Technologies for Swimming Performance Monitoring: Race Analysis and Profiling | Bjørn Harald Olstad
welcome back to uh Ty Motion and the strengthening numbers webinar series uh my name is Peter Holms i'm um heading up tenny motion in the United States and uh I am so uh grateful today to invite Vjernh Harl Ol from uh Norway in Oslo um you're working there at the uh Norwegian School of Sports Science uh doing research and um all kinds of stuff uh and a lot of focus on swimming so welcome to the show Bjornhara thank you very much Peter delighted to be here thanks for inviting me no it's great to have you on board and uh so thank you so much for preparing for this and uh what I know will be a very interesting presentation on swimming um and how to monitor or utilizing u uh technologies for for swimming performance um before we get started I only want to remind people that you can uh type in your questions there's a little uh toolbar on your webinar uh screen uh where you can uh there's a little you click a button there and you can then uh type in your questions and we will make sure those are answered what we're going to do is to uh do a separately recorded Q&A session uh just to make sure we utilize this time well and then we're going to record that uh afterwards and publish it online together with the recording of your main presentation uh also this is our uh last session in this series that we have scheduled and and run here during the spring uh uh we're so grateful for all the speakers that have uh you know spent a lot of time preparing uh we started with Matt Price from the LA Kings in early April and we've been running a session every week since um now we're going to take a break um the for a number of reasons uh primarily is that people are starting to get busy again uh the COVID 19 situation is of course not over by any um you know by it's continuing and it's a serious situation for for everybody uh but people are starting to uh get back into some kind of normal sports are starting to to happen again and and we notice that people are getting busy so we we're going to take a break with the webinars and and plan some new sessions down the road and let everybody know uh when that will happen and uh the interesting topics that we will we will um we will schedule with that thanks again Bjorn Horald i'm going to let you take off now but why don't you uh tell us a little bit more about your background and kind of where you're coming from before you dive into Thanks Peter well thank you for the kind introduction too i now 46 years old but uh been a competitive swimmer for yeah most of my life even though I started quite late as a 12-year-old but still it was not too bad in terms of making it to a fairly high level i lived in the states for quite many years came there to swim in college as my big dream finished my collegiate career but didn't feel like going home so continued to study masters and started as a swimming coach for college and club team then back in 2002 I was uh appointed by USA swimming headquarter up in Colorado Springs and got to work with the national team in terms of performance testing and education very exciting times and also at the beginning of Michael Phelps's uh career and after 8 years in America I decided to come back to Norway and then I started to work as a performance director and a daily manager for a swim club here in Oslo and then afterwards I transferred over to yeah the Norwegian school of sports science in 2009 where I did my PhD in swimming on muscle activation and kinematics in contemporary breaststroke swimming and together with uh many of my good colleagues here we are doing a lot of applied research in swimming with a special focus on performance wow you have an impressive background you've seen a lot in a lot of places um that's I've been fortunate yeah thanks for that background i'm going to let you um you know uh move forward now with your presentation i'm going to go offline here uh or off mic and video and leave you with with the audience so again welcome everybody and thank you Bonjora for being here thank you and welcome everyone so we have uh quite an extensive list of topics today because when our school got rehabilitated about two years ago we were fortunate to acquire a lot of new state-of-the-art technology that we try to implement and into our applied swimming research so I hope it will not be too packed and I hope we get through most of it and in case something is or in case we're going over time we will take this back in the Q&A session later but the two main technologies we will spend the most time on is the race analysis which is a video camera system with automatic movement recognition and it's the velocity force and power as well as drag measurements that we get from a portable vinch system and a land strength system we will go into some muscular activation with surface electromyiography and then we will try to connect the dots to daily and live monitoring and feedback then during training in terms of stroke kinematics from inertial measurement units and heart rate from optical heart rate sensors but first I would like to show you this uh slide which comes from professor Buas in Portugal and it's basically showing their approach to analystic approach to swimming research and looking at this figure you can see that there are quite many factors that are determining for the performance and the reason I would like to show this one is mainly because with the technology that we were able to purchase we can now measure not all but almost all of these different variables so if you go to the middle of the screen into the yellow square saying race analysis which is then considered the optimal performance test I mean it's the race itself and as we know in swimming the winner is the fastest swimmer but in order to have a great race performance all of these other factors play a role so then if you go to the bottom green square we find physiology energetics anthropometry biomechanics and motor control many underlying components there we cannot measure all of them with our technology but at least we can measure one or more so technology gives a lot of possibilities in terms of determining and monitoring performance so let's start with the race analysis technology it's basically an integration between two systems it's the AIM athletes in motion camera system and then Swiss timing which provides the automatic timing from the touch pads and from the starting block so the camera system itself has 11 cameras five you will see placed above the water surface five are beneath and they are all fixed cameras five meters apart and the 11th camera is also underwater at the middle of the starting lane at the starting site so it provides you with a back and front view we also utilized then this uh big sports specific scoreboard that we have in at the pool and also a portable monitor when we are then utilizing the cameras in individual mode and it's mostly then for students and the swimmers to swim a lap swim a certain distance and afterwards they can look up and thanks to the delay function they get to see themselves so what do we then get from this