The Use Case for Heart Rate Variability with Marco Altini and Corrine Malcolm | Koopcast Episode 106

trail and ultra runners what is going on what's happening welcome to another episode of the KoopCast as always i am your humble host coach Jason Koop and on this episode of the podcast we have a little bit of a tag on to episode 102 where we talked about the aura ring three and is it worth it with our coaches stephanie howe karen malcolm and ryan anderson and after that podcast came out somebody reached out to me via twitter who is actually the author of that paper marker altini who just also happens to have his own heart rate variability of measurement company where he has an heart rate training for athletes or hrv for athletes and i wanted to bring him on the podcast not only to discuss the validation study that he was the primary author on but what the utility of measuring heart rate variability is because this is one of those areas where i think that the amount of utility that is being presented for heart rate variability in the marketplace by all these wearables far outstrips far outstrips the amount of actual utility that you can get from measuring heart rate variability so i wanted to bring marco on the podcast to discuss that as well as a lot of these combination readiness and recovery scores that we're starting to see out in the marketplace although marco is an

advisor for aura i think you guys will find him honest and very pragmatic and he kind of pulls no punches with what he thinks about some of the things that actually go on on the inside of the algorithms and the data that is actually being portrayed with a lot of these wearables i had a lot of fun with this conversation as always i bring in people that are way smarter than me into the room and so i decided to bring in corinne malcolm again to this conversation so here it is my conversation with marco altini and corinne malcolm all about heart rate variability i kind of want to start there and and and talk about the app and talk about why you started it and kind of more if you could if you could summarize those last several years of you designing all the features in the app because i think that paints a really good picture for kind of the action ability side of things which is what we're ultimately going to get into yeah yeah for sure so you know we started this uh yeah eight ten years ago i think at the beginning uh you know much of it is just timing it was when you could finally just link phones to sensors right we couldn't even do that before even with chest traps like before bluetooth 4.0 which is you know the first one that allowed you to talk to sensors and and porous traps and things like that we couldn't even get heart rate data easily to an app so that was the

beginning of you know trying to get physiological data into an app so that we can browse it and show it to users in a way that it can help them tracking things like physiological stress then um we evolved that towards trying to make it easier for people to use people that people don't like to use chest traps first thing in the morning so the camera technology uh started to be something that was um a bit more let's say explored those years so you had this company out of mit that had an app that could measure heart rate with the camera so i thought well maybe we can do a bit more than that and look at the b2b differences and get into heart rate variability and that was indeed doable and we developed you know the first algorithms to measure heart rate variability using the phone camera so we had something that we could compare to chest traps and ecgs so just reference systems to do this and it was as accurate uh you know when you don't move and obviously there are some um things to consider right with this technology that is a bit more prone to noise as we know also from optical sensors that we use these days and watches and things like that so maintaining it accurate uh making it super affordable right just a few dollars to buy something that allows you to measure physiology every day easy to use so that first thing in the morning

you wake up you take your measurement you don't need anything else any other sensor or device and then that was the beginning the first few years was a lot of this trying to get this very accurate very simple um and then the interpretation of the data i think that came a bit later because we also needed to learn that i think thanks to the technology right uh before hrv studies were like okay let's get a bunch of people in the lab and we measure hrv once then we did this intervention and then we measure hrv again after four months and we learned that that does not make any sense because there is so much day-to-day variability right today and tomorrow is so different so what's the point of of doing that uh in any study as a matter of fact so there were the first interventions looking at day-to-day data and also there you know people were just trying things for example today your hrv is a bit lower than yesterday so maybe there is more stress but then over the years we learned that it's not really like that we need to identify for example what is a normal range for you that means you know just what's the day-to-day variability in which uh nothing relevant has happened that's just a normal change today is a bit lower than yesterday but it's not a change that is significant right it's uh you know similar to many people can relate sometimes to blood pressure uh because there we are used to have a range in which you know we know that very high is bad and very low is bad and we want to be in this normal range

now we see hrv similarly but the range is very specific to you it's not just a population range in which you know you need to fit so that's how things have evolved i think a lot during the past few years and also how we try to stress the interpretation should be done that's why you need to measure many days ideally for a month or two so you know what's your normal that's your range and then when you deviate from that then we know that something um a stronger stressor let's say is present so that's when you might want to make some adjustments or move to the actual ability piece um so yeah a bit these elements i would say over the years i'm really glad that you mentioned kind of the origin story of how heart rate variability initially was very clumsy to get captured right you had to put on an external advice and getting that information from an external advice device into something that would then analyze it and then use that data between a coach and an athlete i mean i've been coaching long enough where i've seen that pathway in many in many different training devices from heart rate monitors to power meters to gps technology i mean they've all kind of run that similar run of show where at first the data the data capture to action ability piece was really clumsy and as that pathway starts to get paved a whole lot better that's when athletes start or or don't uh the mass adoption piece of it and i think

heart rate variability is one of those where you where we're definitely seeing that pathway get smoothed out a lot quicker and now we're kind of in the we're in the the the mass adoption acceleration phase where so many people have access to this the technology is simple it's relatively affordable it's easy to get into the hands of everybody and it's easy to produce relatively i say this i know you're like in the background going it's really not that easy it's really easy to produce the data to to produce the visualization side of things right as well make things slick for people to look at and stuff like that my first exposure to what you do as as i as i mentioned earlier is just the simplicity of being able to put my finger on this camera and measure heart rate and heart rate variability and i i remember that as a transition point to where i could say okay now it's now it's at the point where the invasiveness of of capturing this physiological measurement is not is is is not so intrusive and so complicated that people can actually start using it in an actual way because as you know compliance is the key with any of this stuff and if it's not easy to to to start to if it's not easy for athletes to actually do these things they're not going to actually do them corinne what was your first exposure to using heart rate variability we've talked about this before a little bit yeah i um i wanted to utilize it doing some

research during my graduate studies and so we actually my advisor would not buy into the idea that this was going to be accurate enough like you refused to buy into that right you're like this being the phone you're holding up we're we've got a failure of video right here so you're holding up a phone this being piece of accuracy yeah so like he was like no no no like if we're going to add hrv to this study you have to like we have to be collecting better data than that and so trying to get a bunch of runners doing the knee knacker race in vancouver to wear to wake up every morning and put on a chest strap to link it to i think we used um hrv elite maybe because we could get the raw data exported to cubios um and so to convince them you know every day for several weeks to wake up and do this for five minutes and to lay in bed and to like not fall back asleep but to be very calm um was so much more tedious than being like okay i need a minute reading with your finger over the light of your phone camera and um you know we'll like we can export that easily so it's it was very clunky in that sense despite the fact that that was when papers were coming out about the validity with the phone lights um for like specifically for hrv and when you're talking about needing 40 to 60 days to like actually weight trends with heart rate variability that's a lot of days in a row to try to convince athletes to wake up and put a chest strap on i've also seen it i come from a nordic skiing background and so first beat

which is a sunto i think partnered or affiliate like hrv group they used it a lot for tracking um kind of non-functional overreach um over training in a number of athletes who were like were experiencing mono who were sick with mono and they used that in their recovery and there was some stuff published in the nordic skiing community specific to that so i had read all that stuff and in my mind i'm like well they're going to bed with a chest strap on so maybe that is the right way to do it so we've come a long ways when it comes to wearables and the clunkiness of these devices i had athletes when you just reminded me corinne i had athletes when i first started coaching that would wear their heart rate monitors to bed the chest straps and not just the like the nice ones that we have that are like fabric now like the old clunky plastic like super thick ones i'm like this is dumb like you're compromising your sleep in a way that's not worth the data capture after that anyway let's let's not get sidetracked too much okay so we're going to fundamentally have a conversation about the utility of heart rate variability in training um but i think to set the table uh for for the listeners and a little bit for for ourselves right here we we fundamentally need to go over what heart rate variability is from a physiological perspective and so marco we're going to kind of lean on you as the expert who has the app you've studied it way more than corinne and i uh have and you also use it from for yourself and for and and for athletes can you

