Sports Science Q&A’s with Dr.Davidson – Volume 2 – Analytics
Last time we discussed a bit about sports science in general. Let’s go deeper. Many public and private sector organizations have been, or are delving now, into higher levels of technology and deeper levels and data analysis for player monitoring, program design purposes, etc. to the point where now many organizations hire a staff person who does nothing but data analytics for the coaching staff and trainers.
1. Generally speaking, what role do you think technology and data analytics play in our jobs as trainers/coaches and for the sport/skill coaches?
In the first installment of my answers to your questions, I stated that sports science is grounded in the empirical method, and the empirical method is walled in a quantitative paradigm. The end stage of the empirical method is making evaluative statements about observed and measured phenomena. These evaluative statements are based on comparison to norm and criterion referenced perspectives. Ultimately this means that our ability to make any type of declarative statement about the merit of something is entirely based on comparing numbers.
This quantitative paradigm is only as strong as our ability to make accurate, reliable, and precise measurements. Advances in technology provide coaches and scientists with better tools to measure phenomena. The better the tool with which you are using to make measurements, the more trustworthy the data. Testing environments where coaches, trainers, or scientists consistently make accurate measurements on relevant bio and performance markers ultimately leads to a situation where coaches, trainers, and scientists have a clearer picture of what is going on, and the ability to appraise, formulate conclusions, and decide on future courses of action for the athlete are improved.
What we must remember is that the ultimate goal of the coach, trainer, or scientist is to be the best possible informed decision maker for the individuals they are working with. An informed decision maker tends to choose courses of action that lead to success more often than decision makers who make choices based on whims. This is true whether we are talking about an informed decision maker who watches the weather forecast the night before and brings his/her umbrella with them on their morning commute vs. the whimsical decision maker who is clad in flip flops and a tank top for the day, or sports performance coaches who test the movement qualities and capacities of their athletes prior to designing a training plan vs. a trainer who puts every person they work with into a generalized training plan. If the goal is to increase the movement qualities, capacities, and sports skill of an athlete, then the informed decision maker coach, trainer, or scientist will utilize every tool that will help him or her gain a clearer picture of the current state of the athlete in question. The informed decision maker will reduce the chances of injuring or not improving the performance of the athlete in the short term and long term training plan.
Technological advances in the equipment used to measure the performance of athletes in specific tests can be very helpful; however, what we must keep in mind is that tests are only tools used for the purpose of measuring something. Every tradesman or craftsman has a tool box. Each craftsman’s toolbox contains the tools best suited for the jobs that individual is likely to have to perform on a daily basis. Often times these tools are simple. Fancier tools do not necessarily need to be utilized. Hammers, screw drivers, and wrenches have been around for a long time, and a good craftsman is able to utilize these devices with incredible precision, accuracy, and usefulness. The best tools are those which are efficient, easy to use, and give the individual wielding them the results they were hoping for. In the modern world of sports science, the technological advances have in many cases bogged practitioners down with excess data. Perhaps the potential for discovering something previously unknown in this data exists; however, the route to getting to this information is often convoluted, lengthy, and one that can sometimes be a greater hassle than was truly necessary.
Technology is certainly not going anywhere. Generally speaking, technological advancements have ushered in some unbelievably useful things. Accountability and informed decision making has improved, and new discoveries will continue to be made which will reshape what we know about the exercising human body. Coaches, trainers, and scientists should utilize technology to the best of their abilities. In doing so though, they should keep the following in mind: 1. Know exactly what it is that you want to measure 2. Know what the right tool is for the job 3. Be able to compare the collected data to something 4. Interpret the findings carefully 5. Continue to master this process.
2. Do you do anything with training quantification?
I coach athletes who train and compete with Springfield College Team Ironsports. We are the only mixed resistance training college team in the country as far as I know. We have strongman athletes, bodybuilders, powerlifters, crossfitters, and a Highland Games athlete in our mix. My main focus is on the strongman athletes because they make up the majority of our team, particularly from a competitive standpoint. The common denominator that brings all of my athletes together is iron, and the metric that is most important is load. Now that being said, I am tremendously concerned with the movement quality of the athletes on the team. Based on this concern, if we find that certain exercises are causing problems and leading to regression of FMS scores or pain, then we delete those exercises from the athletes’ playbook.
Our team trains at different facilities, but the primary facility that we train in is a collegiate weightroom. We do not have access to strongman equipment in this weightroom. Due to the lack of contest equipment, I have to use other training lifts as indicators of training status. The four lifts that I have chosen to be our cornerstone/indicator lifts are the deadlift, the push press, the single leg contralateral loaded deadlift, and the dumbbell clean and jerk. I believe that these are the four lifts that are closest in terms of specificity to the demands faced in strongman.
