Evidence Strong
How to measure clean performance in weightlifting - with Angela Sorensen
there's several variables that are reliable and correlate to performance of cleans at like 90%, which you can relatively safely say is, you know, heavier load. 90%, you start under approach, the higher end of people's abilities. So with that, we wanted to make sure that we investigate everything from the floor up until, you know, basically peak bar height, essentially. So the entire duration of the pull, which would also kind of include the entire duration that the bar is moving upwards. Okay, Angela, welcome to evidence strong. Please introduce yourself. Thanks for having me. I am Angela Sorensen. I am currently a PhD student at Middlesex University in the London Sport Institute. After a lengthy career in strengthening conditioning and a lot of that time also being spent in weightlifting, I decided the next logical steps would be to get a master's degree and then take that master's degree on into a PhD. I am now in the final throes of that PhD investigating kinetics and kinematics of weightlifting, specifically the clean, just kind of went from the practical side. Now we're going into the academic side of weightlifting. That's super exciting. So we will be talking about your recent paper intra and intraday reliability of weightlifting variables and correlation to performance during cleanse. And right from the start, we have to talk through a very important thing you did in your study. You have named the phases differently that we would usually name them on the practical side. So why and how? So if you actually dig into the literature for weightlifting, the original way that, especially in research, the original way that the phases were defined were according to changes in vertical ground reaction force. So if you're looking at a force time curve, if you're looking at that curve, the start of the pull is when you basically created enough force to overcome gravity and the bar actually comes off of the floor. So that's the start of the waiting one. And then as the bar transitions past the knee, the force tends to come down and dip below that. And what that horizontal line is is system weight. So your weight plus the body, your weight plus the bar weight. And then in that second pull or the waiting to is when we get that nice spike in it. So these were all created back in the late '70s, early '80s by Roger and Noka. And those phases tend to hold true. But also through literature, a lot of times it's easier to define the phases according to joint angles. So that's where you start to get that first pull, the transition, the second pull. But being in the lab, we're on force plates. So we define them according to those original definitions, those changes in vertical ground reaction force. So that's where you get the waiting one, that slight dip as you pass the knee is the unweighting. And then the waiting two is where you finish that nice big aggressive extension. The phases, the waiting one, unweighting, waiting two, and the first pull transition second pull can be used somewhat interchangeably. It's a little bit tricky because when the unweighting starts, the bar isn't quite to where you would say the end of the first pull is. But from a practical standpoint, yes, you are correct. It's a little bit easier in the gym to call them first pull, second pull. You know, that's the pretty common nomenclature now. -One more thing, before we go into the study, you mentioned the system and the barbell. Could you elaborate a little bit on that? -Yeah, so system weight, pretty straightforward. It's the body weight plus the barbell weight because yes, we are lifting the barbell, but we have to work as one unit. So knowing how that unit functions together is going to be super important. It's nice to know like, you know, what's the speed on the barbell or, you know, what's the power output, but it's also important to understand that in the context of you and the barbell are essentially one unit, you're working together, you have to work together to move that bar. So it's important to think of it as a system rather than individual components, really. And I think that's kind of overlooked sometimes by some coaches. We have to work together. It's not just you versus the barbell, it's you plus the barbell. -Okay, very sense. You guys, so I think we are ready now to ask you what was the aim of the study and how you went about doing it. -The aim of the study is really to almost take a step back from some of the current research. A lot of the current research is based on weight lifting derivatives, stuff done from the hang, you know, powers, you know, anything and everything except for like going from the floor into a full lift. But it was kind of like a way to peel back the layers and find out if there's something that we've been missing. So we took a stab at it by seeing, basically, let's calculate all the possible variables. Let's, you know, let's see, one, are they reliable? And then if they're reliable, do they actually relate to overall, like, clean performance at heavier loads? And as it turns out, there's several variables that are reliable and correlate to performance of cleans at like 90%, which you can relatively safely say is, you know, heavier load. 90%, you start under approach the higher end of people's abilities. So with that, we wanted to make sure that we investigate everything from the floor up until, you know, basically peak bar height, essentially. So the entire duration of the pull, which would also kind of include the entire duration that the bar is moving upwards, basically. So we started out with 70 variables. And then we had a, we had a pretty strict criteria for reliability. The