 Training, load monitoring and prescription has been a hot topic recently within both academic and applied circles within sports science, and really there's been a lot of arguments and discussions around what are the advantages, what are the limitations, how do we move forwards as a field. And I think a lot of it comes down to different people coming at the same problem, but from completely different perspectives. So hopefully it's really useful to get an academic and a practitioner together to discuss this and think, yeah, how do we move forward? What are the limitations of our own current approaches? And that's the idea of this chat today, really. So it's a bit of an introduction. I'm Stuart McEarlane-Nailer. I'm a lecturer in sport and exercise biomechanics at Loughborough University. My PhD was very theoretical in nature, looking at the simulation of shockwave transmission through the body after people land on the ground. So purely theoretical biomechanics. I then took a job at a very small, new university with a much more applied focus where I got involved in a bit more wearable tech and training load monitoring. Having now returned to Loughborough, I'm trying to merge these two worlds and say, how can we use wearable technology as an input in place of some of the laptop or computer-based things that I've done previously to try and advance the training load monitoring field, whether that's in sport or rehabilitation. So that's the background I'm coming from. And I'm really glad to be joined today by Joe Club. And yeah, I'll probably let Joe introduce herself rather than me butchering it. So to you, Joe. Thank you, Stuart. Yes, I'm Joe Club. I'm representing in this conversation, the applied practitioner. I have been an applied sport scientist in professional team sports for a number of years. I worked for Chelsea Brighton in English football before moving to the US, where I worked for the Buffalo Sabres in the NHL and then the Buffalo Bills in the NFL. And now I'm working as a sports science consultant with teams, individuals, organisations, sports tech startups all over the world. And I also just love to discuss sports science insights and that translation from the research to apply practice. And I do that through my company, Global Performance Insights on my blog and my YouTube channel. So I'm really keen for interested to get into this conversation today. Great. Thanks, Joe. And yeah, definitely check out Joe's YouTube channel while you're here or maybe after this conversation, there's some really useful videos over there in terms of training load monitoring as well. I guess to start the conversation, then, it'd be really interesting for me and hopefully people watching or listening to this, if you could just give me a bit of a background in how you or others within your area typically monitor training load and what you think kind of the pros and cons of that are at the moment. Yes, so I will admit firstly straight out the gate that my bias, my experience is more so towards team sports than the individual kind of endurance based sports. So generally speaking in team sports, external load monitoring, so player tracking of distances, speeds, accelerations, decelerations on a whole body level has desperately become the focus. There are a lot of technologies, predominantly GPS now, that is very commonplace across a lot of these settings. And even now in the last 10 years or so, we've seen competitions permit league wide tracking or perhaps I say permit, but sometimes these technologies are determined by the league. So there is a league wide provider on game day, which may then differ from in practice. And given the ease of those kinds of technologies, particularly RFID or optical tracking that doesn't burden the athlete, those have become the go to external load tracking in competition. But that might not be the same provider or type of technology as being used in practice or training in the week. And then internal load, it's interesting, right? Because I think early on in applied practice that was came first in terms of heart rate and GPS, sorry, heart rate and RPE. And perhaps now in certain sports, we're seeing that movement away from internal load tracking. I think that's just the nature of perhaps external load is maybe now less invasive and just gets so many numbers. That internal load tracking has gone by is becoming less common in certain sports. Yes, when you say external load gets so many numbers that it's kind of overwhelming internal load measures. What kind of numbers are you talking about on a kind of daily basis? What kind of things are people actually paying attention to? Definitely bias towards distances, total distances and high speed distances running distances above certain speed thresholds. Now, there's going to be so many rabbit holes. I think we could go down in this conversation, right? Because immediately there, OK, we're trying to turn continuous data into buckets to say, OK, above this speed is high speed running or above this speed is sprint distance. And that also varies across sexes, ages, individuals. So there's a whole bunch of complexity there. We have seen in recent times, perhaps more awareness around acceleration and now catching up is deceleration. There's more discussion in our field that perhaps decelerations are such a damaging movement. And I'm sure you have thoughts on that biomechanically and associated with injury risk and performance that we need to pay a bit more attention to that. And then I would sort of say the third bucket are accelerometry derived metrics. So obviously, these are provider specific. But if I pull out perhaps the most common one is player load that's trying to capture those lower speed but higher intensity type movements. But as I alluded to, whilst that has benefit in applied practice, because those wearable devices aren't always used in competition, then we may be limited with when we get access to that. Perhaps we only get it during the week in training and not come game day. So what other constraints are you working within? There's a key one there in terms of what you're actually allowed to wear in competition, but even in a training or rehab kind of based