give the listeners like a freshman undergraduate level synopsis of what heart rate variability is so that we can then use that to to try to dig into the utility a little bit yeah yeah for sure so when we measure heart rate variability we are looking at a measure of physiological stress which is derived from how the body responds to the stresses we face typically via the autonomic nervous system so the part of your body that you do not control yourself you know everything that is happening in the background uh you know like you don't have to remember that you need your heart to beat or that you need to breathe and those sort of things are always happening and they stay in this state of balance that is required for optimal functioning when we face stressors there are disruptions in these mechanisms and one of these disruptions is a change in heart rate variability and that's what we measure with this technology so if we face a stressor there will be a reduction in heart rate variability your heart rate becomes a bit more constant so the difference between consecutive beats is a bit more similar and that's a clear marker of physiological stress and that is why we use heart rate variability it is not possible to measure the autonomic nervous system especially the parasympathetic branch directly which is the one in charge of you know

recovery or relaxation what we care about here so we use heart rate variability as a proxy of that so something that is impacted by that process and therefore can give us some insights on physiological stress level does that make sense it totally makes sense and i i i'm trying to play the role of the listener here a little bit in asking some probing questions um we use proxies a lot in training and corinne and i we kind of beat our heads you know we kind of beat our heads against the wall every once in a while when we're using these proxies because that's really all they are and sometimes those proxies are very good and sometimes those proxies are okay and sometimes those proxies are actually pretty pretty terrible you're using a proxy for quote unquote stress that has a wide range of inputs uh from neuromuscular stress muscular stress cardiovascular stress life stress i stayed up with a barking dog all night corinne's dog peaty is bothering her during work and that's kind of stressing her out and we're using not only but we're using in this conversation heart rate variability as a proxy to distill down all of those pieces of stress to say okay how how stressed is the actual athlete and what i think a lot of the listeners are thinking of right now is how good is that proxy given the fact that you're taking all of these branches of input and kind of distilling them down

into into into one channel i know you've heard this piece of feedback before so what what do you say to that to to that process we're taking all these pieces of stress and we're distilling it down into this one thing yeah so i think the first thing that is important to understand which sometimes is um forgotten when we think about athletes and training is that it is definitely not just about training right so training is one of the stressors um if there are other stressors linked to a lifestyle and you know the most obvious are you know alcohol intake poor diet things like that those will have an effect to a point that you might not get anything meaningful in the context of training from the data because it's changed so much because of your lifestyle choices for example so this is a global marker and even in that sense it still does not capture everything that you mentioned for example so if we think about um you know from a data perspective we have clear associations between things like training of various intensities of various intensities and hrv or the menstrual cycle different phases and hrv or sickness and hrv alcohol intake um traveling those kind of things you know show clear associations but there are things like uh muscular damage right which is key to athletes and that is not really captured well

right so if you go a high intensity session with very short intervals for example something that maybe has a lower load from other points of view but has a high load at the normal solar level is something that you might not be able to capture with this matrix so that is why the way we tend to look at the data or inform people about how to use the data is always to use it as an overall marker of stress related to anything that will affect the autonomic nervous system such as the stress as we mentioned but to look at it not as the one marker but you know in combination with other things the most obvious being how does the athlete feels right so the subjective uh you know perception of performance or training or all of that the other piece is obviously training external load whatever you did that is you know the important part of the training plan um and then look at these things together so that you can get a bit closer to the bigger picture uh but not obviously just using this one market of stress as the one market that will capture everything because there will always be something that is not really reflected in there muscle of soreness always becomes the most obvious one in my experience you know corinne i'm going to want you to jump in here for a second but let me kind of set this up okay i think i think that this area that you just mentioned marco is the one that confuses people the most

because they're looking at a singular metric and we're going to get into the composite scores i promise you because i know marco you want to talk about that but we're but in a lot of times when we're looking at heart rate variability we're looking at it as this true surrogate uh of of quote unquote stress and as you mentioned it correlates very well with stress in some areas and not so well with stress in other areas and we all know stress is stress is stress right i mean if i go and i do a really especially for trail running right a really muscularly damaging workout a lot of descending and things like that it might not show up as very stressful on some ways that we measure it heart rate variability might be might be one of those but on other ways how do i feel i feel like shit because my legs are super sore that's good like that imbalance i think is is part of the reason that athletes get so confused when they're looking at it because they're because they're seeing this thing that's supposed to alchemize all of the properties as being or they think that it's more like weighted equally or true representation or whatever but in reality what they're sensing is is different and that discrepancy is hard for athletes to go through karen i want you to kind of jump in here because i know you've got some thoughts on this as well yeah so first i think it's i've got a couple things okay so first i think it's really important that the audience understands that the heart rate variability metric that is now being used in

apps and through wearables the the the like the specific metric like that number there was a ton of research that went into which which number of from heart rate variability should we be using to court to correlate to correspond with stress they looked at high frequencies and low frequencies and all these different mathematical interpretations of hrv to get to what is now used in apps as this thing that we can say okay this is going to be our marker our thing that corresponds best with stress including like when do you take that reading so tons and tons and tons of research has been done to get down to this like quote unquote simple metric and i think that's really important because there's a lot of nuance that we're skating over there um from heart from the heart rate variability standpoint there's like grant money on the line for this stuff because people think it's the fountain of youth all sorts of fun stuff so a lot of research has been done to get to this like simple metric and that's important the next thing that i think like the biggest complaint and you guys just both talked about this was that delayed onset muscle soreness that dom's effect doesn't correspond necessarily with feeling like you're ready and so when you get this readiness score in an app you're like well i feel like garbage how can i possibly be ready to go again that's like a that's a contextual thing where that stress isn't picked up well by this one metric and maybe this is the thing that i like champion the most in any of these conversations is that context is

so important and i think that's the thing that we struggle with with our athletes when they want to use when they want to bring in a new tool to the toolbox be it hrv or whatnot um is that what is the context and does this metric does this app whatever it might be take in that context well is it like is the machine learning is the ai there to understand the like the nuanced context of someone's life how much information is going into that to give you that readiness value and i think that's honestly there's like through lines here to how coaching works in general right like i provide much better coaching when i know more about my athlete and i think things like readiness scores can eventually give back more to the athlete when it knows more about the athlete so i think that there's like this idea that oh we're going to take this metric it's going to be super easy to use and in reality there's so much more nuance behind what any one metric means to a coach to the athlete and what should be used moving forward so i mean in that vein corinne sometimes i think it's harder sometimes i think it makes it better but it's also harder and this is the conundrum right we want it better and easier meaning you get better information to drive action and you also want that information easier to understand and apply and with when we start getting into things like heart rate variability or any one

of the myriad of things that any of the groups or aura rings actually measure you might get better information but it is certainly more hard more complicated and and more i was going to use the word precise but it's just more complicated to to distill down into action because you have to know kind of what everything's going on i i think that the a really interesting lens to look at this through marco and this is where you can jump back in is how you have intentionally decided to do the data visualization side of things because corinne just mentioned we glossed over a lot of the nuance and i think we kind of have to to keep this like a freshman level course versus like a PhD level course because nobody will be able to keep up at that point maybe that's like four years from now we can do that but i i think that one of the things that i've appreciated about your app is just that is what you are displaying to the athlete and also i think more importantly what you're not displaying right um to to add to the confusion so why don't you set that up and then with that background of why you intentionally chose to do things the way that you do yeah yeah for sure so i think that here there are a couple of points that are really important and really easy to get confused about and you know the technology improved a lot right we said we started from all these difficult methods and then we went to the phone cameras and now we are