My methodology for monitoring performance in these four lifts is something that many may find unusual. On training day A, our athletes pair the deadlift with the dumbbell clean and jerk. I like this pairing because you have a slow bilateral pull and a fast unilateral push involved in the same training session, which I view as juxtaposing one another and creating balanced development. On training day B, our athletes pair the barbell push press with the contralateral loaded single leg deadlift…again, similar juxtaposition of movement and velocities. We perform training day A and B twice a week. In every training session for A and B, the athletes begin their resistance training session by performing a 1RM test for both lifts. I ask the athletes to not tap into emotional arousal for any of their test sets, and I ask them to try to find their max within a maximal of 4 sets. The athletes have become extremely adept at this process by now, and most of them find their daily 1RM within about six or seven minutes for both lifts.
So, yes, I ask all the athletes on my team, including myself to hit 8 maxes a week on four pretty big lifts. Why would anyone do this? Doesn’t this cause CNS burnout? Well I don’t really know what CNS burnout is, but I do know that the Bulgarian weightlifting team maxed the snatch and the clean and jerk six days a week all year. I do this for two reasons. First, I want the athletes to build the skill of performing 1 rep max movements. If you do not believe there is skill to this, then you have never competed in a strength sport. I want them to develop this skill in an emotion free environment. The athletes understand what this means. It means no yelling, no slapping each other on the back of the head, no pacing around and snorting like a bull, no shaking the bar. You walk up to the weight like a professional, execute the technique with no expectations, and accept the result whether positive or negative. Second, I want the training percentages for the athlete to be as precise as possible. 6 sets of 3 @ 88% of what?…of what you were able to handle today in a controlled environment. In my mind this ensures the proper stimulus for the goal of the training session within the phase of the training plan. From what I have seen, strength gains are fairly unpredictable, but when they happen, they are often substantial. This is fairly obvious when thinking about 18 year old freshmen, who seem to increase strength session to session, but from my experience this is true for my seasoned lifters as well, and some of these young men are deadlifting in the mid 600’s to mid 700’s (belts and straps usually) with no lifter at a body weight higher than 240 pounds.
3. What metrics do you think are significant in indicating fatigue?
In terms of what I believe good indicators of fatigue are, I will answer from a theoretical basis and from what I do on a practical basis. Clearly, measuring heart rate variability with the Omega Wave is the gold standard for assessing preparedness. Heart rate variability is indicative of the status of the supersystem in the brain that assesses the total stress load imposed on the organism. The supersystem is primarily made up by the ventromedial prefrontal cortex, the amygdala, and the medulla. When this trio of structures in the brain interprets the inner and outer world as one that the organism is in control of, then in general, HRV stays high. When the environment is perceived as one that is unfamiliar to the organism or one that is highly unpredictable or out of control, HRV ultimately will measure as being low. Low HRV is associated with an increased likelihood of sickness, decreased performance, and increased risk of injury during training or competition.
I do not have access to any device that can measure HRV currently. This does not mean that I cannot take advantage of the underlying concept that governs HRV though. Ultimately, the most important variable related to increasing HRV is the feeling of being in control of one’s self and one’s environment. What increases feelings of control? Awareness appears to be the key. How clear do you understand what is happening around you or inside of you? In truth, knowledge is power when it comes to HRV. How well does the athlete understand the technicalities and the tactics that are associated with the movements that you are asking them to do? Do they know what the purpose of the program design is? Do they know the length of the phase they are in? Do they recognize that the changes and fatigue in their body near the end of a phase are normal?
Based on this knowledge, the most important thing that I do with my athletes to monitor their fatigue is talk to them. I have lunch with my athletes every weekday at school. We talk before and after training sessions. Many of them I teach in the classroom as well. In short, I spend as much time with my athletes as I possibly can. I end up knowing them very well as people. I can look at them and see as much information as any device could ever give me. Would this work in every coaching situation? No, as it would not be feasible in certain situations. If you have hundreds of athletes then you simply cannot get enough interaction with them to gain such familiarity. In such a circumstance, a paper and pencil survey may be a great idea. From what I understand, psychological mood state is one of the earliest markers that athletes are heading towards overtraining conditions. Athletes whose mood state falls within what sports psychologists call the Iceberg profile appear to be fresh and in a good place with their training status. Athletes who display the Inverted Iceberg profile are likely moving towards overtraining.
