Understanding Training Data (part 1)

Once upon a time we were excited to see our heart rate appear on our FitBit and using nothing short of magic it also “counted” our steps. But the world of fitness training has moved on a pace, and with modern devices, and ever more sophisticated software, it can sometimes feel like you are drowning in a tsunami of data and buzz words!

So what to make of it all? Without a doubt, focusing on the right metrics that support your training purpose, will help you improve faster and avoid unplanned plateaux’s and sessions that just don’t work for you. In this first post I look at some of the key metrics that I use and why I think they are important.

Which Software?

Truth is “which ever one you find easy to use”! Even the social platform Strava offers some degree of usefulness in this regard. But pretty much any fitness analytics software will give you a version of the data you need.

I use TrainingPeaks both as a Coach and Athlete. Partly because that is what I have become accustomed to, but mostly because it gives the most complete set of data and some very helpful analytics tools. If you are interested in trying it, then I offer a 14 day Free Premium Trial which includes Coaching and Support .

The Fundamentals

Whilst performing a cardio based activity (for our purposes running or cycling) we produce a limited set of metrics; our heart rate, power output if we have power meters on a bike and cadence whilst running. These data points are set against time and also (through the magic of GPS) terrain.

More sophisticated information is then derived from this for instance the time spent standing rather than sitting whilst peddling.

But from these few metrics we can start to build a picture of what happened in a particular session as well as the general training trend.

Zoning Out

I have written previously about setting training zones, so won’t go into that again here. But suffice to say that, regardless of the model you choose, knowing what represents “easy”, what is “hard but just sustainable” and what is a “vomit inducing near death experience” is important. Of particular importance is knowing the “zone” in which you are neither working “easily” or “hard”… Therefore avoiding the no mans land that athletes gravitate to when they have no plan at all!

Whilst zones are super helpful for planning and designing sessions, they also play a role in the “post match analysis”. If a session subjectively felt hard and yet was performed in an easy zone you have a red flag to worry about (over training? impending illness?). Of less concern a session that was upper threshold according to the data and yet reported as feeling easy (zones out of date? macho ego leading to unreliable feedback? broken hardware / inaccurate data?).

Looking beyond pure zones we start to think about some key ratio’s and metrics. For the rest of this blog I will be using the terminology and metrics offered by TrainingPeaks.

Training Stress Score (TSS)

Every time we perform a training session or an event we incur a level of physical stress. TSS attempts to quantify this. Within the world of TrainingPeaks the method used depends on the type of activity, for our purposes we can use this for running and cycling but is not meaningful for strength (or functional fitness activities).

For cycling, a power meter is needed. TSS being calculated using activity duration and intensity of the ride. One hour spent at Functional Threshold Power (therefore an up to date FTP is needed) would equate to 100 TSS points.

Running is slightly different (rTSS); Power being replaced by Pace. As with cycling TSS, we use Normalised Pace (or Power) to account for variation caused by gradient.

The above example (run), we can see that the planned rTSS was estimated at 70 (time / pace) but the actual result was 115. Why? Whilst the duration of the sessions was good, the average pace was too fast. Further investigation showed that the issue was in the 1km pickups that were too fast and at too high a heart rate. Therefore changing the nature of the session. This insight then gives options as to what to work on from a technique perspective and also how to adapt the rest of the training week.

Chronic Training Load (CTL)

An indicator of how fit you are. It combines duration and stress to provide an indicator of your historical training. The calculation uses an exponentially weighted average of Training Stress Score for the past 42 days.

What is a “good” value? That all depends on the person, it is relative… Focus on the trend and past performance (when that becomes available to you!). Is your fitness trending up, down or plateaux? And is this reflective of the training block you are in? Also if last year you were at a CTL of 120 and felt good and competed well, but now you are on a CTL of 40 and three weeks out from your big event, could be a point of concern… Used in that context (rather than absolute values) I think CTL is helpful.

To get a sense of how “hard” a particular session was, compare the actually TSS with your Fitness (CTL) and calculate the % change. In this example there is an approximate increase of 67%.

  • Hard Session : 50% to 100%
  • Moderate Session : 25% to 50%
  • Easy Session : 25% or lower

You could reasonably conclude that this was a “hard” session, but not crazy! Just a bit “keen” 😉

Intensity Factor (IF)

As noted previously, TSS is calculated based on Intensity and Duration. To calculate IF, we use the (normalised) power or pace achiveed as a ratio of the athletes Functional Threshold (power or pace). And can be calculated for an entire session, or more usefully, for isolated segments (for instance for the effort phase of individual reps).

To give some meaning to IF:

  • Relaxed “Recovery” type session : < .75
  • Endurance pace (Z2 Easy type run) : 0.75 to 0.85
  • Tempo paced work / Intervals : 0.85 to 0.95
  • Lactate Threshold Intervals : 0.95 to 1.05
  • Anaerobic Sprints : 1.05 to 1.15
  • Maximal Efforts : > 1.15

Going back to our example run, you can see the IF was planned as being .75 (Endurance Paced Run on average) but transpired to be more like high end Tempo and into Lactate Threshold Intervals. Again, consistent with the 1km intervals being run too hard. The consequence being that rather than building Endurance Economy, the session targeted Lactate threshold and Lactate Recovery. Useful, but not what was planned!

In Summary

Coaching, and training, relies upon a blend of qualitative and quantitative data to order form a wholistic picture of what the heck is going on! We can use TSS, CTL and IF as measures for a planned session and then obviously as part of the feedback post session. But equally (or more) importantly, we can use them to help answer some key questions:

  • Am I on track to be peaking for my desired event?
  • Am I doing enough, or just as importantly, am I doing too much?
  • Are my training sessions valuable in terms of achieving my goals or am I wasting my time? (Remember that a session can be “valuable” in many ways!

In part two I will explore some additional metrics that we can use to further understand what happened in a training session and get a better insight into how well our training is going.

Till then enjoy your running, cycling and training!


Ed Stivala is an England Athletic qualified Event Group Coach specialising in Endurance Events. His practice specialises in performance coaching for runners of all levels. He also holds qualifications in Personal Training, Nutrition and Swim Instruction. 
When not Coaching, he competes in Duathlon and plays golf very badly!