aware of what we just sleep and we don't do anything right and they collect the data so that is great especially for athletes that otherwise would not measure i would not have high compliance and you know if you get two data points every 10 days then you better not use the data at all right so if we have these tools then there are two additional aspects that i think are really important and we can try to discuss them like high level but still um mention them because there's a lot of confusion that comes from using tools that can potentially measure accurately and they still provide you with the raw data let's say hrv not a composite score or a readiness score or whatever but still the data is not useful because it's not sampled at the right time and you know the easiest example is you know an apple watch right there are i don't know 100 million apple watch devices out there so it's very commonly used the data gets into your health app you don't do anything and you have your hrv in there at the same time this data is not used typically to measure first thing in the morning and in the night it is sampled randomly so you get these data points that are maybe once at 1am and then once at 3am and that's very problematic because during the night there is you know the circadian rhythm your physiology changes a bit so you know your heart rate reduces throughout the night hrv increases a bit so if you sample at different times

the second night maybe you were not more stressed it was just another hour and then another thing is that sleep stages have a huge impact right autonomic activity during a sleep stage is very different during deep sleep and REM sleep for example that is actually the whole point in when we try to estimate sleep stages from hrv right if there was no strong association there we could never try to do that but that means that if you get again a few data points and one night they are in REM sleep and the other night they are in deep sleep then again this data becomes meaningless so my point there for you know the consumer whoever is using a device is just to make sure that you know you use a device that is either allowing it to measure thrusting in the morning or that gives you the full night of data because if you have the full night of data you average out all of these issues the circadian rhythms and sleep stages that's not a problem anymore because you have all the data and you know some say that maybe you need to collect data in a specific sleep stage right so maybe deep sleep is better because it's a more stable state but unfortunately the technology does not allow to do that we cannot be certain that any given point of the night you are in a certain sleep stage right even with the best algorithms we can build we can maybe say you were 50 minutes this night in deep sleep or something around the time but we can never say in this moment you were in deep sleep right that level of precision is not there so we cannot say we measure

your hrv at the time because you were in that stage because we are never sure about that and there is also wait a minute hold on marco i want you we might come back to this but i want you to repeat that last piece just for the listeners because you literally just wrote the paper on this right you just wrote a paper that i'll link in the show notes having to do with aura's new mission the new aura uh three ring which we had a podcast on that which you know this is how we got synced up on twitter for the people in the background this is how these things happen um but you literally wrote this you literally wrote this paper and for you to come out and say we cannot pinpoint this stage of sleep even though we have all of these different inputs coming into it i think i think is quite powerful so before we kind of go on any further i want to like pull that apart really quick because it's going to come back with the readiness score piece pieces of things what did you do during that study and why is that so why is this sleep stage scoring so problematic yeah so let's start from the beginning of why sleep stage is very problematic and that's just that we don't even have a reference like when we say reference for sleep staging is uh psg a device that measures your brain activity uh your eye movement and your muscle activity that is uh data that is collected and then a person scores it which means that they look at it

and they say okay there's 30 seconds you are in deep sleep and there's 30 seconds you are in lamb sleep and they do that throughout the night and then another person does that and then they agree 80 85 percent of the time and that's what we call the truth like the reference for us to develop an algorithm so i think there is the first problem it's not that we have you know an actual reference if you develop an activity recognition algorithm we know that you know this time point you were walking and this time point you were running and that's the reference but if when we do that for sleep we don't really know that you were in deep sleep and your time sleep you know that's a person's opinion uh that you know typically is 80 85 percent accurate and that's the best you can do so if we were to develop an algorithm that was a hundred percent accurate that would actually be a hundred percent of that 85 percent right because still the reference has a margin of error so that's you know the first problem in all of this is that it's very difficult because even the reference is not perfect then from there we use autonomic activities so again heart rate heart rate variability temperature movement any sort of thing that you can measure um at the finger with the ring or at the wrist with another device and then we use that as a proxy of you know brain activity because that's where you know sleep stages are actually measured normally and there is a relation between those because otherwise it would be impossible to get um you know the results that we obtained recently but at the same time

there is again a margin of error so in my opinion the best we can do even with the latest algorithms and the latest results is to spot trends over time maybe to see if you know there are larger changes in situations where your behavior changes dramatically again maybe you are traveling frequently or things like that or you know you have a kid you know things that change behavior and then maybe we can capture those changes and see what could be the implications or all we can try to address those but still uh given all of these errors that you know accumulate eventually we i think it is simply not possible to pinpoint a time point in the night and say okay at this precise minute you were in this stage it's just we just don't have the capacity to do that and apart from very obvious maybe sleep stages like maybe a ram uh it's very difficult to do that even if you have psg data because it's just challenging to you know to do this we're good so we're gonna come back to the to the how how your app visualizes the heart rate uh variability data but just just for the listeners the the title of the paper is the promise of sleep a multi-sensor approach for accurate sleep stage detection using the aura ring and this is just their latest um that it's just or as latest validation paper right which you always need to do whenever you iterate

either the algorithms that go into um the different devices or actually the hardware that's that's that's actually put into the device i'll leave the links to that on the show notes but i want the listeners to kind of remember this fundamental issue that we have when we're scoring sleep that the the gold standard really isn't all that gold that's the way i've put it to athletes before because it's still subject to human interpretation bias and subjectivity that's only 85 accurate and in uh in in your or in that's recognized i guess it's not in your research but that's the that's the recognized amount of error so let's go back to your app right and kind of getting back to what you display and why you've intentionally chosen to display that i totally derailed your initial thought process so you can go back a little bit karen's laughing because i do this to her all the time yeah but this i think is really it's really key because uh it fits back into why we do things this way as well uh you know the way we report the data is always to report the physiology only uh and then to use the rest as context not to combine it in some sort of score and we do that for different reasons so you know we would show you your heart rate and your hrv and then we would show you your normal range so again where you expect your data to be unless there is a strong stressor or something odd happening right if uh you are within this range then that's just normal day-to-day variability and everything is normal and you shouldn't bother if you are outside of that then uh there is something you might want to be more cautious

about or you then again you might want to try to look at the other data again the context the training and the subjective measures to see what could be driving that uh deviation so we do that instead of combining things together um for two reasons one comes back to things like sleep uh for example as we said sleep stages and sleep in general sleep quality uh is something that we try to estimate right it's not really something that is measured because we measure some parameters and then we use them to estimate sleep stages for example we measure heart rate and heart rate viability and then we try to estimate sleep stages while hrv is actually measured so if we put them together we basically confound something that is measured with something that is estimated and i think eventually we end up with more noisy data instead of better data and i get it that you know we would love to put everything together and get this perfect marker of what is going on but i don't think that's realistic because we will never have the context like the all the context that corinne also was mentioning right the more context the better these tools can be but at the same time it's so difficult to have that context right um they measure a bunch of things and you can enter manually other things but still so many other things happen that you know are just not there um so it becomes very difficult then to get to this um unique score that is supposed to uh include all of this and another aspect there