session, what are the kind of constraints of what is and isn't doable? A league wide provider, for instance, the NBA have a committee that assesses different technologies. And even though in that instance, you might have some variety in those that you could pick from. You have to pick from the ones that they deem suitable, which is quite interesting context. One thing I've written about on my blog is this precision practicality trade off. So it's perhaps not a legal constraint, but I think in applied practice, we are always trying to balance this desire for precision and accuracy with the practicalities. And I think this is where this is exactly where our conversation intersects. Right. So one example I'd give for that, although it's less biomechanical, is body composition. So DEXA being the gold standard, the most precise methodology, but not particularly practical to either go to or have in your environment. That said, some environments with more well resourced to have it versus perhaps pod pod or skin pod calipers, which are more practical tools, but then you lose some of that precision. And so I think for any kind of technology, but in the context of talking about low monitoring, it's this, how do we best compromise on precision to maximise practicality or get the best trade off there? So, for instance, often now GPS devices or wearable devices will be worn rather than in those vests. They will be worn in custom made pouches. That are sewn into the back of player jerseys or shirts. It makes it more practical. You can put it in there in the morning long before the players get in. They don't need to put anything on. There is no burden to them whatsoever. But of course, that approach loses some of the precision in terms of the reliability, the additional movement of the device. If I think back to my time working in American football, some practices will have your full pads on. Some will have what's called shells, which is like a lighter equipment. And some might just be the jersey. So in all those instances, there is a difference in what's being worn underneath. But in order to try and maximise compliance, that's the approach that's taken. Yeah, that's a really good point there as well, in terms of the separation between GPS based metrics and accelerometer derived metrics as well. Because I think if you're looking for total distance covered within reason, whether that GPS unit is in an elasticated vest or on top of other clothing, the distance covered is pretty much the distance covered. Whereas once you start getting into things like play load from Catapult or dynamic stress load from statsports, if these are based on peak accelerations, then we know from just pure biomechanics research for getting training load from in it that the attachment of an accelerometer to the participant is incredibly important to accelerometer based research. How tight that attachment is is going to have a big impact on not just kind of peak acceleration, but even not wanting to get too technical when we get into frequency components of that signal, the way you've got high frequency noise, maybe low frequency of the thing. Actually, you know, the limb moving moving through space and then somewhere in the middle, you've got the actual shockwave from the ground traveling up through bones, skin, muscle, clothing. They'll travel at different rates through different materials. All of that is going to be changed by the actual mounting of the device. And I think that links quite nicely on to another question I wanted to ask, which is how accurate is accurate enough because as academic, there's a big tendency for us to try and almost assume that everything has to be as accurate as possible. And it has to be lab based or it has to be the absolute gold standard or it's not good enough. But like you said, with the precision and practicality trade off, if something's very accurate, but can only be done in the lab, that's not good enough. If something can be applied in the field, but it's not accurate enough at all, that's also not good enough. And maybe we'll come on to this later. But one of the areas where I think we need to work together is actually establishing how accurate does it have to be? What questions are you trying to answer? Is it just how did their training load compare to last week? Was it higher or lower? That only needs a certain level of precision. If it's comparing two drills or two athletes, there's a certain level of precision. I think I don't know if you've got any answers to any of that, but just how accurate or how much error would you be willing to accept in some of these measures in order for it to be kind of applicable in a field based setting? Or how would you even go out making those decisions? I think it's a great question. And one I wish I had the straightforward answer for. But I think this is the beauty of this conversation, right? Is perhaps academics and researchers would be surprised in certain settings at how much of a bias and a lean there is towards the practicality standpoint, because if you just consider the logistics, I'll use American football as an example because it's just such an extreme example in terms of logistical demands of in certain times of the year in the offseason, preseason period, you have 90 athletes, nine zero and, you know, perhaps four or five strength coaches and sports science staff trying to run around and manage all of that. And obviously the technology and the training load is just one aspect on top of everything else that's going on. And even like you were talking there about the accuracy of the mounting, you know, yes, the company's custom fit vests are preferable. But then even if you were in an environment where you use them, insisted on burdening the players with putting that additional equipment on, well, then you're juggling the different sizes of the garments and, you know, alignment, however big, 150 kilos, 300 pounds is not going to fit into a small or medium. But then you get the other guys with an extra large that's alignments worn. And so I do think there is this need for more collaboration and communication as we move forward here of like what the amazing work that's being done in the