race analysis system so if Peter would be kind to play the video I don't know if you guys see the video but I guess I probably don't all right so I think Yeah it's going oh here it comes so basically what you see then is a split screen from the above and underwater cameras and it stitched them together forming an almost seamingless uh movie in addition on the right side you find now the left uh quite some of the important stroke matrixes you have find the instantaneous velocity you will see the hip and head depth the stroke rate all pinpointed to the location and time for the race and in addition you will have the block time the turn time the brake time brake distance cycles and of course the lap time so in addition to this video we also get the graphical representations of the race and sorrying uh I think you have to accept become the percent in there so we can see your screen that means that it probably went out sorry click start yeah now you're back back oh it was back for a split second there but it disappeared again okay sorry now it should be back hopefully now it's back thank you yeah so yeah so here you then see the graphical uh representation of the race as well as numerical uh numbers and as a coach and a swimmer you get a lot of fancy data how do you utilize it do you just go blind on all these numbers or how do you start to work with it but of course I should also say in addition to what we get from the system we also use MATLAB to further programming so we can go even deeper into the race itself when we want to look at the performance deciding factors but basically we used to look upon this as the normal way of doing race analysis you take the swimming race and you break it into the free swimming part which you call often the clean swimming and it consists of the stroke rate and the stroke length and the product becomes the swimming velocity on the right side we find what we call skills which is the start often defined as then zero to 15 m the turn often from 5 m before the wall to 10 or 15 m after guess depending on the stroke and the underwater distance they are traveling and the finish which is normally the last 5 m but of course during the start the turn many other things will decide how fast you are for example the start what is your takeoff velocity what is your reaction time how long do you glide when should you start your first underwater motion when should you break out and similar goes for turn also including the pivot so we usually look upon this from four perspectives and one is segment contribution like we identify the contribution of each segment to the finishing time it's not a very sophisticated way of looking at it but it's always nice to go back and revisit this topic because it puts things into perspective how much time do the swimmers spend in a different segments in a race here is some examples of three different uh studies that we did we will not go into all the details in the results and I guess I forgot to say that in the beginning i will present quite a bit of results but I don't think we have time to go deep into them so I would like you to then leave a Q&A if you want to hear more or please feel free to contact me at the after the session with the email provided at the end but one example is from the men's 100 meter shortc course breaststroke so here we looked at the six best swimmers in Norway at the championship final they came to our pool the day after for testing in the race analysis system and what you can see maybe not very surprising either is that the turns compile the largest part of the the race but what I think is very interesting is when we look at the break time from starts and turns and here it's the individual measure of the swimmer meaning when the head breaks the surface after the start and after the turn so it's not really related to the clean swimming at all more than 38% of the 100 meter shortc course breaststroke is related to either the airborne phase or the underwater phase so what we then like to ask the coaches and swimmers when it's so important or contribute so much to the performance time how much time do you actually spend on this practicing it with 100% focus during your daily training if you go over to the men's 50 m short course butterfly we see that the contribution is changing quite a bit the blue numbers indicates that it's increasing from the 100 meter breaststroke race and the red that it's decreasing so of course the start will here be more important in terms of contribution because it's only a 50 m race but also here we see that uh the start and the turn break time contributes to almost 35% of the race but also more interestingly here is that look at the standard deviation which is quite large 6.55 and again that shows us that elite swimmers I mean these were swimmers also qualified for the national championship they have quite different strategies in terms of how long they stay underwater in their underwater kicking and we will not further go into the women's 50 meter shortc course front crawl but you will have the numbers on the screen component contribution And this is probably maybe the most interesting part in that sense because which segments and the components in these segments are determining the performance so again going back to the men's 100 m breaststroke study we found a nearly perfect correlation for the 15 m start time with the finishing time and saying this very simplistically this means that the swimmer that was number one two and three after the start was also the swimmer that was one two and three at the finish meaning they could have stopped the race given out the medals and the race would have been over of course in real life it's not that simple because we have to swim the race and things can happen but what this shows us is that what you what the other swimmers lost during the start they were never never a able to require after the start so then we were curious what part of the start is crucial so then we broke the start into five meter segments the first five the middle five and the last five meters of the start and here we can see that the last five meters the 10 to 15 m mark showed a nearly perfect correlation together with the starting time so then it's important to go further into that part of the starting segment what constitutes the important factor there and here we found that the peak velocity showed a very large correlation with the 10 to 15 meter time and also we see here that the peak velocity has a range of from about 2 m to 2.5 m/s and this is then generated from the underwater arm pull which appears to be the crucial part of the starch how how high peak velocity do you actually get from your armpool and it can of course be related to are you gliding too much before do you generate uh velocity from your underwater dolphin kick or are you simply not enough strong enough or have good enough arm pull technique to generate this high velocity we did the same thing in the 50 m short course butterfly where we also looked at start and turn performance and maybe not surprisingly the mean underwater kick velocity has a large impact on the 15 m start and the 10 m turn time and again since butterfly has the rule that you can kick 15 mters underwater most of that part will anyway be underwater but we can see that from the start it had a point a negative.689 correlation but almost a perfect correlation for the turn