i think that others do um i think also in auras readiness is like this but also in garmin's um body butter i think they call it where you mix the physiology and behavior right so if you exercise the lot then your score is lower like you are penalized for doing more but that is not really helping me as an athlete or as a coach to understand if exercising more had a negative impact on my physiology or not because my score is lower so maybe for the consumer that's like a way to keep them engaged right you see that you did a lot and your score reflects what it expected to be so that's great but then is my physiology really impacted by that like if i do a back-to-back long run and you know is the second run uh starting in a state that was physiologically not optimal or is it still optimal because maybe i'm used to do that uh and then it's not a problem right so we put together things and then we have this illusion that that's more informative but instead i think we are diluting the information and it's better maybe to look at the single pieces of information not combined together so that we understand what is going on physiologically and with everything else that is maybe behavior but this is a bit how i see it yeah this is why i think that you like we have more information but it's also more complicated not the other way around we have more information and it makes it easier

and what i always try to weigh is is how much is like how much does that complication actually matter which i think is going to be kind of how we wrap all this stuff up because that's the practical piece of it but to summarize it's really simple you're collecting heart rate heart rate variability you're comparing it to normal and you're showing long and medium term term trends and what i mean by long term trends is months and medium turns several days accurate synopsis yeah that's that's it so your spec sheet right on you can fit it on a business card right and that's not to say it's more or less valuable i think it's more valuable i think it's the right value i think that's that that's that's the biggest thing because the the thing that the consumers have to ultimately we weed through is how much utility is in the data that's actually being collected right and you've taken an approach of maximum utility for the the things that you are actually or you're looking at maximum utility and choosing to collect just those things yeah yeah exactly less is more in my opinion in this case yeah and it takes up the speculation i think too right like i think when you add when you give consumers a score it's easy for them to rely solely on that instead of maybe their own personal knowledge of themselves with just the simple data that's being collected instead i don't know here we go we're gonna we're

gonna get into it okay so all right set it up marco so ultimately you're collecting heart rate heart rate variability you're comparing it to normal and you're trending it signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling signaling you to do and because then we'll get into the bigger picture conversation what's the utility in just that what's the utility like if i had no other training information right let's just like play you know weirdness world where we don't know what else is going on if we had no other information to kind of go off of aside from what hrv for training is showing us what fundamentally is that going to tell the athlete and or the coach so in my opinion what you get from this type of data is information about your response so you would have no idea about you know the stimulus and the load even if you know the athlete because it's really about the response and i think that's why it's interesting for example you know we talk about the acute stressors the high intensity training will cause a disruption that will be reflected in hrv if you measure it right after training for example right there is research from um you know steven siler that i think you interviewed as well in the past who showed very well this right you measured before and after a workout showed that more than the

duration the intensity of the stimulus would cause a larger autonomic disruption right but that's right after the workout if you measure you know several hours later in the morning and you know you are used to the stimulus because you know that's training that is appropriate to you you do want to see you don't want to see that um acute suppression right everything should be within your normal you expect to have a positive response so for a good athlete normally um you have confirmation that the process is going well when you look at the data because things should be pretty much always within their normal range right i've seen you know triathletes that you know are maybe top 15 20 in kona and they would not have a single acute suppression until the day after kona you know what i mean it's like that's the day in which they go hard and they train a lot and they train high intensity also the whole year but then they are always in that optimal zone because they take care of their training and they are used to that kind of training and that's what they do it's you know it's their job and if you don't have any uh unexpected event sickness or things like that then you are always in this optimal zone for other people you know there's always other stressors so you see how you might not be in a time period in which certain stimulus in terms of training like high intensity work might not be assimilated well

because you are already in a negative state for example if you see a long-term suppression of hrv that's you know the practical use would be okay now my body is not responding well to whatever stress there is in my life so if i add additional high intensity stimulus for example then i might compromise things worse in the longer run um and that that's a way in which for example all the hrv guided training studies studies in research operate right now right they look at this deviations from your normal when you are below your normal age in terms of baseline so it means you know seven days moving average so not just an acute suppression but a couple of bad days then it will say okay now it's not a good time to add more stress but that's not to say that this should be frequent right this should be a very unfrequent situation it should almost never happen unless again something is really wrong either health-wise or stress-wise because of major concerns it's not what you expect to see in most situations another one maybe is simply um to start a conversation i hear this a lot from coaches also of elite athletes you know people that know everything about the training of their athletes and then sometimes you have a trend that you don't expect things are not going in the direction you think and then it's maybe it's just a tool that that point you use to talk to your athlete about something outside of training

and then maybe you learn that you know there was something there indeed that might have caused the additional stress so oh yeah i stand up till two in the morning working every day i mean clarence you she's shaking her head she's like yeah i have athletes that like five days later they told me they got four hours of sleep you'll last four days it's like why did you tell me that the next day exactly yeah their kids their kids been sick all week they exactly they broke up with their significant other i don't know there's they're telling you at the end of the week not yeah you get that story like a month later and you're like oh by the way by the way this happened five days ago thanks thanks man oh yeah no or like i didn't want to tell you that i was going to be at altitude because i still wanted to do that workout and now i'm calling you to tell you that the workout sucked and i'm like well i wouldn't have had you do the workout if i knew that you were going to be in breckenridge so yeah there's yeah communication it's like a spy right it's like a spy into these other things you know that there's a disruption but you can't explain it but no i mean that's an insightful point marco it's a it's a conversation starter right in a lot of ways and we encourage athletes to be as transparent as possible about what's going on and that's an easy statement to make and sometimes if you have levers that you can pull or tools that you can use to facilitate that transparent that transparency hurry variability being one i think that that can be a good piece of utility for it but fundamentally i know i know i know i kind of cut you off but i kind of want to

wrap this up to talk about the next piece is that the utility comes in you're the it's a it's a as as we mentioned at the at the onset it's a reflection of the stress and you're using and you're using that kind of distillation of how stressed the athlete is to modify training and to start a conversation fundamentally those are kind of the utilities of it yeah yeah i think you know description yeah for sure and i think we are still in a good spot as you know coaches or athletes like at least you have something to play with which is training that can be adjusted because these tools are used you know for research in any sort of other issue uh you know psychological research and any sort of chronic disease and in many other cases when you have suppressions and clear signs of stress it's so much harder to actually make a change and again if you're going through a breakup or there's a lot of stress at work and things like that maybe that's how it is you cannot really do that much about it sometimes while with training is always a bit easier uh you have the choice to make some adjustments so i think that's partially why this is more adopted in let's say the sports industry more than anything else despite the fact that is a global market of stress it's just that you know there is the motivation to do it to measure because you have your objective and then there is the action ability it's a bit more action whether in other cases i want to talk about the art of this at some point because we started we

started mentioning you know seven day rolling averages and i do think that there there's an art form to everything in coaching but this is one area in particular because you're using it in conjunction with the entire kind of training picture so i'm giving you guys a heads up to kind of collect your thoughts on on that but before i know corinne's like oh my god but before but before we get into that we we'd really be remiss because we we've danced around the subject a lot i want to spend a little bit of time talking about the composite scores and you guys know there are no sponsors on this podcast i don't have to adhere to you know anything i'm not trying to you know collect sponsorships or anything like that in any kind of way and i think this is this is one of the areas where that's the strategy and that really kind of shines really kind of shines through because i can authentically say kind of whatever the guest wants to say and whatever i and my other guests want to say uh about kind of about anything but very specifically we're gonna probably a little bit unfairly lump in whoops recovery score and aura's readiness score kind of into the same category although they're telling us or they think that they're telling us slightly different slightly different things as the names imply recovery versus readiness but i want to broaden that lens back out to composite scores we're taking a lot of different physiological information heart rate variability heart rate the time spent sleeping body temperature and we're i'm intentionally using the word

