor  1:37:26

Absolutely.

 

Trevor Connor  1:37:27

Do you see those sorts of things being incorporated into the software?

 

Armando Mastracci  1:37:30

Absolutely. So you know, when we talk about the instrumentation, this is what we’re talking about is, are we going to be able to collect sleep data, are we going to be able to collect HRV data, are we going to be able to collect other data related to how our various muscles are performing or being used like here, maybe motion, there’s a whole bunch of information that we can collect. And with the availability of this other information, and the ability with the computing power that we have now available to us, we can start to use that information to better identify what is best suited for a particular individual, given the types of exercise that are performing, with the information that you’re gathering, and then the types of athletes they want to become. All that information is now going to be much more valuable and useful, rather than getting a very narrow view of an athlete, but now you’re getting a greater 360 view of that.

 

Trevor Connor  1:38:17

Now, what do you see the software doing? Because man, it’s gonna be a ton of data, obviously, they’re not gonna want to be sitting there trying to analyze all that data themselves or interpret it. So what do you see happening with the software in terms of using that data? What’s the software going to do with it?

 

Armando Mastracci  1:38:33

Well, you know, there’s new when it comes to things like machine learning and networks, and capacity of software uses computing power, certainly, there’s a lot of opportunity. We do have a lot of data. But the challenge with a lot of the machine learning processes is that, how do we identify what we’re trying to optimize? That’s the biggest challenge, right? So I was reading this paper recently, by Tania Churchill, who’s based out of University of Canberra, so part of her PhD thesis, she did his sort of machine learning for training, implemented, what’s called an artificial, hybrid artificial neural network to analyze all this training data for cyclists, which is, which is really good. If you’re ever interested in understanding kind of how machine learning will be able to be applied, how it what the challenges are, she really gets into a lot of great detail, so I’d recommend anybody who’s interested in this to read to read her read a report, you know, there, there are a couple of really big challenges, right? The big challenges is, what are you trying to maximize? What’s the what’s the goal? What’s the objective, right? These are really difficult to identify. So we’re so that’s one of the challenges is to identify, within the data itself, What are the goals of that particular training? And those can be a challenge. And then the other challenges is that you know, the data is very individualistic, everyone comes into training with some level, some context, and a U.S. coach would know that when you first have an athlete, you need to understand what you have to learn the athlete first, before you can really tell them what to do. Right?

 

Trevor Connor  1:40:12

Right.

 

Armando Mastracci  1:40:12

So part of this whole process is gathering data first about an individual, so all this data needs to be collected first, before we can start to use it and make meaningful information out of it. So if you’re just going to show up and you have all these new devices attached to you, you say, “Okay, I’m gonna get the best training.” Well, you’re not there yet, right? You’re gonna have to collect the data for a period of time before we can really prescribe the training. So everything is very individual, so you can’t really solve that problem by throwing more data at it, because each individual is different and  where they’re starting the process is very different.

 

Chris Case  1:40:48

Well, Tim, it’s been a couple years since you’ve been on the program, but hopefully you remember that we like to close out with a take home messages, give you 60 seconds on the clock. This was a long, deep discussion about a lot of complex things, but what would be the one or two most important things you think people should take away from this episode?

 

Important Takeaways

Tim Cusick  1:41:12

I think people should understand what training load is, and what Training Stress Score since we ended up discussing that, as a you know, a training load measurement, what it is. And I think the most important thing to take away and look, I’m a data person, you know, as TrainingPeaks WKO product leader, I deal in data analytics. But when you look at these numbers, it really is the art form of using it. The one thing I’d like listeners to take away is, please use the metrics, but take the time to learn what they really are and what they’re not, and invest in your own learning of how to use these metrics as not as a data science, but as a decision science, that you are taking these metrics, it helps you learn about the athlete or yourself, you’re investing your knowledge of what that learning is telling you, you’re understanding what it means. And then you’re using that information to make better training decisions, based on how you apply training load. I think it’s important that people take that away. CTL is not fitness, higher number is not better. You don’t get a trophy for high CTL you’ll get a trophy for winning a race, right? They focus on that. And you’re going to do great.

 

Chris Case  1:42:33

Trevor?

 

Trevor Connor  1:42:33

Oh, I have to adjust mine because that was very similar to what I was thinking of saying. So what I’m going to add is, numbers are not performance. There is no number that you’re going to hit and go “Okay, now I’m going to win races.” And for some reason where I’m going with this, I give us an example our last, last guest on the show was Emile Abraham. Emile, in his time was one of the best crit riders in the country. So he knows how to come off of the line going super hard, he knows how to push through those corners, he’s got a great sprint. Though we asked him about Zwift, He’s like, “no, I hate it. I can’t last more than five minutes.” And Zwift is the one place where I get to say, “yes, numbers are performance,” because that’s the only thing that Zwift is taking in is your power. And even though it feels pretty close to real racing, it’s not quite the same, and there’s a whole bunch if you saw this with a meal. There’s a whole bunch that numbers will never show you, so you have to use them as a guide, they’re a very valuable guide. They’ll tell you a lot about yourself, but then you have to take that further step that they can’t show you. As you’ve said a few times through this this episode, You have to still take responsibility for your own training.

 

Trevor Connor  1:43:58

Chris?

 

Chris Case  1:43:59

To follow on that point, I think it’s worth noting that you can go to the extreme and overemphasize the importance of these numbers, and then you take an example like maybe myself, who has basically neglected to look at any numbers for quite some time now, and goes solely on feeling, what you want to do really is some hit somewhere in between. You don’t want to go on either side of the extremes, just look at numbers and say that is an absolute, that numbers going higher and higher, that means my fitness is going higher and higher, that’s not true. And not look at any numbers, Well, if you’re not into racing, that’s fine, If you’re just into riding the bike, and numbers aren’t your thing, that’s fine too. But if you’re really trying to see progress, improve performance then neglecting all of these things would be like throwing out the map and just trying to wander your way to the destination you’re looking for, which isn’t maybe the best approach either, so somewhere in the middle is that place where you really want to end up, using everything you’ve got, comments field, CTL, numbers, TSS, etc, etc., to really hone in on the destination.

 

Chris Case  1:45:24

That was another episode of Fast Talk. As always, we’d love your feedback. Email us at fasttalk@fastlabs.com, or record a voice memo on your phone and send it our way. Subscribe to Fast Talk wherever you prefer to find your favorite podcasts. Be sure to leave us a rating and a review. The thoughts and opinions expressed on Fast Talk are those of the individual. For Tim Cusick, Jared Berg, Payson McElveen, Ned Overend, Brent Bookwalter, Armando Mastracci, and Coach Trevor Connor. I’m Chris case. Thanks again for listening.

Related Pathways