but what was interesting here is what comes in red that this was regardless of the initial velocity meaning that the velocity that the swimmers entered the water from the start or from the push off during the turn had no influence on this mean kicking velocity at the same time too the velocity deceleration from the underwater kicking did also not have any influence on the underwater kick velocity and I mean expecting that having a high peak velocity going off the wall and off the block is important it was kind of surprising and if you look at at the first kick velocity it appears that that this one is essential for the mean underwater kick during the start so what this means is that the fast swimmers are starting their first underwater dolphin kick with a higher velocity than the others and this can be related to the body angle when they enters the water like some swimmers might have a pretty steep angle traveling a bit downwards before they are flattening out to have their horizontal body position and of course others might glide too long but to start the first kick with high velocity is important for a good start but in the turn contradictory well not contradictory but oppositely here the last kick velocity show to be very important and again when you push off in a turn you are almost more or less horizontal so we don't need to be so good in manipulating your body angle so here it's much more the underwater kicking technique itself to then maintain the high velocity and come up with the breakout on a high last kick velocity and then one last example of this because I think this is very interesting and it's not coming from our race analysis system but it shows how important this is for performance this comes from the world championship in 2018 and not much race analysis has been done on the long-d distanceance event so this was uh done on the first second third and sixth place finisher in the men's 1500 meter front crawl and if you look at the figure and towards the bottom you have some red and blue numbers we see here that the winner of the race the first place finisher had a total turn time of 295 seconds and a total swim time of 553 seconds and what's very interesting here is that you see the second place finisher this person swam about 4 seconds faster than the winner of the race but the turn time was 5 seconds slower meaning that this person lost the gold because of a slower turn time and also to make it precise the turn here is classified as 5 meters in and 5 meters out even more interestingly when we look towards the right for the third place finisher and the sixth place finisher here we see that the fifth place no sorry sixth place finisher swam roughly 7 seconds faster than the bronze medalist but this swimmer lost actually almost 15 seconds on the turn leaving this swimmer in sixth place while the other swimmer got third place thanks to an excellent turning performance which is almost identical to the winner of the race so this is just an important uh perspective in terms of why race analysis is important for performance another way of utilizing race analysis is to use race modeling which I think is a very practical and applied way of using it so here you can compare yourself to the gold standards you can compare yourself to others and to yourself so again going back to the 100 meter breaststroke study we did a race modeling on top you will see the fastest swimmer in this study and below you will find the fastest time so regardless of which swimmer had the fastest time that's what it's plotted against so when you see that the fastest swimmer has a green box what then the swimmer was within 0.099 seconds of the fastest swimmer meaning this swimmer is good at this yellow between.1 to.199 seconds and we classified red if the fastest swimmer was more than 2 point more than 0.2 two seconds slower than the fastest swimmer and we can see that the fastest swimmer is quite good in many of the aspects of the race but one particular parameter the swimmer scores quite poorly and that is the pivot time and in breaststroke we give pivot time from when the hand touches the wall until the feet touches the wall so by looking at the time that this swimmer spent here this swimmer could have swam swam almost 1 second faster if he was also the best at the pivot time so it's quite something to think about so it's important to use race modeling in that sense to see where are you scoring well and where are your main factors for improving performance and one last way we can use a race analysis is to compare yourself after training so basically you come to the test you get your race analysis you identified some parameters you need to work on and then you go back and do it so in this example from 50 m front crawl we have the left and right stroke distance and when this swimmer came to the initial testing we see that the average distance from the left arm was slightly above 1 meter looking at the right arm it's below one meter it's quite a difference between these two so then the swimmer went home to practice their distance per stroke on the right arm in terms of improving it later they came back to from for testing and here we see they're almost equal both more than one meter per stroke or per arm sorry again leading the swimmer to swim with a longer stroke distance some future perspectives or directions that we would like to go in terms of race analysis is to look even more into start and turns and from the perspective of how to optimize the underwater work when should you time the dolphin kick how do you time then the dolphin kick to the arm and leg motion in breaststroke the velocity fluctuations the decline and what is the ideal breakout point what should your velocity be compared to your swimming velocity when you break out from your underwater work in the four competitive strokes we would like to implement the inertial measurement units because then we could get even more detailed stroke analysis and of course it would be nice to do some intervention studies in terms of effective training on race analysis parameters and performance and also it would be very nice if we could include force measurements in the starting block and the turning plate to see how the forces are distributed throughout the start and the turn and that takes us over to the next topic inertial measurement units and IMUs we will not go into detail in terms of what an IMU is but on the right side here you see the IMUs that we collected throughout this process and you can see it's a very small and light IMU weighing roughly 5 g and also with very small dimensions but of course swimming is in the water and we have many challenges with technology in the water here we have something that we need to be aware of when using imuse in the pool and again we will not go into the details about this you can go to the references if you want to read more but I guess one big challenge is