Q - Power

Indoor Rowing Team

Intensity (3 of 6)

(Above) Diagram 2. Heart Rate Reserve (HRR) can be expressed as a percentage, being the percentage of the gap between the resting heart rate of 34  (RHR) and maximum heart rate achievable whilst rowing of 190 (MHR).

(Below) Diagram 3. Example athlete's VO2max test results. The left y-axis shows O2/ml and the right y-axis shows VE/L, HR/bpm and power/W (i.e. the right y-axis represents 3 different units concurrently).

(Below) Diagram 4. An extract from Dalleck and Kravitz's findings as to the tolerable relationship between %HRR and %VO2max and the good relationship between %HRR and %VO2R.

(Below) Diagram 5. Example athlete's %VO2R/%HRR.
We can now start to add in our second currency – percentage of VO2max.

First, what is VO2max? It is the amount of oxygen that your body is capable of metabolising. This refers to the oxygen that you breath in, which then passes into the blood stream, which is then delivered to the cells that need it, and which then chemically reacts with a fuel (normally fat or glucose) to produce energy. “V” is for volume, “O2” is for oxygen, and “max” is for maximum.

It is important not to confuse your VO2max with the idea of the air you are breathing in and out (only 21% of which is oxygen on the way in, and of which about 16% of it is oxygen on the way out, at least when you are at rest), or to confuse VO2max with the size of your lungs. It is perhaps an unfortunate coincidence that the maximum amount of oxygen a person might be able to use in a minute is the same sort of amount and unit as vital capacity (say 3 to 7 litres depending on age, gender, genetics etc.).

For example, the example athlete's vital capacity is 5,740ml and his VO2max is 5,800ml. For reference, it is reported (unverified) that Sir Matthew Pinsent had a VO2max recorded at about 7,500ml in his day and a vital capacity of about 8,500ml. (It is often reported, incorrectly perhaps, that his VO2max was 8,500ml, rather illustrating the confusion between two very different concepts).

These VO2max figures are absolute figures. Absolute? Well, VO2max is typically expressed in two ways: relative and absolute. Absolute is simply the total amount of oxygen metabolised in a minute; relative is the amount per minute per kilogram of body weight. So James, who weighs 89 kg (or did) and has an absolute VO2max of 5,800ml (or had), and a relative VO2max of 65.2ml/kg. So we can write that figure into Diagram 2. Rowers tend to be far more interested in absolute VO2max whereas runners are far more interested in the relative figure – click here to see why.

I said above that example athlete's VO2max was 5,800ml for present purposes. The reason for the caveat is this. As can be seem from the results of his incremental step test which appear in Diagram 3, at maximal effort his oxygen consumption reached a plateau of about 5,800ml. In rowing, unlike most if not all other endurance sports, the muscles used in the mechanics of breathing are actually required to participate in the exercise action itself. In rowing, the torso needs to remain rigid to allow for the transmission of power from the legs to the handle. At the very end of the test after he put the handle down, he was able to continue breathing as hard as possible and, it seems, without the interference of the rowing action was actually able to suck in more air (VE increased from 140l/min to 147l/min) and more oxygen (VO2 reaching 5,860ml). (This would suggest that he is one of those athletes whose blood oxygen saturation falls below 95-100% at VO2 max, but this is a digression into another topic).

We can also write into Diagram 2 a figure for the amount of oxygen used at idle. Now admittedly this varies slightly from person to person, but getting it spot on does not make any significant difference from a training perspective. The standard figure which is used is 3.5ml/kg/min. Given that this represents your basal metabolic rate, a VO2/kg/min consumption of 3.5ml/kg/min is also referred to as 1MET. So his relative VO2max of 65.2ml/kg/min would be 18.6METs.

So is there a correlation between VO2max and HR? Sort of, but it is not very good. That is perhaps not all that surprising given the large error factor brought about by RHR. However there is a tolerable correlation between %VO2max and %HRR (which strips out the RHR component) and a good correlation between %VO2R and %HRR. (VO2R is VO2 reserve and in James's case represents the gap between 3.5ml/kg/min and 65.2ml/kg/min).

This really is tremendously convenient, because it means that we can now measure the oxygen throughput of our muscle cells when they are working just by looking at a HR monitor (well, now that we understand what it is telling us), and it is this rate of oxygen throughput which (generalising somewhat) determines the adaptation stimulation being applied to the muscle cells (noting that the duration of that stimulus is also a very important factor).

However having said there is this very good correlation between %VO2R and %HRR, I have not said, at least not yet, that the relationship is 1:1. One could have all sorts of other theoretical correlations which might be very predictable, but not linear (for example as %HRR increases, %VO2R might come up slowly at first, and then much quicker later on to catch back up). This is a bit of an aside, but it is worthwhile exploring this a little.

Is it the case that 35%VO2R correlates to 35%HRR, 50%VO2R correlates to 50%HRR and 80%VO2R correlates to 80%HRR? Well, Dalleck and Kravitz found that to be so in 2006 and the linear relationship is well described in their graphs shown to left in Diagram 4. The problem is that historically we are told that there should be an inverted deflection in the HR/work graph when this threshold is reached – the line climbs at a constant gradient until we get to VT2, at which point it flattens out a bit and climbs more slowly until MHR. Their conclusion was that further work was needed to address this inconsistency and they leave open the possibility that there is a better way of describing the relationship than by a purely linear regression. However in 2007 Lounana J. et al repeated Dalleck and Kravitz's results and found the %HRR / %VO2R regression was indistinguishable from the line of identity (i.e. x% HRR = x% VO2R) where x is between 35 and 95.

So where does this leave us? Well, we are faced with the possibility that below LT the %VO2R we are working at lags a bit behind %HRR and that when we pass VT2 it starts to catch up, only drawing level when we get to around 100%VO2R (which equates to MHR).

For no reason other than passing curiosity, below in Diagram 5 is set out the relationship between %VO2R and %HRR from the example athlete's testing. Interestingly, but frustratingly, this does not show anything like the linear correlation expected until 95%VO2max. In fact during this test, 86%HRR was needed to achieve an arbitrary 75%VO2R (at HR165; VO2=49.7ml/kg; 76.2%VO2max). In order to have the actual data roughly fit the line of identity, RHR would have to be set at about 100bpm, and plainly that is not right.