CV on all of the variables had to be 10% or less. So like, some very, so the coefficient of variance variation, always get there incorrect. So basically, how much does that vary across, you know, all of the testing conditions? And 10% means that you're a lot closer to what you're saying is true. Like, there's not as much variation. If you have, say, 20% variation, then that's not as reliable, because that's a lot of chance for that variable to change. You know, so that was one of the criteria. And then we looked at other reliability statistics. So we did the interclass correlation coefficient. And according to a paper by Ku and Lee, if you calculate the ICCs and you actually take that 95% confidence interval, so now we're getting into the kind of the boring statistics, but that's what this paper is. It's a statistics paper, basically. But if you take that 95% confidence interval and you take that lower bound, so the lower number of that 95% confidence interval, all of our statistics, all of our variables had to be 0.75 or above. So we wanted to make sure that even at their worst, which would be the lower bound of that confidence interval, even at their worst, they're still considered good to excellent reliability. So we had some very strict, you know, so like, even though we started with 70 variables, we narrow that down to these are very specific. They don't vary across the variability within their performance isn't massive. So like, you know, they're going to be reliable. And then we took those 11 that we found, I think we ended up with like 16, and then we wanted to make sure that, okay, that's nice if these are reliable, but do they actually relate to performance? And so we did a Pearson's correlation coefficient, and we found that 11 of those variables are super reliable, and they relate to performance of cleans at 90%. So like, you want to make sure that when you're working at those higher loads, that the variables that you're kind of looking at are indeed telling you what they are meant to tell you, which is what the reliability is, and that they're actually going to relate to the performance outcomes that you're looking for. Needless to say, my spreadsheets were massive from an academic standpoint. And we solved this problem in the paper. It's a bit more challenging to obtain that system weight that we were talking about, the lifter plus the barbell. So to be the most accurate with getting that system weight, because you need that system weight to help calculate some of the variables and the changes in the phases, you need to have that lifter and the barbell standing motionless on the force plate. Well, off of the floor, especially if you're going heavy, they have to pick that barbell up and hold it and not move. So if you're at, you know, heavier loads, that's going to start fatiguing the lifter. So with weightlifting derivatives, you know, it's not that hard to pick it up off of blocks and hold it at mid thigh level if it's already at knee level, you're already minimizing the fatigue. What we did to solve that problem in regards to obtaining system weight is we actually handed the lifter, the lifter stepped onto the force plates. We handed the lifter the barbell for them to remain motionless, and then we took it back from them. So it's the lifter got to work out and myself and my colleagues as investigators, we got to work out because, you know, we tested a lot of lifts, you know, they did a lot of lifts working their way up to 90%. So picking that up, hold it. All right, we'll take it back. So we got really good feedback on that method. Important to note for people is that the force plates usually are not big enough to accommodate lifter and the weights on the side. So usually the lifter is standing on two separate force plates, but the weight, the barbell, the blades are outside. So that's why the whole issue of obtaining the weight of the system is existing. Let me know, please, who are the participants and how many have you tested and how many times all the setup of the study? These participants were from a local weightlifting club and they were all competing on, you know, a regional to national level, you know, so they're all experienced weight lifters. The coach is like a national international level coach, you know, I myself am a national level coach as well. So, you know, their proficiency was there. Unfortunately, we had a small sample size. We had eight lifters, but given our parameters of them having to be proficient in weightlifting, having experience, you know, and competing on that regional to national level, which is kind of based on being able to hit qualifying standards. So yeah, we just recruited them from a local weightlifting club. And then each of the participants came in and completed three testing sessions over the course of a week. And they had, so we basically, we tested everybody on a Monday, Wednesday, Friday, because we wanted to give them enough time in between each of the sessions to fully recover. And we tested them three times because, you know, as the paper says, it's an intra and intraday reliability. So it's great if they come in for a one-off and all right, you know, okay, cool. Within a day, this is reliable and relates to performance, but like, does it relate across multiple days, you know, because training in a practical setting, training is not a one-off you do it, then boom, we're competing, you know, you consistently, we need to know how these variables react. So, you know, with all of that robust reliability criteria is reliable within a day and relates to performance and is reliable across multiple days and still relates to performance. So these wonderfully patient athletes came in, these lifters came in on