usually how you seal the sensors how do you keep them waterproof luckily enough the sensors we have are waterproof so we don't have to worry about that but again there is still no consensus yet in the swimming world about where should we place these immuse for collecting the different types of data we want and of course in swimming when you have the drag from the water it's quite different than doing on land exercises so of course if these IMUs are significantly in size and weight they will imp impose a drag on the swimmer and of course again the feel for the water might be altered depending on where you place them and of course again the size too but if you look at the commercial market today you find many of these so-called IMUs or at least included in technologies like sport watches or other ones that can track quite a bit of stroke metrices on the left side here you see a system called Triton and it's the one that claims today to be the most accurate and if you look at some of the validity measurements from this system this was then compared to video recordings you see that the system is uh quite or very good in terms of coefficient of variance being lower than five or six for split time speed stroke count and stroke rate but on the other hand when you look at distance per stroke the time the swimmer was on the water and turn times is quite a high coefficient of variance so in that sense IMUs are starting to become available they're also starting to be more and more accurate again depending on the purpose can they be utilized or not here's an example of a collaboration we did with some researchers in Portugal and it kind of shows the possibility or the future possibility of using IMUS because here we were trying then to develop or create algorithms to detect more stroke kinematics besides what we would get from the video capture at least in 2D and here we have then the body rotation which we know is quite important in especially front crawl and backstroke like how much does the body rotate in the longitudinal axis on the other hand too you can have excessive rotation and of course asymmetrical rotation and of course in breaststroke and butterfly you want this rotation to be minimal more or less at zero trunk elevation on the other hand is interesting in terms of butterfly and breaststroke when you're looking at the undulation technique meaning how much body undulation do you see in the stroke and if you look at especially breast stroke over the last uh decades it has gone through many changes from first being a very flat stroke then a very undulating stroke and now breaststroke is still undulating but more towards the flat and also in butterfly I mean we want to move as quickly as possible horizontally so again excessive up and down movements will slow the swimmer down and finally we were also looking at the body balance or what we call pitch angle and again we know that in front crawl and backstroke we should be as horizontal as possible to have minimized drag and of course if this angle increases and the legs are going down the drag will increase so here's just a way of how to utilize IMUS down to the right you see a movie from Triton and they have placed their sensor on the back of the head down to the left you see me trying our IMU for the first time doing butterfly but when it comes to IMUs I think it's the missing piece to swim coaching because it's the way of providing instantaneous feedback on stroke metrics to the coach and to the swimmer as you can now see in the movie down to the right you can have the live streaming out to an iPad or to a monitor in the pool because for everyone who has coached swimming having 15 20 swimmers in the pool at the same time it's quite complicated to follow each one in a very detailed manner many coaches are left with a stopwatch providing split times the occasional stroke technical feedback of course swimmers can count their own strokes but it's very limited how much you can get out of it so of course if something could track the swimmer's daily training and then continuously monitoring this data over long periods you will get a great tool to see first and foremost are you improving on the parameters that you want to improve and also in other cases you want to be better in underwater kicking the coach and the swimmer has an agreement you should do six underwater kicks after each turn the swimmer is a little bit tired today so they cheat on the opposite side of the pool when the coach can't see it doing only three or four kicks then of course if you have this IMU daily or live tracking this the coach can say hey you're only doing four kicks today why so basically the point is I think we have a lot more to offer in terms of stroke technique and improvements by having this continuous data and I don't think we are quite there yet but I think we are moving quite rapidly forward in terms of being able to utilize these IMUs in a full possible way so what I think would be interesting in this case is for us to then integrate this with race analysis and then we could acquire additional parameters such as the body roll and the undulation you can look at the pitch angle and also you can then divide the arm stroke and the leg kick into different phases to see actually what is occurring during these phases when you swim of course instantaneous velocity is a very nice parameter from race analysis we have it but not from the daily training because I mean it is the best uh estimating of swimming performance and if you have very large velocity variations when you're swimming there are par there are things in your technique that you have to go and investigate and of course what would also be interesting is then if you can take the IMU to combine joint angles i mean today it's very tedious and laborious of course with 3D kinematics and automatic tracking it's quite much or quite doable but not from the normal 2D and the normal video recordings that we see so in that sense we can then go into the armstroke faces the interarm coordination and of course intersegmental elbow and knee angles cycle per cycle and of course then you can see how how consistent is the swimmer cycle burst per cycle do they have an optimized automatic stroke pattern or do they have a lot of stroketo-stroke variability strength and power land and water technology so to the left you see the 1080 quantum and on the right side you see the 1080 sprint two technologies that we also acquired during this uh rehabilitation process and why is this interesting for us well one reason you see right here the swimmers have changed through the last 100 years i mean looking at tarsen to the left Mark Spitz in the middle and then to the modern swimmer it's clear that they are now much more powerful and they're physically stronger than ever before and also when we look at the way swimmers train today compared to like for example in the 80s many many things have changed i mean back in the 80s the purpose of the land training was to build strength endurance that was the crucial part and if you look at the the highlights in red you see that one repetition maximum with unspecific velocity not