alchemizing in this state in this stage because it literally is alchemy we're taking all of those things and put in giving one number that says on on whatever scale it uses yes you are recovered no you're not recovered yes you are ready no you're not ready and whatever you know green to red light bandwidth exists in between those two points and i've always had a hard time wrapping my head around why i throw those in the garbage bin or trying to articulate why i throw those in the garbage bin so marco why don't we give you the first shot at maybe explaining how those scores come to light first because you've got the you and your role have a little bit of insight into how those actually get produced and what if any information they are actually telling us yeah so i think um let's say that here again we have one major issue which is there is no reference so we don't know what's the ideal reference that you would use for recovery or readiness or anything that is supposed to tell you how you should be feeling today um because again it could be subjective but then also there we know that how we feel subjectively does not always match uh our body's abilities for example on a

given day uh again you're getting sick maybe and your physiology is already impaired but you don't realize it yet you think you're okay and then you're not okay there are mismatches right so uh the first issue i would say is there then the second issue is that all of these algorithms they try to fit let's say a genetic model in which you put together all of these things and they give you an output and that is also again genetic is not specific to you so maybe for you a certain set of variables are more relevant than the genetic variables that are used for everyone like maybe for you as an ultra runner uh muscle soreness should be an input and for someone that is not running or not exercising maybe you know they don't need to use that as an input so these um attempts to capture context still miss much of it and even if they were including more still it would be generic and not person specific so that's also another issue that we have there that is really interesting because i i don't think that gets a lot of um i don't think that gets a lot of uh scrutiny um the individual variability does because we oh we're all unique people and you know that that just gets kind of thrown around ad hoc and it is it is extremely important but what you mentioned on the sports

specificity side of it i think people need to take equal take of note equally when either of these device manufacturers is producing a readiness or recovery score they are they are not discriminating amongst are you a rock climber are you a runner are you a triathlete are you this or are you that and some athletes switch between that but you switch between all those different sports at different times and you and it makes logical sense we might not ever tease this out of the research but it makes logical sense that even if you wanted to have a perfect readiness score you would do it differently for an endurance athlete versus a strength athlete versus a power athlete versus a skill athlete that makes it that makes like coaching 101 sense all the sense in the world but yet that discrimination is is is is in its current form not there and i don't even know how you would do that yeah exactly so that's you know one of the many problems i think of this course uh together with again mixing physiology and behavior right so what you did and what your physiology is you put them together and what do we get there uh is that you know this score is it helping us understanding how we are doing or maybe just creating more confusion on what could be the cause of the issue that we are seeing um what else i think when we look at consumer tools another issue that we cannot ignore

is that when we move away from the raw data things like heart rate hrv temperature even or breathing rate or things like that and we go to this course then we are also they are they are also subject to change right there is a new version and it's different and maybe we are aligned on how this worked before and now it works differently it provides a different output they tune the algorithm or something like that and that could be problematic because then what are we relying on right is it something that uh from this day on we cannot use anymore the same way we were using it before even if we somehow made it work for us because now it's different and we never know when that happens so i think it becomes very problematic in a sports setting or as a coach or as an athlete or when we do research to do things with this course instead of going to the source which is typically the signals that are measured that's why you know all of these tools that you mentioned i think at least all of them provide this data so you can look at it which i think is more informative at least most of them um some of them don't which i find a bit annoying like i have a garmin for running and then i see all their made up scores but i cannot see hrv for example so that's i think it's a pity because it's a device that you could also use you know for research or to actually look at your physiology but you are not able to do it and can only look at composite scores that are built on top of that and that to me are

not informative for you know all the reasons that we discussed so far everybody who's ever had a garmin here's mine it's my phoenix i love my watch i think it's i think it's one of the better showing watches out there but everybody who's had one of these and trained seriously has had the same experience they stop the watch and they see some absurd amount of recovery time presented either you do a one hour recovery run or one hour recovery activity and it tells you to take four days off or you've just done the hardest five hour training activity of your life and it's like recover is normal like everybody has had that experience but i bring that up kind of jokingly is is it breeds a lot it breeds a lack of confidence when it's that egregiously wrong and an untrained i can look at it and go no i'm i'm annihilated i'm not taking 24 hours off i'm taking four days off i'm absolutely annihilated right now our mind says i'm detraining all the time it's like detraining it's like activity or training status detraining and i'm like come on that's not fair so yeah i think there and also i've noticed too like because i i'm altering back and forth between cycling a lot and running a lot because i'm coming back from injury and i found too with a lot of these metrics that like the the way it weights running so like uh like using a whoop or an aura ring or my garmin versus the way it weights cycling like i went for a long ride on sunday and was

destroyed but it weighted that so much less than like a standard you know two hour run for me and it's like they're not it's not a one-to-one it's not a two-to-one but like it's it's so frustrating as an athlete trying to weight some of that with with the tools versus just like knowing what i'm supposed to be doing yeah for sure and you know sometimes it becomes dangerous almost because um you know i've had people reaching out to me like i use garmin again and then after a marathon or something or longing around some things like that i look at the data and it tells me that i'm really stressed and they are super concerned because the algorithm is actually very simple so they actually give it a lot more credit than they should give it because it's simply showing that your heart rate is elevated right your heart rate is elevated because you need a big effort and then it's going to be higher for several hours even with respect to your normal and so this body battery and these other features will always show like this enormous amount of stress which is completely normal after a marathon or any other big effort right so they take something that you know it's very simple and then build all of this on top which lacks context even though you're in the marathon with you using that device still that context is not taken into account right it will just look at your elevated heart rate and then trigger unnecessary concern i would say due to the oversimplification that is

context that you cannot have yeah that's that's key the oversimplification is the context that you can't have because individually most of maybe not all we and that's probably that's debatable most most of the things that are being collected that are getting put into this composite score could have some utility and that's where people start to get super confused and they want one number to kind of rule them all that puts together all of these different things but in reality yeah i would use temperature like as a separate data channel i would look at that and say okay if that's i can treat temperature is that as a coach i can treat heart rate with this as a coach i can treat heart rate variability in its own little channel and kind of figure out what to do with it and marco you used an analogy in one of your blog posts where it's like combining apples and pears but i just came up with a better one for you and i'm not very good at analogies nor did i like them nor did i like them very much so this might get completely butchered but the thing that i think about if we want to stick on the food analogy is it's really combining two foods that don't belong together like chocolate and ground beef like i really like chocolate for what it is right i love i love chocolate i'll eat it for dessert i really like a nice hamburger i really like a burger and i like the ground beef that goes in it but if i combine those two things it just doesn't make sense and i try to make a meal with them it

doesn't make sense like you have to appreciate body temperature because we keep using that for what it is just like we have to appreciate heart rate variability for exactly what it is and resist the urge to combine them into it into one thing that's the point that i'm trying to get with it so my week let's go around the room i'll give you guys my opinion last is this worth are the composite scores worth using with that corinne you get to go first by the way the composite scores even worth looking at yay thanks for throwing me under the glass i've known you for longer corinne it's only fair to marco okay okay so i'm gonna for in full disclosure right i've used the the hrv for training the elite hrv those those tools you know i i grew up basically in a heart rate monitor strap as a nordic skier i have used both whoop and currently testing out a gen 3 aura ring so that's a disclaimer there right so i'm seeing this data um presented to me and getting to be like well what do i actually know about my training and what do i actually know about how i'm responding and the other things that are going on in my life and i still think that it's an accountability tool like i to me that at most is what i'm getting out of using these products is like a oh yeah i tell myself i'm going to bed at