a Monday, Wednesday, Friday. They came in at the same time to make sure that we're not having any changes, you know, in their failure effects, you know, they weren't training beforehand. We wanted to make sure that they had the same warm-up. So we did the same warm-up, we had the same barbell warm-up, and then every lift was tested. We did three sets of one repetition, so they did starting at 50%. They did the lift, brought it back down, okay, take about 30 seconds rest, do it again, and then they were given, you know, two minutes between loads, but then when we got to the heavier loads, when we got to that 90%, we extended that rest interval up to a full minute, which is kind of in line with how you would do it on competition. If you're following a lifter, it's about a minute. If you're following yourself, it tends to be more like two minutes, so you want to make sure that, you know, you're kind of getting somewhat of that competition by even though you're in a lab, and it's really hard to recreate a competition vibe in a lab. They did 50, 60, 70, 80, 90, three sets of one at each of those loads, so we had a pretty good, pretty good range of weights there in terms of, you know, how many lifts they were able to perform over three days, and this was all based off of their one repetition max clean and jerk, so we took it off of their clean and jerk, because obviously we're talking about competitive weight lifting, and you're not just judged on your clean, you're judged on your clean and jerk, so we were kind of basing it off of that, and they had obtained their 1RM within two weeks prior to commencing the study. It was just kind of as part of their regular training program, and they were just come in and lift heavy three times and then call it a day. Okay, so you mentioned before about using force plates for measurement, can you go a little bit deeper into what, where are you measuring? We calculated, you know, the force time variables, so that's going to be some of those changes in vertical ground reaction force, but with all of that, then we were also able to calculate some of the system variables and some of the bar power variables, so you know, we were able to actually see what the bar was doing, and you know, we ended up with those 11 different variables, could list them off if you'd like me to, if it makes sense to anybody, but we basically ended up with those temporal force, so the time and the force, and then the bar power, which is basically what if you've got, if you've got a set of force plates, you can do all of that, some of the original studies kind of did it by hand, because they knew what the weight of the barbell was, and they knew what the weight of the lifter was, and then they knew how long it took to do all of that, we just luck out, because we can get all of the force and everything calculated directly, because we've got those force plates and that they're lifting on them. Okay, so you tested 70, got 16 that were reliable, so they were the variance between one and the other, from one left to the other was pretty small, and then out of these you had 11 that correlated with clean performance, so what do you mean by clean performance? So we ran the correlation compared to the cleans being performed at 90%. How does each of these 11 variables, how do they correlate to the actual, just the full performance of the clean, like just the clean itself, and that's where like 11 of those variables came back as saying like yes, this directly, we have a really good strong correlation to how that clean is performed, and we went so deep as to run a multi-colonarity statistic on that, meaning that there's a massive matrix in the paper, if you look at it, of all of those 11 variables, some of those variables are actually going to be reporting the same thing, so like for example, if we take the average bar power of unweighting to peak bar height, obviously it's going to be reporting something very similar to average bar power from liftoff to peak bar height because it includes all of the same phases, but taken in the context of like a performance setting, some of these could actually have some importance to a coach in a practical environment, so you know we can track, it's easy to track bar power from the point of liftoff until peak bar height because if you've got your handy little phone app, you can tell at what point the bar comes off of the floor, and you can tell at what point the bar hits its peak, but as we were kind of talking before with some of the derivatives, you know, knowing that the average power from the unweighting until peak bar height, so basically like a hang, if you will, you know, that it's also reliable and correlates to performance at those heavier loaded cleans, so you know, you kind of have to look at it in in the context of what you're needing in a performance, like it's interesting from a research standpoint, but it's also, you know, how does this apply in a practical environment? With all the false ones, coaches don't have access to false plans, so they cannot see how one is pushing, but they can see other things, are there any of the variables you have found easy to see enough to be useful in daily training? So some of the variables that we found are easy to track, like the average bar power lift off to peak bar height, that's a pretty easy one to track, but with some of the forced ones, especially the ones that we found, they included the waiting ones, so that first pull, what that means for the coaches isn't so much that, you know, they need to be able to like monitor it, it's more that they need to know that the application, like force application, in that first pull, that in that waiting one, in that initial lift off phase, is