In closing, I will say that I have had very poor success in trying to assess preparedness with performance measures. Trying to gauge levels of fatigue with 1RM testing has proven fruitless. From my experience, it is nearly impossible to kill the 1RM no matter what you do to someone. I have also not found jump tests to be tremendous indicators of training readiness for the day either. I would recommend that coaches and trainers be very careful in making major alterations in the design of a training day based on jump height or 1RM testing results.
4. More importantly, what metrics do you think are significant indicators of adaptations to training?
This question is difficult to answer because it does not include a specific type of athlete or sport; however, I believe that I can still answer this effectively. I will start by talking about the athletes who I coach, and I will end by talking about athletes involved in more common field and court sports. Prior to answering, I would like to point out that I think this is an excellent question thematically, because I do view the purpose of training as a process of trying to stimulate the appropriate adaptation in the athlete for the specific demands the athlete has to face in competition.
The sport of strongman is a weight class based sport. I do not coach any athletes who are heavy weights, so tracking body weight is critical simply because the athletes cannot afford to not make weight for the contest. Based on this, the most important indicator that a positive training adaptation has occurred is increased strength in competition performances and the four cornerstone training lifts previously described without significant increases in body weight.
In this paragraph I am going to say something that may initially appear to be contradictory to the last sentence of the previous paragraph; however, I hope that I can adequately explain my views so that I can shed light on something that I believe to be critical. During specific phases of training, the marker that I am hoping to see is a drop off in performance. Again, I am examining the 1RM in the deadlift, push press, contralateral loaded single leg deadlift, and the dumbbell clean and jerk. When my athletes are engaged in a concentrated loading block, at some point, perhaps at the end of week two or three, I am hoping to see their numbers start to decline. I base this view on my interpretation of Zatsiorsky’s Fitness-Fatigue theory.
Zatsiorsky has proposed a two factor model of adaptation that appears to be a more complete explanation of what happens to the human organism during training compared to the single factor supercompensation model. We typically only think of training adaptations according to the idea that the athlete trains and depletes the systems of their body, and then after recovery, a new higher level of homeostasis is reached. If the athlete conducts the next training session while at this new higher level of homeostasis then they will display increased abilities. Based on this model, coaches believe that if the athlete cannot outperform or at least match their previous training session’s numbers, then the program design is incorrect. In contrast, Zatsiorsky has proposed that the instantaneous ability of the athlete to display ability (preparedness) is based on the net of their fitness and fatigue. Fitness is the true underlying capacity of the athlete within a specific realm (limit strength fitness, speed-strength fitness, power endurance fitness, etc.). Fatigue is the degree to which training induced stress masks the ability to display fitness. The greater the fatigue, the greater the inability to display fitness; however, this does not mean that fitness is gone, or that fitness was not improved in a training session that appeared sub-par. Also, based on this model, the greater the increase in fatigue from training stress, ultimately the greater the potential to increase fitness once the fatigue is removed from the picture.
Based on my interpretation of the Zatsiorsky Fitness-Fatigue model, I want to see fatigue mount to high levels at various times of the training process. If I am truly driving what will result in a positive adaptation, I need to see decline first. This process appears to be intimately tied into the endocrine system’s function in the human body. The role of the endocrine system within the organism is to respond to deviations in homeostasis. The stress of exercise causes deviations in homeostasis. The training principle of overload is also intimately tied into this concept as well. According to the overload principle, the body must depart from homeostasis in order to incite the internal repair mechanisms which will ultimately rebuild the system at a higher level than where it previously existed. The further the body departs from homeostasis, the greater the ultimate repair response. Sometimes this repair response can be delayed (the concept of delayed transmutation); however, it appears to always come. So I need to know that I have in fact driven my athletes from their comfort zone. I need to see their performance decline from time to time.
There are other areas that I examine with my athletes to try to get a sense of whether they are adapting to training, but I will end the discussion regarding strongman for now, because I believe that the drop off phenomenon is tremendously important, and I do not want to undermine its relevance in this answer by delving into any more specific topics.
In regards to traditional field and court sport athletes, the answer regarding what are important metrics for adaptation indication is that it depends. While I always find it annoying when I hear other people repeatedly answer questions by saying it depends, in this case it really does depend on the individual athlete. Are we talking about a college freshman baseball player, or are we talking about a 15 year veteran in the NBA? Now, even though I have taken the route of being vague at the start of this paragraph, my next statement will be extremely specific. The best metric for training adaptations for an individual athlete is seeing an improvement in whatever it is that is the ultimate weak link for that athlete.
