The Data Revolution in Training

Heart Rate versus Ventilatory Threshold

We try to train the best we can with the tools we have.  What is the ‘best’ training?

The challenge we have had over the years is that there is a gap between the available data and what we need to measure to find ‘the best’.  It would be great if we could wire up a real-time feed of all our body’s vital bio-chemical-electrical outputs, but we can’t.  We have to work with what we have.

The good news is that, with what we have, we can get close.  We can get very effective training with what we have today.  The bad news is that it is still an approximation.  The really good news is that we are on the verge of getting much better data and with that data, much better tools for training will be enabled.

But, first, how did we get here? And where exactly are we?

The history of training has been a history of approximation.  Challenged by a lack of perfect data most coaches do the best they can with what they can get.  Historically this has meant treating the human machine of the athlete as a black box.  A coach would try a certain input, like a level of training, or a specific cycle of training, or a specific type of training and then measure the outputs.  They could correlate inputs to outputs.

This has led us to where we are today.  Coaches over the course of their training lives discovered certain types of training and certain cycles of training produced better results.  It is a valid form of experimentation, but due to the complex nature of the human black box it is also rife with questionable conclusions.

Chief among these questionable conclusions is the tendency to deconstruct to a specific input and making sweeping extrapolations about that one input.  When you hear anyone saying ‘the most important thing’ you can do is X, whether that X is a diet or a training element that is over-simplification.

What this approximation has allowed us to do is to come to general conclusions.  If you train more you get better results.  If you train harder you get better results.  But how much more and how much harder?  Where are those diminishing returns?  How is it different for each athlete?

I can confidently tell you, as I often do, that if you have never done any quality speed work as part of your training, if you add that to your training you will get faster.  That’s a generalization.  I can also say, with confidence that if you’ve been training in the 20-30 miles per week range for your marathons, loading that up to 50-60 miles a week will dramatically improve your marathon results.  It’s still an approximation.

Miles and time, how many and how much, is the basic data that we have always had to work with.  This is a very non-specific and rough implement to build training plans around.  Still, it probably gets you 80% of the performance, but it leaves a lot on the table.

The next big leap in applying data to training is using heart rate.  Heart rate brings more granularity into our training.  What it really gives us is a way to track our effort level.  Why is that important?  Because in training your body has this threshold where the biochemistry changes as your effort intensifies.  You know it as the aerobic threshold or the lactic threshold.  This is the point where the muscles can’t keep up and start lactic acid buffering.

What we discovered is that training below this threshold moves the threshold.  HR is a better, still an approximation, but a better way of determining where you are in relation to that threshold.  We can move that threshold and run longer and faster.  HR doesn’t determine that.  HR is the best way we have to approximate it.

What we really need to know is what is the athletic capacity of our body?  How far can we go?  How deep is that well.  We can actually measure that.  It’s called a Max V02 test.  This directly measures your aerobic capacity by measuring the volume of your lungs and correlating it to effort.

It’s been hard for athletes to access this data because up to this point you needed a lab to be able to measure Max V02.

We are on the cusp of being able to take this Ventilatory Threshold data gathering out of the lab and into training.  This will give us a direct measurement of what we have been approximating with heart rate.  Being able to measure the Ventilatory Threshold real time will allow a fine tuning of the athletes’ training.

We will be able to move the conversation beyond are you training too much or too hard to specifically where you are training in relation to that specific athlete’s capacity to perform on that day, in that moment.  We will have the specific real-time data on that athlete’s response to the training effect and recovery.

It’s not just the data.  It’s what we can do with this data. We will soon be able to set loose learning algorithms on that data.  What used to be my coach telling me to take a day off will be an AI telling me mid-run to lower my effort or call it a day.

We are shaving away at the perfect training plan.  We are getting closer.  We’ll never get there because it’s Xeno’s paradox, reductio ad absurdum, Achilles and the tortoise, there will always be a better set of data.  There is no endpoint, but these are exciting times for endurance athletes.

We are on the edge of a data revolution in training.