lab, but the realities, again, of the of the applied setting. Yeah, I completely agree. But I think that's a really good point of where I think as academics when necessarily even giving practitioners the information that they need in order to make those decisions, because all of the validity and reliability work where that's even possible to be done is in very controlled environments. Yeah, we're even back to that mounting example again. But reliability work will typically be in, you know, the manufacturer provided harness. Nobody is saying this is the reliability of the unit if you wear it in this particular way. Or nobody is saying, you know, we might have a study that says this is reliable or valid to within certain limits. But then it's being extrapolated outside of the range in which it was tested. And you almost can't make a decision around is it OK to accept this precision practicality trade off because you don't actually know how big that trade off is unless we're measuring those things. I'm interested along those lines of what isn't research currently doing that you wish it was. But if you forget about practicality for a moment, what would you want to know about an athlete that you don't currently know if you could design a training load metric of some kind? What is it you can't currently measure that would be your kind of holy grail that you really want access to? Oh, I don't know. Well, obviously holy grail is the performance in the injury prediction pieces, right? I mean, clearly we are biased towards these whole body metrics in training load, total distances and play loads and acceleration base. But of course, we know even from the textbooks and appreciating the complexity of the human body that there's so much more going on inside in terms of the tissue types in terms of of the joints. And I think that we've got a great starting point now with these global whole body external load measures. But actually, you know, there is so much more in going on inside the body in terms of what's going on with the different tissue and structural components and, you know, linking with that that trade off again in applied practice. We're we're always seeking to simplify things. And this is perhaps the difference, again, of of the the lab where you're going into more and more detail and you can disagree if you want. But, you know, as I said about external load monitoring, we're getting hundreds, if not thousands of metric. And we want to know the one, two, maybe four at max measures that we then report to coaches or physios, medical staff or strength conditioning staff. But in doing that, I think as practitioners, it's important we realise that we are oversimplifying it, that we are reducing complexity down to a handful of numbers. But I would love to have more access under the hood to those, you know, more specific measures of, well, we're talking about training global, but what impact does that have on the tendons, on the muscles, on the joints? Because that when it comes to rehab, that's the level we look at. But we don't tend to look at that level on a global training load basis. Yeah, you raised the point that I was going to make, which was around simplicity or the number of metrics where I think if we separate this into a performance or rehab context, you can definitely see why in a rehab context, it would be useful to know what is the load on through the Achilles tendon or the hamstring or ACL, etc. But then from a general kind of, I guess, from that rehab perspective, you're targeting a specific tissue and looking for relationships between stimulus and adaptation. But from a general training perspective, where does that balance lie between? I guess you're looking for a meant from what you just said, you're looking for a metric that describes the stimulus to adaptation for the whole body. But then even if we were able to accurately give or quantify that stimulus, it would be specific to tissue types. Like you said, are we talking about tendon, ligament, muscle, bone? What is it we're talking about? But even within that is what muscle are we talking about? What ligament? What tendon? But if we were in this main belief world that might not be too far away, but if we were able to even estimate what is the load through the ACL ligament, for example, if we could do that, you'd then have this even longer list of metrics than now has every part of the body. So I guess, again, you probably won't have an answer. But what is it? Are you looking for a load on specific tissues? Are you looking for a number that represents adaptation at muscular tendon level? Something that represents this is ground reaction force. And we're going to assume that it correlates with loads of other things. So this is an overall measure of muscle activity within a particular part of the body. Yeah, I don't really know what the question is, but where's that? Yeah, without spamming you a hundred more metrics. Well, that is the problem, isn't it? I'm asking for simplicity, but then I've asked for more detail and more numbers. I guess rather than starting with the numbers and the metrics from an applied practitioner perspective, what are the questions I'm being asked right around training load? And that is, you know, questions around minutes. How, you know, how long should we train for? How long should we do these drills for? What kind of drill should we be doing in, again, American football? That's more rep based. So certain sports is minutes, certain sports is rep based. How do we design preseason training to progressively overload the different tissue types and help our players adapt so that they are exposed to optimal training load and be ready for the first game of the season? How much recovery should we give them? And then, yes, the rehab piece, obviously, is where we get more intricate and think about it. So, you know, right now we're answering those questions on training, planning and monitoring based on very global measures, as I said, of total distance or player load. And perhaps that next layer underneath, which would open account of world of complexity, but it would give us more knowledge of that training load planning process. Yeah, I'd argue we