Determining the most important metric comes down to testing results. Good coaches collect data from a combination of subjective and objective sources. Often times the subjective sources of information are overlooked. Coaches should trust their eyes, and they should also have conversations with athletes to get a sense of what the athlete believes are their strengths and weaknesses physically. This practice may save time in constructing a battery of tests, and it also empowers the athlete to feel like an active participant in the process. When objective data is collected from the testing battery, the coach should have some idea about the hierarchy of test results. Movement quality is generally thought of as the most important objective testing variable. In regards to quantitative performance measures the weak link is often most easily seen by comparing test results to norm referenced perspectives for other athletes in the same age level and same sport.
5. Relative to the above three questions, what data visualizations do you find most effective/telling for coaches?
If you are looking for drop off or for improvements in realms of fitness, I would recommend doing a legitimate statistical test, like a T-test on Excel or SPSS. I think that would be the easiest thing to do. Simply compare scores from an earlier time in the season to a later time in the season. See if there is a significant difference between scores. Sometimes there can be fluctuations in performance that are not really meaningful. You cannot possibly know if fluctuations are statistically meaningful or not unless you use a legitimate test. Chaos theory tells us that there are quantitative windows of normalcy for all measurable phenomena. Nothing in the world appears to be truly static. There is always variability, or what you could think of as good days or bad days. The difference is whether something truly has changed dramatically. This is where statistics comes into play. Statistics can tell us that something is in fact different, and the difference in measurements is 95% or 99% not due to chance.
Once you have determined if something is truly different from something else based on statistical significance, feel free to make charts and graphs and whatever else it is that helps you visualize things. I would simply caution coaches to not become to exuberant at data visualizations unless some sort of statistical technique has confirmed that there really is something there. Scientists understand that finding significance is easier said than done. This leads scientists to being very cautious people in the way they interpret data.
6. How do all these metrics implicate performance? Should they?
What we must keep in mind is that metrics are only as good as the phenomena which we have legitimately recognized in the world around us and the tools with which we have attempted to measure these phenomena with. If we do indeed identify valid measures and we can measure these things with precision, then metrics can be devastatingly effective. Yet, it is important to understand that we have not identified everything that is relevant yet, and we probably never will, so we must remember that metrics can never tell the whole story.
The ability to grow food increased dramatically with the advent of fertilizer. Fertilizer is NPK (nitrogen, phosphorous, and potassium). If you put a seed in NPK, it will grow. This caused early scientists to conclude that this was all that was necessary to be present in soil. What we have come to realize is that natural soil is rich with ingredients that we previously could not even measure. When you grow food in nothing but NPK, that food is in many ways malnourished. We thought we knew it all regarding the science of soil, but as it turned out, we did not, and we still do not know it all in that area.
In terms of metrics as markers for performance, I see a parallel with soil science. We used to believe that sprint times, agility performance, 1RM, and jump heights would tell us everything. We were wrong. Then movement quality testing came along and the theory behind that form of testing seemed to explain some of the missing links in the traditional performance quantity testing paradigm. In my mind, the problem is that the movements of most field and court sports are so unbelievably specific that they cannot be mimicked close enough in sports performance testing environments.
The lesson of specificity is something that I have become intimately and directly aware of with my foray into strongman. One would think that a log clean and press would be fairly similar to an axle clean and press. They are not that similar. If you do not train the specific movement with the specific implement in that sport, when the contest comes, that event will be one that you will struggle with. I have learned the hard-way that a car deadlift and a straight bar deadlift are not the same thing: that different types of loaded carries are completely different from one another: that a barbell clean and a stone load are not the same thing. So if these lifts, which are kinetically and kinematically almost identical to one another could vary so much, how could we really expect weightroom numbers to translate over to sports performance prediction?
The one thing that is sticking out in my head right now as I write this is something that I heard Bill Hartman talk about. He was talking about the importance of taking still photos of athletes while they are in competition performing their sports moves. He showed side by side comparisons of high level performers, like Adam Vinatierri kicking a field goal against a lower level kicker. He showed that the joint-by-joint function of the high level performers appeared much better while they were performing their sports movements compared to the lower level performers. Bill Hartman also talked about how fatigue changes everything, and that we should get a sense of the way the person moves from a joint by joint perspective when they are fatigued as well. These two simple procedures stand out to me as being incredibly brilliant recommendations. We need to see how the body moves in terms of the most specific demands it has to face, and we need to see what breaks down from a movement quality perspective in the presence of fatigue. Are these, “metrics”? Well maybe, and maybe not, but they are invaluable pieces of knowledge that may bridge the gap between the weightroom and the field of play.