meaningful and also there are low maximal strength requirements in swimming looking at today's strength training now it's development of maximal strength that is the basic and you use methods with high to very high loads and I think this quote from Adam Pedy summarizes it quite well a swimmer's physique is built in the gym not in the pool so then we were curious what are then the requirements for strength in swimming so we went to the Norwegian swimming federation and our Olympic uh uh government and the only parameter that we have classified in Norway as important is the squat jump meaning squat jump is important for the start and turn performance so they've gave some recommendations in terms of high how high you should jump in squat jump in terms of reaching the Olympic Games final for men and women in the different strokes and distances but we think that there's more to it than just squat jumping so we went out to look what else is out there and then we found that the British swimming have classified three other exercises as uh important if you want to be an elite senior swimmer so they identify that chin-ups and then they provided uh how heavy you should be able to lift up body weight plus additional kilos they used bench press and back squat and then they classify that for female sprint middle and distance so then how can we utilize then this technology so I think the the topic of force velocity load velocity has been covered quite well in the other webinars so we will not go into too much detail about or the specifics about it but we will talk about it from a swimming perspective so basically what you can do then by generating a load velocity profile is that for example you can see if a swimmer is load or force dominant if they are velocity dominant and what are their production capabilities so we were kind of interested to see can we see some parameters from a load velocity profile in dryland strength exercises that could be important for the swimming performance so we have only done one experiment so far where we used the 1080 quantum and there we had the butterfly swimmers perform a load velocity profile in leg squat and in bench press unfortunately we could not find any variables from the load velocity profile correlating with the 50 m butterfly performance but we found that squat jump as identified previously had a high correlation with the performance in the start and turn segment it doesn't mean that we cannot find important findings from land exercises but from these two perhaps they were too unswimming specific or it could also be other reasons for it so therefore we kind of moved more over to the pool and here we utilized the80 sprint machine and basically what you see here is our experimental setup for semi-etered swimming and when we talk about semi-ettered swimming you will see that the swimmer has a belt around their waist and then it's connected to the wire going out of the sprint machine which again is placed on the top of the starting block so it's one meter above the water surface and then we use this to predict and assess swimming performance so over on the right side here you will then see an example of the load velocity profile that we generate in swimming so here you see the different marks 1 kilo 5 kilo and 9 kilos load so that means that when we are testing this swimmer they are swimming 25 m with maximum effort and on the first trial we give them 1 kilos resistance the second trial the 5 kilos and then 9 kilos resistance and then we collect the data from the 5 m to the 20.5 m mark and then we plot it into the and get the regression line and the load velocity profile so basically what you will see here is that the L that's the predicted maximum load that the swimmer could pull and VO is the predicted maximum velocity so before using a performance test it's important to know if it's reliable or not and what you see here now is some preliminary data from a study that we conducted in terms of reliability and on the left side you will have again the L predicted maximum load we have the RL which is then the LO relative to body mass of the swimmer we have VO predicted maximum velocity slope and then you have the adjusted R r 2 which is basically how well the measurements are fitting the regression line so if you look at then the coefficient of variance almost every parameter is under five 5% and I'm sorry I forgot to tell you what again if you look at the first the table you see the load velocity profiling when we did five different loads here we tested the swimmer at 1 3 5 7 and 9 kilos and on the bottom you see the same calculation only using three different loads it's based on the same data but here we excluded the three and the 7 kilos measurements and as you can see it's very similar almost identical between using five and three different load for calculating load velocity in front crawl swimming and again looking at the reliability results you see that the interclass correlation coefficient also have excellent values almost every measurement is over a 0.9 and moving over to the right you see SEM which is then the standard error of the measurement and on the right of that again we have the minimal detectable change you should be quite uh careful in using this minimal detectable change from this table in terms of using it for predicting performance changes because in this study there was boys girls men women different performance levels and of course if you have 2.8 8 kilos as a minimal detectable change to see an improvement in performance for LO and you as a swimmer can only pull 8 kilos which was the case here it's quite a large improvement you need to make on the other hand we had male swimmers pulling 23 kilos for them it would be much less but the reason why I'm showing it is because you see that the MDC is quite larger than the SEM which is a good thing in terms of reliability because then the standard error of the measurement is classified as low versus the possibility of seeing actually a change in performance we also use this to predict performance meaning here you can see 50 m butterfly swimmers and on the xaxis you have VO which again is the predicted maximal velocity from the load velocity profile and on the yaxis you have the vmax which is actually their swimming velocity from a 50 m butterfly race so in this setup we first tested the swimmers in 50 m butterfly all out in the race analysis system and thereafter they provided a a load velocity profile and what you can see here is that then the relationship between the predicted maximum speed from load velocity and the maximum swimming speed during 50 m swimming has a very large correlation and also a very high intraclass correlation so basically what that means is that the velocity predicted from load velocity corresponds very well with the actual swimming velocity during a race so again if you have a minimal detectable change in your velocity outcome from load velocity that should also then transfer into improving your swimming velocity here you actually see a mix of three different studies that we did and here we try to look at the parameters coming from the load velocity profile and link