probably don't necessarily even have the estimation or prediction of the values that we would want before doing that. But I think it comes down to a really key point of establishing the causal framework or pathways here. Even if it is take player load and we're not coming up with any new metrics, but saying, can you predict injury risk from player load or can you predict some positive training adaptation from player load? I think we need to know what are we? What's our kind of pathway of player load correlates with this particular stimulus applied to the body? We're then saying that stimulus is going to correlate with this particular type of adaptation. And the more of that adaptation that you get, the more performance will improve or the more injury risk will increase or decrease. But even then it gets complicated with, I don't say we're talking about loading on a bone or a tendon, but we need to know not just necessarily an overall number for this session or for a week or kind of training cycle, but what's the relationship between the number of loading cycles and injury risk or the magnitude of those loading cycles or the recovery between them? So even something like the tibia in running, you know, what's the relative importance of how many strides you take versus what the load from each of those strides is? And even the recovery period between those phases, you kind of need some kind of model of how does this all tie in with injury risk? But then again, that's possibly getting into the academic side of overcomplicating things. We kind of want all of that to then work backwards and say, well, which bits actually matter? Can we predict that endpoint from the start point and just take out the bits in the middle? I don't know, but I think it's just, yeah, it is challenging and that you want to know what's the load on this particular part of the body. But on its own, that doesn't really tell us a lot when we talk about validity. It's valid for what, what kind of something isn't just valid or invalid. It's valid for a specific purpose. It's completely feasible that a training load metric can be valid for one application and invalid for another, or even then valid and invalid. Isn't this dichotomous kind of yes or no? It's how valid or invalid or what's in what context. And based on what we said earlier, how accurate does it actually have to be? But to try and come back from this over complication, I think the key point really is for us to maybe work backwards from speak to practitioners and what do what do practitioners want and how accurate does it have to be? So if you said, I want an estimate of load through the Achilles tendon during a run and it only has to be accurate enough that I can compare 10 different rehab exercises and see which ones are loading it more or less or correctly rank those exercises in order or compare to data last week and see if I'm progressively overloading that tissue. Kind of if we work backwards from there and then say, well, what would you be willing for the athlete to wear? Can they wear 20 sensors on every segment of the body? Or are you restricting it to just one GPS unit? Or are you restricting it to something on both boots and the GPS unit? And I think, yeah, establishing what are the constraints that we're working within? And then from there, you can start your kind of research process which might come down to, like, can we predict whole body? Even before we estimate load, can we predict how the whole body is moving from some sensors worn on just a couple of locations? There's been some really nice work recently looking at pressure based insults in a boot or a trainer. In combination with an IMU on the shank or the foot and saying, if you put those two together, we can estimate load through the tibia or Achilles. You know, you've got two sensors. Can you predict the rest of the body? Or are we looking at once we can predict those, can we predict the load on specific tissues that those movements result in? I think it's coming up with specific research questions that actually benefit practitioners and keeping that dialogue throughout. So you're not working in silos, but it's a collaborative process. And you come back and say, don't do that, because even if you do manage to do it, we're never going to use it too many sensors. Or, like, why are you hovering to get it that accurate? It doesn't need to be that accurate. I guess the last thing for me is just a massive thank you, because regardless of what anyone else thinks, I've got a lot out of this. And it's not the first conversation we've had, I've really enjoyed it. Where can people find out a bit more about you or you've mentioned various videos and blogs and things that you've written or filmed during this conversation? Where can we find those? And as fellow academics, where can we continue learning from practitioners like yourself? My company, my website is globalperformanceinsights.com. And there's my blog on there. And then that's the name of my YouTube channel. And then I think I'm at Joe Club Sports Cy on other channels. I think we've got so much to learn from each other. So I love these conversations around, OK, research, applied practice, how do we bring it together? So thank you for the conversation. No worries, yeah, thank you. Again, this has been quite selfish from my perspective, and I just want to chat to Joe and kind of discuss these things. But if anyone else does have any questions for either of us or kind of, I guess we were going to say at the start of this and I probably forgot that I don't think there's a claiming to be experts in any of this area, but it's just a case of trying to learn from each other where we can and discuss where should research go in the future? But if anyone does have any ideas or even anything you want to refer as to where you disagree or think, oh, have you seen this thing that actually we're closer than you thought? Or here's a really cool application. What do you think? Just kind of even getting in touch with either of us or leave a comment below the video. And I'm sure we'll both be really happy to do some reading and brilliant. Yeah, thank you again, Joe. And yeah, I guess I'll leave it there for now.