super important, so if you want to improve performance, you're going to need to do, and you're using weightlifting derivatives or weightlifting performance from the floor, you need to get stronger in that, that's what those forced variables mainly mean, you're not necessarily going to be able to track them, but you're going to need to know that a lifter needs to be stronger off of the floor, it's super important that they need to be stronger off of the floor, because that sets the stage for every subsequent phase, but also that ability to generate force off of the floor relates to overall clean performance in the end, but with something like the bar power variables, like peak power, those are something that can be monitored through the use of the tracking apps on your phone, you know, and as we kind of keep talking about the lift off to peak bar height or the unweighting to the peak bar height, those are those are points that you can definitely track through there, but the big thing with the force is that basically, as we kind of know in general, but sometimes need reminded, you need to be strong, especially off of the floor, so like one of my favorite variables is the average resultant force, and the interesting part about that is that one was reliable during the waiting one, unweighting and waiting two phases, but basically the entire duration of the pull, average resultant force is important, but to discuss what resultant force is, resultant force is basically the product of your vertical force and your horizontal force, so if I have a large vertical force and a tiny little horizontal force, the resultant force kind of goes this way, so we're talking about barbell trajectories, that's important, yes, but if it goes the other way where I have a large horizontal and a tiny vertical, that bar's looping out, right, so that's why this one's so exciting to me because with this particular variable, it talks about not only is force generation important, but the direction in which you apply it, obviously in weightlifting, we want to minimize horizontal force displacement, we want to keep that bar over our base, we want to keep, we want to keep that system compact, we want to keep the bar and our body pretty close together, and this kind of confirms that, especially off of the floor, you need to be strong, but you also need to pay attention to the direction at which you're applying that force. When you put it this way, it's very intuitive, we always want to keep the bar as close as possible, but also generate force, so if we think about, okay, if I move the bar forward or backward, it will change the force production because now some of these force I produce have to be wasted to go forward or backward, so it makes sense. Okay, any other results you think are important to discuss, and then we will move into advice for coaches and athletes from you as an expert. Yeah, so I think one of the biggest take homes from the research is how, like we've basically kind of been alluding to this whole time, that how the bar comes off the floor is important, what happens in that initial lift off, in that waiting one, the first pull, whatever you want to call, that part where it comes off the floor, that's super important, you know, and that's kind of evidence in our paper because there's most of those variables started from the floor, so if we're looking at the temporal force variables, we had five temporal force variables, and three of the five were that first pull, that waiting one, the bar power, five of the six variables in the bar power started off of the floor, so, you know, that like, you know, how you perform the lift off of the floor is important. Good, your frame, now the results of the study and some pointers maybe you want to pay attention how the bar should go, I don't know, some things athletes and coaches can apply to the next training session. That's pretty simple, and it's pretty intuitive already, and what most coaches should already be doing is that, you know, weightlifters need to get stronger, you know, they need to be able to generate high amounts of force in short amount of time, so if you're early in a training cycle, then you're going to want to emphasize, you know, ability to exert force, you're going to want to emphasize like hitting maximal levels of strength, and then, you know, especially as it pertains to this study, you're going to want to try to overload and get stronger in that first pull, so this is where you can do, say, just like, you know, heavy loaded lift off, so like, lift off from floor to knee, so from like lifting just above the knee level and getting strong in that position. For more advanced lifters, you can do sort of the deficit lifts where you're standing, you know, slightly elevated, so you have to pull just a little bit longer, you know, and during that unweighting that lift off phase, and then, you know, obviously with bar power, you want to be able to speed things up, so as you transition from being able to go be really strong off of the floor, then as you progress through your cycle, you're going to want to emphasize speed of the pull, and being able to translate that force into moving with that level of force quickly, you know, so you're going to want to emphasize anything that works on that power development, so you're going to, you know, you're going to want to make sure that you're checking the loads, because it's fun to pull heavy, but, you know, if you're emphasizing power, you're going to need to move it, you're going to need to load the bar as such that you can still move quickly, you know, I've previously done some like, super max pull studies for potentiation, and spoiler alert, when you hit like 120%, you move slow, you