9 30 but really i didn't go to bed until 10 15 type of thing over oh i'm i'm taking this readiness score and i'm going to use it to to decide what i'm doing for training today because i feel like i'm smarter i'm smarter than the composite score yeah no no i get what you're saying my pushback on that is how interrelated is the compliance to the metric that you're getting meaning if you're complying more i'm sleeping more am i getting better better scores exactly exactly because that's the compliance that's the compliance feedback right because i've heard i've heard this argument before marco's probably heard this argument before that these types of things do have a psychological compliance associated with them and whenever that is the case you have to make sure that that association is has a high r squared value right otherwise it's otherwise otherwise you're not motivated to continue the behavior okay so here's an example i on tuesday morning when i woke up my aura ring app said you should be careful today you you like your things are not trending well today sort of thing and i said well i know that i'm getting a rest day tomorrow like i want one more day of stress and then i will take a rest day and so i overrode the readiness score and was fine turns out i'm totally fine so i think that there's like in like like previous conversations that we've had right like i'm not

one to say oh here's this one score i'm putting all my faith and all my value into this one thing and i'm going to make a decision day to day based on that like that doesn't that's not what the science says to me that's not what my brain says to me that's not what the mostly compliant athlete says to me so like that like that's an example from this week of like seeing this on my phone and then making a decision that might counter what it's telling me does that make sense is that so corinna corinna corinna use it as a compliance mechanism as long as it's convenient for you as long as it as long as it fits your whatever you've got going on for the day if i don't want to do the workout then i can say adam well my aura ring said that i shouldn't do the workout today it's confirmation by it's fine on tuesday i really wanted to run tam with my dog and so i did and then i took a rest day yesterday or wednesday so yeah i mean i mean it tells you exactly how i think i guess but no i don't know i get it i get it i followed my plan i get it all right marco you get a shot at this the composite scores i think that um like just to add a little to that i think that that also again speaks just to the lack of context of the device right you know that you have a rest day tomorrow so for you it's easy to do the math but the tool does not know that that's also why even without the composite scores just looking at hrv people that get started with these things they or with anything

that is hrv guided they start from hrv when you should never do that you should start from a plan right you should start from a training plan and then maybe you make some adjustments based on the data but the data should not be the first thing because otherwise what do you do you always go hard until your hrv is suppressed because you are already in a terrible state right so it does not make any sense to operate that way you start with a plan maybe you make an adjustment also based on you know the plan the athlete level and all of that then back to the scores um you know i have my conflicts you know i make hrv for training i advise order so i also like to be honest so honestly i think that the data should be the important piece here more than the composite scores uh you know i use the tool as well and for me there is little use in uh readiness scores recovery scores and things then but you know body battery and all of that that combines things together um i see it and maybe i see a low score and maybe it triggers my interest i'm like why is that right and then i go and i look at the data and the signals and the physiology and was my temperature elevated and you know that was my hrv suppressed or how is my heart rate with respect to my normal but then maybe you know in another tool i just look at the data directly without that extra step which um yeah it's not maybe helping me particularly but

again this is also me uh which probably i have a different experience and i've been doing this like looking at this data i can sign out for 10 years and i understand why these these scores are there right you try to simplify also things for a tool that is supposed to be used by you know millions of people that look at the data and get some insights and not everybody has you know four degrees to look into the data and understand what is happening so totally normal uh and fine and it can work for people like if you are that average person in terms of the um how the physiology and the activities impact your body it's basically the person for whom the tool is made eventually right not the pro athlete not the um nfl team is probably you know uh what we think that happens to the body and to your readiness when the average person does not sleep much based on studies when the average person trains a bit too much when your physiology is a bit impaired and all of that gives you a number and i'm sure many people have um found decent guidance in there it's just for me personally that i don't find it because i come from a different angle and probably yours is much closer to mine that that is yeah so that context marco i think is the most important part because i don't want everybody to like reach out to aura and ask for a refund after this after this podcast no me either i'd be

fired yeah uh you you're in a worse position than i am but here's why here's why and i'm going to emphasize a couple words very intentionally i Jason Koop as a coach find zero utility in the composite scores and that is because i as an experienced coach who i you know i'm going to go out on a limb here i know what i'm doing most of the time i've got the experience i do i do my research i do my homework i can do a better and far more accurate job looking at the totality of the information that's in front of me who the athlete is what they're training for how long they've been training what they're presenting in their post activity comments what their training load has looked like over the last three months what i want it to look like over the last month i can put all of those to come together with the individual variables that i want heart rate heart rate variability temperature and maybe the sleep borders as being the top four there i can put all those together better way better with all due respect to the people who have come up with these scores way better than the algorithms to drive training action that's why i say that as an as a coach as a professional of over 20 years i would throw them in the trash if i were joe schmo out there and i didn't know shit or i was a horrible coach and i didn't really kind of know what i was doing i might find a little bit

more utility in these recovery scores because you know maybe if that you know maybe if that directional arrow gets a little bit better from a recovery score i'm not gonna necessarily say that that's a bad thing and if it leads to you getting better as a coach and analyzing all of those things maybe that's an outcome of it but i i get i get the desire for people to want to take all of this stuff and give them a directional arrow i totally get that i can i completely get that what i'm saying is is me as a as a professional coach i'm more accurate taking all the individual things and all the things that can't that the that the devices can't capture and putting that into actionable information versus an algorithm kind of doing it for me when the algorithm gets better than me i'm going to be the first one to raise my hand and say yep take it over machines like i'm i'm going to sail off into the sunset and let all the machines do my coaching for me so i'm going to be the first one to do that trust me but as of today it's it's and it's not close and i kind of i and i wanted i'm trying to say that without coming off as an arrogant prick but when i look at the council that you could potentially derive from specifically the composite scores it's not close to what i would want to do and he and i think we're going to get into why in this next segment right we're going to talk we're going to talk very

specifically about kind of the art of using heart rate variability in combination with everything so corinne i put you on the spot earlier i'll put myself on the spot this time we'll go the opposite way around marco can be the sandwich in the middle of everything but there is all of this information that we as coaches and we who advise athletes to do things can take into consideration and i kind of want to first get a little bit of a rank order and then we can get into where heart rate variability is because i think that that context is key so when i'm personally evaluating what i want to do with an athlete i first look at good training architecture i think that that matters the most if you get good training architecture right most of this stuff kind of takes care of itself to be honest with you second i look at the individual athlete what are their strengths and weaknesses what are they obviously specifically training for what does their previous training look like do they respond well to high volume high intensity three days a week five days a week that kind of stuff that individual variability i then look at the workout performance so just day to day week to week are the intervals going up down sideways and how does that compare to to to what i uh what i expect in conjunction with the workout performance so these are two things that are in parallel of each other

i look at the post activity comments so those two things are like side by side so let me recap the order really quick smart architecture individualization side by side workout to workout evaluation and post activity comments and then after all of those and this is a rank order prioritization after all of those i'll take the physiological measurements that we're seeing from heart rate during training heart rate at rest heart rate variability temperature and sleep border and i put all those into my own little basket and i don't i i can't tell you there's not an algorithm because it's me looking at it i put all of those and i kind of give them equal weight across the board and if i see the trends predominantly pointing in one way or the other and i'll explain what that means in a second that's when i'll adjust the architecture at the top so if i see a post activity comment or two or three in a row hey i'm tired hey i felt like crap hey whatever i'm gonna bring the training down a little bit let the athlete get his or her legs back underneath them if i see a little bit of that combined with a little bit of the heart rate variability which is what we're going to talk talk about point in the same direction the athlete is really ready to take on work like i feel awesome today okay and the heart rate variability is kind of showing me the same thing okay let's pour it on let's like you don't have enough training load let's let's add let's