it towards u performance measurements in the swimming race so on top here you see then theif the V50 FC and the V50 BFF which is then the velocity from swimming 50 m front crawl and the velocity from swimming 50 m butterfly and again we see strong relationships between the velocity that you make during your race to how strong you are meaning how high your LO is and if you look at the 50 m front crawl study you will see that this is even higher when you relate it to percentage of body weight so when LO is related to percentage of body weight it has even a higher correlation significant importance for the performance on the bottom here we have a study where we used girls age 11 13 and 16 and we didn't measure them in a 50 m race but we measure them in a load velocity profile and basically here we have the predicted maximum velocity from the load velocity profile in FC front crawl BA is backstroke BR breaststroke and BF butterfly so in all four strokes and here we looked at it in relation to percentage of um body weight and interestingly here is that if you look at the front crawl and backstroke there is a significant relationship between their predicted velocity max and their predicted LO in relation to body weight but this is absent in both breaststroke and butterfly so what this points towards is that the alternating strokes of backstroke and front crawl it is important to be strong and generate high propulsive forces to get a high velocity while in breaststroke and butterfly in this age groups this might not be the deciding factor here it could be much more related to how do you minimize the drag and how is your stroke technique and here we have an example again from the load velocity profile of the 50 m butterfly swimmers and here you see all the swimmers profile in in the study and if you go back to the slide before when we see that there's a stronger correlation when you do when you take the predicted maximum load and relate it to the percentage of body weight it would probably have been more interesting to look at this also for this study but unfortunately here you only see the actual true values so what's interesting here is if you look at the blue line that swimmer is extremely strong and it's also the second fastest swimmer in the study the swimmer in red is also strong but not close to as strong as the blue one but it's the fastest swimmer in the study both of these swimmers are international top level or international level swimmers and they're also extremely different in terms of their physique so if you wanted to give some recommendations to these swimmers what would you do well probably first and foremost we would relate it to the body weight because we know that the blue swimmer is much taller it's much more heavier and muscular than for example the blue swimmer but at least from analyzing this we could say that the swimmer in blue is extremely strong perhaps even too strong because we know that added body mass from muscles also increases or can have a good impact on the drag because your body will be heavier in the water and also is not really able to utilize all of this force to kind of generate high velocity so at least to put the swimmer on more explosive strength or swim training could be beneficial in in order to improve their maximum predicted velocity for the red swimmer was quite a steep slope again swimming the fastest which is the main output anyway how could this swimmer then improve their performance most likely why probably increasing a little bit of their maximum strength possibilities and again probably improving on their peak power output and this takes us over then to the peak power training which is very popular in many sports and I just read from rowing some time back that rowing claims that the peak power is the number one deciding factor to rowing performance together with the rowing technique much more important than for example V2 max or lactate threshold and then what is the case in swimming is it important to train and develop your peak power is it important to practice uh resistive training specifically in the water at least if you look towards the right figure here we see one male breast stroker that we tested and in the the red line it's the load velocity profile and the blue it's the power curve and basically if you look then towards the top of top of the screen here we see that this swimmer's LO is almost 23 kilos so quite a strong swimmer the Vax is 1.53 m/s so going to the theory behind
load velocity and peak power we see the L optimum is classified as 11.45 45 kilos and that is the load corresponding to where the vmax velocity has decreased 50% and in the v optimal that's the velocity where this occurs further on we see this l deceleration 10% that is the load corresponding to a 10% decrease in the velocity max and the V deceleration 10% is that actual velocity so far I only know about one study who tried to do an intervention using resistive sprint training with the80 sprint coming out of Sweden testing their elite swimmers elite sprint freestylers they were using the L D deceleration 10% as the resistive training which for this breast stroke would have been 2.29 kilos but after 6 weeks of intervention together with a control group they couldn't find any significant differences in the velocity they of course they saw individual differences some swimmers actually had quite large improvements in the experimental group but not as a group and one challenge that they mentioned from that study was that there were swimmers from different clubs and they were not able to control what they did outside of the pool for example in terms of land strength training how much they actually swam and what they did besides the experiment and also one theory in swimming is that if you were to train at L optimum which would decrease your velocity by 50% it would perhaps alter your stroke technique too much in terms of being beneficial but these are still questions that we don't know too much about yet but at least the purpose of by generating this power curve at least that you could then choose and pick which spot you would like to do and then programming the,080 sprint to let you train at that certain resistance with that velocity so again how to utilize this technology here again we see the power curve from stroke to stroke cycles in breaststroke girls and from the top swimmer you see that it's a very flat mean power curve indicating that this swimmer is very good in keeping her power cycle by cycle throughout the test both at 1 kilo's load which is the blue and at the heaviest load she tested the kind of orange 5 kilos swimmer on the bottom on the other hand you can see that her mean power is decreasing throughout the swim this means that even at 1 kilo the swimmer is not able to keep up her power so this swimmer might have a challenge in actually generating good high power in her swimming and even more so to maintain it either due to fatigue or also possible to stroke technique and the drag but at least this graph here will show us if the swimmer is good in generating high power and how well they can keep it throughout the distance just a