know, so if you're looking for power to develop, you're going to want to keep things a little bit lighter, like as we keep talking, make sure that you're paying attention to not just that they're getting strong or that they're getting powerful, you know, that they're applying that force in the correct directions, especially as it comes off of the floor, if you do that, it's going to make your life so much easier. I know that you have other studies coming in up, would you be willing to tease them out a little bit for us? Yeah, my second study, again, as you know, as with any PhD one study sort of leads into the next, so now that we have all of these reliable variables, you know, now we need to know whether those variables, whether they change or how they change over various loads, so like, you know, if you're lifting it 50, 60, 70, 80, 90%, does it change? Does it matter if, you know, the 70 and 80% are so close to being the same, you know, those could be used interchangeably in like a training program, you know, things like that, you know, a lot of times for power, maybe, maybe some studies have shown that power is best at like lower loads, well, you know, let's see on these reliable variables where things lie on that spectrum, and then the most fun part is whether you can kind of change some of these variables through training. So like, if I send somebody, because we've been talking about force so much, if I send somebody through a force block, a force improvement, you know, if I just send them through a strength block, have these variables actually change? Like, have we been able to manipulate these variables by making somebody like super strong? Okay, cool. Now, if I send them through a power block, like, I kind of like in line with what you would be doing to leading up to competing eventually, now if I send them through, you know, something designed more towards maximizing power output, have those variables changed? So if we can determine that, then we can know at what point to apply some of these things when to emphasize or de-emphasize something, you know, if these things change, you know, if I go through a massive force block, put my trajectories all off, then that's a problem, more similarly for a power block. So that would tell us, you know, like, you know, yes, it's important to emphasize this, and you can manipulate it through a training and intervention working at like roughly this percentage for strength stuff and this percentage for power stuff. We've collected the data, so the results are to be determined, but I can tell you that we have a tremendous amount of data to work with, slightly larger sample size this time, so quite excited to see the results on that one. So just to finish off, I wanted to ask you why in the research? I think it's possibly because I'm someone who likes to kind of diversify in their skills and abilities, and I've done so much in a practical environment. All right, great, you know, but then also I kind of want to know some of like the underpinnings to what I know intuitively like, you know, and I could go read the research papers, but it's kind of one of those things like you learn quicker by doing something. So when it comes to research, you know, it's actually it's actually a lot of fun for as much of a headache as, you know, PhD stuff can be or like being hunched over a computer for hours on end. I've actually enjoyed it. I've learned a lot, you know, and that's going to, you know, influence not just like how I program and how I program myself or program my lifters or my athletes, but you know, I'm also I've also done a lot of work in coaching education, and it's going to influence a lot about how like I kind of impart what I have learned to everyone else, you know, especially now having read hundreds of papers, you know, as one does during a PhD, now I know as having been a researcher and got gone into it, now I know how to read those papers. So now I can, as we kind of mentioned, act as like an interpreter like, yeah, that's great, except for this or that's super they found this because they did this, you know, so like, you know, when we all did our undergrad papers were like, huh, I don't understand any of this, but like, now I'm like, okay, I get it and this is what it means for the average everyday coach, you know, so they don't have to go in and, you know, figure out what it means, I can help interpret what it means. So it kind of helps with everything. I get to do a little bit of coaching, I get to do a little bit of researching, I get to do a little bit of educating. So it's definitely something that I'm, I'm glad I decided to do jump into weightlifting research. Okay, so for your next study, I have to know what is your favorite color? Favorite color? I'm like a green person, like maybe like a forest green or possibly like a teal turquoise aquamarine type color. So your next study will get an infographic with green tail or turquoise. Kind of color. That's beautiful. I love it. Where people can find you? People can find me on Twitter, which is where most of my research is. It's, well, I'm, I'm not on there a lot, but that's, that is where I share my research. It's @&sorensen. So A-N-G-E-S-O-R-E-N S-E-N. You can find some of my everyday banter on Instagram @Beesus Athletics. So kind of like a lioness, but I'm a beastess. Beastess Athletics, feel free to reach out to Alex and she can give you my direct contact info. We can have a chat anytime. I'm happy to speak to anyone about any of this or, you know, discuss ideas or, you know, just have general weightlifting chit chat. Thank you so much for making time for this conversation. And it was a long awaited and I'm super happy that we had this opportunity. Thank you. You