add a little bit kind of with the within reason that's how i'm distilling it down and i go through that order pretty much every day and every week with every with with every athlete but my point with it is is the order and in this basket of the physiological stuff of which heart rate variability is one and it might that might be the dominant player if i'm thinking about it out loud right now within that whole basket that's how i'm kind of taking it and that's the way i'm describing it and this is why people want a composite score is it's super esoteric and it's subject to my my skill essentially of being able to put those in the right context and to give them a word that corinne used earlier the the right weight so marco your point of it's heart rate variability guided not dictated is very that that phrase i think is very much true in the way that i go through things where i'm using it as a singular point of that guidance in combination with other things that absolutely have a whole lot more weight to them so that's how i do it personally and that's how i advise athletes to do it marco you get the next you get the yeah that makes perfect sense um and you know the the reason why also we might want to use this data is it starts i think with awareness right so to start um maybe fine tuning also our subjective feeling the perception of how we

are doing and all of that sometimes is a process depending on the level of the athlete as well of course typically the higher level you know the better that skill um so it can be more important maybe to look at that with a recreational athlete that thinks they need to power through all the time and you know they wake up at five and work out because that's the only time they have and all of those things can become more problematic than for someone that does it as a job and does a bit more flexibility and of course that also comes with a lot additional stressors right typically even just making enough money to eat for example because you know professional athletes outside of some sports still struggle a lot with the day-to-day that is being a professional athlete so everybody has stressors it's not that one way or the other is easier but that's also why it can be relevant regardless of the level but in a different way so the process might change a bit but i agree 100% that training is the first thing and feedback of the athlete on how that training went and then the physiology might help explain some of the issues or give you some feedback there so that you are a bit more confident on the decision you make if everything physiologically looks perfect maybe you power through another day and see if that was just a bad day if everything looks bad physiologically on top of the

negative feedback then maybe you make the change a day earlier so things like that is just the process of looking at these things and indeed as you say it's a bit of work right it's not a readiness score it's you looking at it but i think that's again also the awareness piece like as an athlete or self-coached athlete that might want to look at these things maybe it's more useful to try to do that process uh and actually i had coaches telling me you know we take the hrv measurement but we really like the questionnaire because that's when the athlete self-reflects and they're like okay how do i feel like am i sore am i fatigued these kind of things like the self-awareness piece sometimes needs uh yeah a little help so that we start working on that and that's how these things i think can help a bit or can be integrated in the process but yeah in my view your process is also what normally uh yeah i would um recommend doing to coaches and not least looking at these things the questionnaire piece it's interesting that you've got that feedback because some of the feedback that i get i'm very rpe based and i use a lot of the subjective feedback a lot and i hear this in the athletic community a lot and it's mainly in like the high performance community um it's just that the athletes they want to train so hard so they're going to lie to you right they're not going to be honest about how sore they are

which to me i'm like that's bad coaching because if you're looking at everything you can see when the physiology dissociates from the physiology and the performance dissociate from the questionnaires or the subjective feedback and then you can go back to the athlete and say well why are you telling me that you feel awesome but you're performing like shit like why like tell them and and all of your physiological metrics are you know in the tank like explain that discrepancy to me and so usually that's an art of coaching thing to kind of like bring those things out i've always looked at that feedback and just go and just been like well you're not presenting kind of a complete like picture of actually what's going on i guess is is is is what i'm trying to say okay corinne you get the last shot at this one yeah i would say that athlete needs to trust their coaches more and maybe have an open conversation with them too um but that's neither here nor there i guess um one okay so i'll start backwards and then go forwards and maybe i think with the physiology piece listening to both of you talk about it i was thinking well maybe in a way you're like you're using the physiology to test the null hypothesis right you're saying like if this training is working what is like and and then saying like does the hrv match that right like is the athlete tired what does the physiology say is the athlete can the athlete take on more what does the physiology say you're kind of like it's almost like it's confirming what you're doing or or proving it wrong um as a way to kind of like i don't know go

back and forth between like turning the dial a little bit and i think that i agree you know almost wholeheartedly with your kind of hierarchy of how you approach coaching and i'd say the only thing that i probably do differently is that i probably flip-flop the the human component and the architecture component because i've got my shift work athletes i've got my parents with joint custody athletes i've got my uh i've got my parents or my my athletes that are parents in general or they travel every other week for work or their home you know they they work from home one week and they work from the office one week and they have to commute to the office so i feel like in my mind i almost had to take those human like the human aspect of it first and then say how do i fit optimal architecture architecture with that with those variables as opposed to saying this is optimal architecture let me push your life into it somehow so i think that's probably the biggest difference there is that i have to start with is there any are there any huge life things that scream at me their job their family structure whatever that might be and then can tweak the architecture to to meet those needs because if you're working with a nurse who works three or four twelves a week like optimal architecture isn't going to look the same for that person as it does for someone working a nine to five so i think that that to me is like i start there and then i can make the architecture fit that individual athlete you know it's funny corinne how i act like i practically do this we're going off the

rails of this conversation by the way how i so with the shift workers i coach a lot of shift workers i put the i've designed the schedule as if they're not a shift worker and then i move the things around training peaks like literally that's what i do like literally i put it out there i'm like okay they're not a shift worker yet okay this is what i want to do okay where do i how where are the train traps that i need to avoid and i just move the puzzle pieces around and then i make the you know the puzzle pieces bigger or smaller kind of whatever whatever kind of whatever whatever i want to do at that point but for the like the 20 minutes that i'm doing that i'm going please don't log into training peaks please don't log into training peaks you're going to see something and be like wait a minute i have a 20 hour work day and you're asking me to do a seven hour run on that doesn't like that doesn't actually work but literally mechanically that's that's what i do sometimes i'll do it in my head but most of the times i do it on training peaks yeah it gets it gets to the same it gets the same end but like i have to think about it that way first like i have to put in like here's when they're working how do i make it work or here's when they have their child here's how i make it work because like to me it's like i don't i don't want them to see me putting it on the wrong days that's when you that's when you use the blinding feature coop and you have it all i know and then you i need to i need to do better with that i need to do better with that all right let's bring it back to heart rate variability so we've kind of got the here's the interesting part right we've got this we've got this physiological metric that we could take it's easy to take it might not fall high on the rank

order as we're all presenting it but it certainly can be very powerful and so part of the art and the interpretation is knowing when it's giving you a signal to adjust we've already gone through the thing through this aspect that we've got to look at things as otherwise specified right but i think we can to help the listeners understand maybe we can put some parameters on that marker you know marker i'll let you take the lead on this because you've got the most experience there is uh i'm going to ignore these things and i'm going to take into consideration when these happens like that like the very very clear cases why don't we kind of start with what are the most clear indicators of when we should be using heart rate variability to change training and then kind of move into the murky like the like the murky waters like almost think about it as if if i saw that it might jump all the other rank order things that we just mentioned just because the signal is so strong right it might not be it might not be tops on the list because we're getting such a strong signal we're going to boost it up for this particular piece of evaluation yeah all right so i think uh a couple of things one is always to consider the change again with respect to your normal and you know technically that might be you know one standard deviation in the data and i

mention this just because it it's not a score that is just a bit lower so that's something that you know you need to use a tool that allows you to understand this and when a suppression is really a big suppression so that means that to have an acute change which is not just normal day-to-day variability there is actually something that was different there now in this context still the one day acute change that's for later so for the second part of let's debate if there is something to do or not while what is a stronger signal is when you have a couple of this and that's typically if the tool you use also shows you a baseline or seven days moving average so basically your recent trend not just the daily score not your long-term normal but a recent trend if you look at that and that is below your normal that can only happen when several days showed very high suppression so very poor physiological profile that is in my opinion typically a red flag it should be very uncommon uh over a year i mean unless your periods in which you have poor health or you're sick or something is really wrong this should happen really rarely that your baseline is below your normal for several days or even weeks it can happen uh you know simple examples like in spring i suffer headaches and allergies and things like that and