quick example in terms of breathing and the symmetries that we did from a little student experiment so if you're looking towards the right here you see the distance and the force and the green is when the swimmer did no breeding uh blue is breeding right and red is breeding left and I mean as a coach you keep telling your swimmers when you sprint freestyle you should not breed and this is quite a visual representation of why you shouldn't because here you can see that the force production is much higher when you don't breed but of course also looking at this you could then identify does the swimmer have any asymmetries between breeding to one side or the other do they have a preferred side a weaker side should they then spend time on improving the weaker side or is it even better just to actually go for the strong side and finally what we are now starting to experiment with is then the passive drag i mean we know that from the start and the turn you have a not a significant large amount of time but there is some time when you're spending in streamline and how should your body position be in order to minimize the drag and inc and maintaining the velocity so just quickly here some future directions we want to then try to establish relationships between load velocity parameters and swimming performance in all the different strokes we have quite an extensive database right now so I think we will soon start to go deeper into the other strokes different genders different age groups the second question here would be very interesting is there an optimal low velocity relationship for different strokes and distances would it be possible to identify what is the optimal profile could we also look into lower and upper body contribution in terms of the stroke technique and of course a big dream would be to look into active drag then we are moving over to heart rate in swimming and here you see some different devices in terms of measuring heart rate and if I'm not mistaken I think the first device at least portable came from Polar in 1977 and since then the technology has uh improved quite significantly and for most people involved in endurance sports at least in Norway when you think about running cross-country skiing many many use heart rate as a tool for measuring intensity while in swimming this has been very little utilized up until now at least so one question is of course how accurate is heart rate or optical heart rate we know from uh previous validity studies that uh the P polar H10 chest belt is very valid compared to ECG but we know very little in terms of optical heart rate we know that in the past it has been not very reliable and valid for measuring heart rate in swimming but also in other activities and sports here you just see an example of a figure from one of the testings we did when you see the H10 the O1 and the M600 throughout the training session and as you can see here the H10 and the O1 optical heart rate sensor are following each other very very well if you look at the picture to the left that's me in my triathlon suit in an early pilot stage of this experiment here we place the chest belt u around the chest and the O1 sensor is here placed on the temple of the head underneath the swimming cap and it's also very important not to uh to to make sure that there is no hair coming underneath it so the big question is how accurate is optical heart rate so this is some data from a recent uh paper that we published and if you look at the H10 which is then the criterion measurement and in this study we used both male and female swimmers and the chest belt was then placed underneath the swimsuit and that's mainly because the chest belt for male swimmers is not very practical because every time you push off from a turn it will slide down so what we see here is that the H10 and the O1 has no difference while on the other hand M600 in terms of heart rate max and heart rate mean significantly lower than the H10 and also again the ICC values very strong for the H10 with O1 but also maybe what's more important is is it also able to identify the different uh training zones or the different uh intensity zones so again we look at uh the heart rate zones to the left under 50% 50 to 60 and so on up to 100% we use H10 as the criterion measurement and then we have O1 in blue and M600 in red and again there is no significant difference between the H10 and the O1 while again the M600 for certain uh heart rate zones tend to be lower and also higher than the H10 so basically what we can take from this is that O1 is an accurate measurement of a heart rate in the water during swimming while M600 is placed on the wrist is not as accurate so in that sense the question is how should you then utilize heart rate in swimming i mean swimming is extremely velocity do or dominated we can basically use velocities or time for distance to predict what intensity the swimmer should hold and in that sense we could use V2 max which is very complicated to do in the water so we don't do that very often that's more for land-based activities but we often do lactate in terms of step test or uh critical swim speed tests heart rate could be used in terms of percentage of max heart rate and of course RP can be used as a subjective measurement but again the point is that in swimming we are usually usually mostly using velocity or time so let's say today the swimmers are doing 10 * 100 freestyle corresponding to 110 per 100 m and that should be intensity zone three but we also know that uh training adaptations make differences to how fast we should swim it also depends on our daily shape like uh I became dad uh four months ago for the second time and I must admit my sleep is not extremely good all the time so of course if you're sleeping bad for one night consecutive nights you have a lot of stress your body will not get the same impact swimming 110 for 100 meter freestyle it might get a larger impact because your body is fatigued and of course if you keep doing that day after day for a long time you might even end up in overreaching and in worst case an overtraining state of mind so the purpose here of introducing heart rate is that it would be an objective measurement of your intensity or the effort level that you put in that as you can see in this movie can be recorded live onto an iPad so the coach could be on the pool deck telling or showing the swimmers okay hey you are training in the right zone today great you are training too slow too fast it would be a way of monitoring and correcting the intensity in an objective manner and I think I see time is running very quickly so I will just quickly go to two more slides and then I think we have to maybe cut and that's basically if you are to utilize heart rate in swimming we need to kind of base it on something and then it's the maximal heart rate so we did one study that looked at different protocols to achieve maximal heart rate in sprinters and middle distance swimmers to see if there's a difference in terms of is there a favorable set for a sprinter or a middle distance swimmer so here we used then three protocols one was step test 50 m 100 meter and a 200 m protocol as you can see to the right on the screen here so we wanted to first determine is there a different protocol suited for different type of swimmers and also again to look at the topic of is the maximum heart rate the same during swimming as it is during running or cycling so just quickly go to the results here there were no difference in the protocols between the 50 100 and 200 meter the mean max heart rate was the same there were also no differences in the max heart rate between sprinters and middle distance for swimming protocols so basically you could either use the 50 100 and 200 m for sprinters and middle distance swimmers it would more probably be a prefer preference of the swimmer what they would do but what's important to notice here and it's also what we saw back in the 70s when this topic was previously investigated that the maximal heart rate is lower in the water compared to on land during running and cycling so as you can see here it was about 6.7 up to 5.3 standard deviation difference between land and water and this basically means that it's
