Identifying your 'real world' HRV limits

Alan Couzens, M.S. (Sports Science)

Nov 25th, 2015

In last week’s post, I looked at the relationship between training load and the athlete’s ability to complete a given session. I suggested that before mindlessly planning ballistic load ramps, it is important to do a ‘reality check’ of the implications of such a ramp on the athlete’s ability to complete the training and, for that matter, the athlete's overall feeling/enjoyment of training!

In this post, I want to follow up on that by bringing heart rate variability into the conversation as a further guide to efficient and realistic session prescription.

HRV research has been focused on statistically significant deviations from ‘normal’. While academically interesting, there is a gap between the statistical significance investigated in studies and the practical significance that a coach seeks in determining how to use HRV data to guide the training on a day to day basis. While a drop in HRV of 1 standard deviation may raise the proverbial ‘red flag’ in a general algorithm, what we really want to know is, what does a 1SD drop in HRV mean for this athlete? Does it affect the athlete’s ability to train or affect the response that this athlete gets from the training?

I’ll look at the first of these questions - the impact of changes in HRV on the athlete’s ability to train, in this post.

In my last post, I introduced a simple 10 point session rating scale, copied below…

When we compare ratings on this scale against an athletes RMSSD, we can begin to answer the question, at what RMSSD number (for this athlete) are sessions likely to be compromised, or more practically, at what RMSSD should we consider intervening to change the training prescription?

The relationship for 1 athlete (the same athlete’s data used in last week’s post) below…

While the overall correlation between RMSSD and session ‘Feel’ is fairly low (lower than the relationship between session feel & TSB from last week), if we look a little closer at the data, some interesting patterns appear, the first of which being that the correlation is significantly different from one half of the scatter to the other, i.e. important 'thresholds' are evident.

• The session feel begins to significantly ‘scatter out’ at an RMSSD less than 40ms. When this athlete’s RMSSD is above 40, there is session stability & a very good chance of an athlete rating the session at or above 3, i.e. getting the session done. When RMSSD is below 40, it’s a ‘crapshoot’ & the chance of an athlete feeling ‘dead tired’ or ‘not happening‘ goes way up.


• The highest consistent session ratings for this athlete come at an RMSSD in the 40-55 range. When RMSSD is >55, the chances of the athlete feeling ‘Easy was OK but the main set wasn’t happening’ actually goes up. Therefore HRV that is too high provides its own ‘yellow flag’ that maybe today isn’t the best day for a hard mainset.

I’ve tended to see the same patterns in all HRV/Feel data that I’ve looked at this way, however the particular thresholds or borders at which the session ratings start to scatter or at which that ‘sweetspot’ of greatest likelihood for feeling good during a mainset vary, making this kind of individual analysis of HRV limits important when determining the practical significance of HRV shifts.

Before we get too carried away with this relationship, an important reminder that we are dealing with probability and likelihood here. As the scatter shows, there are no hard and fast rules & our predictive ability, especially at certain points on the HRV scale is quite poor. No HRV algorithm is advanced enough to make things cut and dry...just yet :-) It’s all a game of risk tolerance & an athlete’s (& a coach’s) tolerance for risk may vary through the year. For example, in the early season, the intensity of a session may be secondary to getting the load done. However, as we approach competition, nailing those key sessions may become a higher priority. Then, as we get very close to competition, the impact of a ‘failed session’ on an athletes psyche may significantly affect our willingness to prescribe a session that might be ‘iffy’ in terms of the likeliness that an athlete will get it done & feel good in the process! HRV is a useful metric to 'throw into the mix' to help with these important decisions.

Finally, keep the predictive ability of HRV as a singular metric in its place. There is a relationship between HRV and the athlete’s ability to train however it’s far from a perfect correlation. Put more bluntly, a low HRV alone is not necessarily an excuse for a day off! HRV is best used as one of the biofeedback tools that helps an athlete get to know his/her own body and assess potential risk. It is akin to a compass but you are still the captain of your ship! Despite having the technology for it, driverless cars are taking a while to hit the roads for a reason! :-)

Used alone, HRV's ability to predict how an athlete will do on any one session is limited. However, when used as ‘an ingredient in the recipe’, i.e. when we add additional context from other measures such as the training load/TSB at the time of low readings, we can significantly increase our ability to make training prescriptions that keep our athletes happy and healthy.

Train smart,

AC

  Tweet