obviously i keep training but you know i don't feel well i don't do high intensity typically it's reflected in the data even if you have such a period of um high stress reflected in the data and or maybe you also don't feel great this period can last maybe weeks right so in a period that is so long and you're not sick you're just something is just odd again seasonal allergies or headaches or problems like that that can happen to many of us right in that case on the day-to-day basis you still are going to have better days and worse days right so it's not that you know your training is frozen for two months right but i think in that case you are just aware and you need to accept that it's not the way it was before so that's how it helps you it helps you working with that awareness and knowledge and acceptance of the state you are right now which maybe is not where you want to be but it is where you are and the fact that you have better days and worse days and you can still do you know your sessions on the better days and things like that but the signal in that case is strong and i think that's what we see in practice but we also see it um and in the hrv guided studies where now they trigger the intervention only in that case they don't do anything if you have just a single suppression on a given day um and the outcomes are quite clear uh meaning that you know typically performance is either

not impacted or improved physiological uh data during uh lab tests is also either improved or not impacted which you know obviously it's small adjustment so yeah i would be ridiculous to think that you do this and your performance becomes like a lot better right um to me if there is no decrementing performance because you skipped maybe some key sessions that's really a win because you didn't went in you know a negative state or you didn't you know got injured or got sick because you went really hard when your body was really in a poor state but still the research that has been published shows um typically uh positive results and again we all know also that typically that's how research works meaning that maybe there was another study that didn't show that and didn't go through peer review so we also need to think about you know how science works but in general i think that's a strong thing uh chronic suppression baseline below normal uh several days with very poor hrv data i think that's something uh you need at least to think about in terms of deciding what to do during that phase uh with your trade um let's say yeah that's a good starting point one of the hardest things that people have one of the things that people have the hardest time wrapping their head around is that you've got to sit back and wait for the data to come through or for more readings to come through day after day

like when you see one when you see something anomalous or something outside of your standard deviation it's like okay we're going to take note of it and then you see another one okay am i going to take note of it again or am i going to do something about it and then you see another one and like okay am i going to take a note again all right am i going to do something about it and how long do you run that out before it becomes problematic that's where the art kind of becomes the thing with things and once again we're kind of treating heart rate variability like it's on this like data island that is driving everything and it's really not but i honestly have a hard time struggling with like how many like when does when does it need to change right and i don't know the right answer to it right we say a few so it's not one and it's not 10 but is it three is it four is it five and i i mean that that courage shakes her head she's like she's like yeah i'm thinking the same thing right now i don't i don't know what it is for a baseline change i would say that's gonna require three to four bad days otherwise not gonna happen yeah yeah but so in that sense is it too much of a lagging indicator like because i usually get i mean from a practical perspective i usually get another directional arrow that's higher up on the food chain for me before i'm gonna get that fourth day that's practically how

it it as from a coach when i'm looking at things every single day practically that's almost how that's almost how it works every single time so i don't get the fourth one usually i get something else that's stronger and then i combine it with like the first two or maybe three and say okay we're gonna tweak whatever's whatever's coming up tomorrow based off of all that i never i never get it to the point where it's going to solely dictate what's going on yeah yeah i mean makes sense it's uh i think that's where the research shows all its limitations it's not real life like you have these protocols with these very clear you know things that we do um but that's not how you would use it and a counter example to that is also the acute change like today i wake up and my hrv is highly suppressed and i don't feel well so i'm sick do i wait four days before i make a change because i need the baseline to go below normal that would be very stupid right so sometimes the acute score is very telling and you don't want to ignore that but again you have that context right you know that you know maybe you have a fever as well so you're not just not gonna wait that long so um yeah that's indeed as you call it more the art of trying to integrate this information i would say we all have brains we all have brains and we can use them and the the machines don't have brains and that is our

difference right now oh no that's what everybody's pitching though machine learning we're gonna leave it i think that's a compelling place to to to stop it we've got brains and we need to use them because i i i i do think marco and i think you i think you would agree with me on this so if i'm putting words in your mouth please please correct me is that with a lot of the information that we are gathering from the wearables yes it can make the training more accurate and therefore more efficacious but the the partner in doing that is the person that can create the right analysis from that information to drive what the human is actually doing and whether that's an athlete or a coach or kind of a team of people ultimately that's what brings the information to its maximum amount of utility in a do this don't do do do that perspective is there's got to be somebody that is there that understands the information what it's trying to communicate and what actions need to be drawn from that in order to optimize performance for sure yeah i agree to that corinne do you have any final statements besides i just want to do i'm gonna leave it yeah yeah i just want to do tam irrespective of what my heart rate variability is you know creatures of habit right no i think that i think that people like the tools are designed with

a specific audience in mind right and that might be you and it might not be you and i think that's important to evaluate when you think about adding a new tool to your toolbox and then the next step is that you have a brain or your coach has a brain or you guys have brains when you combine them together and i think that that's an important tool to utilize right you can exhale you can be critical um and it's up to you really to decide you know how you're going to utilize that information and there's tons of information out there to help you make those decisions 100 we're very well put corinne okay we're going to say goodbye for now that was really fun marco thanks for putting up with us all the way all the way from the netherlands amsterdam right yeah correct thank you thank you for having me it was pleasure where can people find you uh on social media you're one of you're one of my favorite you're one of my favorite followers on twitter i probably you probably are in the top like 50 no seriously like the content that you put out is fantastic you summarize the research really well you do it accurately and you do it uh in a way that most people can even idiots like me can like read and understand and draw utility from so where can people find you thank you uh yeah i think twitter is right now the main medium i use for um you know sharing um yeah the work and then links to articles and things we write so that would be at altini underscore marco and we can just put it later in in the notes and that's easier for people all the stuff will be in the show notes links to marco's social media and links to the research

that we talked about really appreciate you guys coming on thanks a lot i think the listeners have gotten a lot out of this all right folks there you have it there you go much thanks to marco and coach corinne for coming on the podcast today i hope everybody got a little bit of a glimpse into what we actually think about some of these recovery and readiness scores as well as what the real utility and heart rate variability is you guys go give marco a follow on twitter he's one of the ones out there that i think is worth it he puts out a lot of high quality information and he's also very passionate about this subject and i think that if you're interested in this area everybody would be well served to go and give marco a follow on a somewhat unrelated note i know that a lot of people have been reaching out to me and sitting on the edge of their chairs waiting for the second edition of my book training essentials for ultra running to come out truth be told it has definitely been delayed and that is one of the faults of self-publishing and me kind of not knowing what all of the time how much time i need to do certain things at the at the part of the process it's close that's all i can say i've got all the files ready they're all uploaded to amazon and wherever else they need to be uploaded to it is simply a waiting game but trust me you all will be the first to know that it is available if you follow me on social media i will let you know the instant that you can purchase it on amazon because i know a lot

of people have been waiting to especially receive it for the holiday season whether it's for themselves or for one of their training partners and friends so it's close that's all i can say i'm not going to give a date but it's really really really really close and i actually have a podcast already queued up that introduces the next edition of the book that i'll release whenever the book is released so i appreciate you guys checking in on it and uh just let it be known that i'm doing everything that i can to get this thing out in the wild as soon as possible i appreciate the heck out of each and every one of the listeners out there you guys and as always we will see you out on the trails you