important in that sense to do sports specific tests if you want to utilize heart rate based on max heart rate than for example in swimming and also like if you're a triathlete you're training running cycling and swimming if you don't use the same heart rate as you have for running when you go to the pool you're quite likely to be training in the wrong intensities so I don't know Peter should we maybe wrap it up and then do this for next round or should I quickly do some slides uh people are still hanging on here so why don't we just why don't you just do uh go through uh the EMG stuff there um yeah I'll make it in five minutes do that in a few minutes yeah and I and then uh because a lot of people are interested here so yeah please i'm sorry for running rushing through the end here but uh time is of course ticking so basically here on the right side in this movie you see the testing during my PhD project I project and you can see that the EMG system that we then used also in the middle picture when they have to use the logger and the phone up in the swimming cap is quite different than the new technology we acquired which is on the left side again very small sensors no need to have an receiver and a data logger present with the swimmer stored in the internal memory and also what you see in this movie is uh that we also used in Qualysis for 3D motion capture during this uh study that was actually I think one of the first times Qualysis was used in the water and it would be nice to bring it back for future data collection basically I'm not going to give you all these results either but I thought I'd show it to you but this is basically the main outcome from my PhD and it's basically to show you what you can get from using EMG or electromyiography in terms of measuring muscle activation during swimming so in this PhD we were fortunate to have four worldclass swimmers present two was world champions one was Olympic medalist and one was European medalist so they were in our worldass group and then we had a national elite group with the swimmers qualified for national championship and then we also had age group swimmers but basically what you see on the screen here now is the recommendations that you can get from using EMGs to look at muscular activation and kinematics to evaluate breaststroke technique so basically what we see here is that the best swimmers in the world they have for example an earlier activation in the biceps femories during the leg recovery of breaststroke and that's basically to decrease the time they spend in this phase they activate it earlier so they can have a quicker leg recovery and we also see in tibialis anterior that they have a later and quicker activation meaning that again to reduce the drag they are not making a premature dorsif flexion before they come up but they bring their heels up and then there is a quick uh activation in tibialis here it's there's some other ways to utilize the EMG in that sense you can see if the muscle is active or not is the muscle resting during the gliding phases of the stroke or is the muscle still active using energy getting fatigued so in this figure you basically see the four muscles on the upper body that we collected data from it's the tibialis brachi bubrachi trapezius and the pectoralis mayor and the red line here is at 100% effort the green at 80% and the blue at 60% effort and we use this mainly to see if the muscle activation pattern changed with the different intensities but here we can for example also look at it in terms of if the muscle is active or not by looking at the stapled u black line across the figures if the muscle is underneath this threshold level it's relaxing or at least uh it's not considered to be active but as soon as it goes up it shows that it's active and then the question is for coaches swimmers scientists to interpret that what is the beneficial way of activating deactivating the muscle and of course you can also look here at the peak amplitude in terms of seeing how active is the muscle and of course you can also look at then the co-activation for example then between the triceps bachi and the biceps brachi how much is the muscles co-activating during the stroke cycle this is also an example of what we collected data from but we have not done so much about it yet but this is in terms of looking at the muscular correspondence between maximal effort swimming and different technique and drill exercises and for everyone familiar with swimming you we know that swimming is probably the sport in the world who do who does the most drill exercises a lot of the things we do in the pool is drills and technique exercises but basically just a quick example of what we then can see in this square here now you see that this uh solid line is uh 80% effort doing two kicks to one pull exercise the stapled line is uh 60% effort and the dotted line is 100% effort and we know that in sprint breaststroke we want to activate the pectoralis brachi and the biceps brachi early in or early during the leg glide to not have an excessive glide and to generate propulsion from the arm pull so basically what we see here is that if you perform this drill exercise two kicks to one pull at 60% you have a delay in the muscular activation but if you perform it at 80% you will actually have the same or even an improved uh earlininess you would be earlier even in generating activation and propulsion from the muscles and also at the same time if you look over on the the right side where we have the muscles on the lower body here again we have in green it's the biceps femist like we talked about earlier and you can see here at 80% the activation is even earlier earlier than it is at 100% but also here at 60% it's earlier so again here both exercises could be utilized in order to improve the timing or at least to make the activation in the bubis earlier during the leg recovery in order to improve the to decrease the time you spend there and I think we skipped this so basically to sum up the last slide here some future directions in terms of EMG is that y
2025-05-01 14:37
